Title:
SYSTEMS AND METHODS FOR OPTIMAL ENERGY MANAGEMENT BASED ON TIME SERIES FORECASTING OF POWER LOAD
Document Type and Number:
WIPO Patent Application WO/2024/064258
Kind Code:
A1
Abstract:
An example method of optimized energy management includes creating a synthetic training dataset, where the synthetic training dataset includes a activity profiles for a period of time; training a deep learning model using the synthetic training dataset; predicting, using the trained deep learning model, a power load for the period of time; determining a projected state of charge (SOC) of an energy storage device during the period of time based, at least in part, on the predicted power load; and controlling charging operations for the energy storage device based on the projected SOC.
Inventors:
KHUNTIA SATVIK (US)
HANIF ATHAR (US)
AHMED QADEER (US)
MEIJER MAARTEN (US)
SWART CHARLES (US)
LAHTI JOHN (US)
JORGENSEN INER (US)
HARDAS SHWETA (US)
HANIF ATHAR (US)
AHMED QADEER (US)
MEIJER MAARTEN (US)
SWART CHARLES (US)
LAHTI JOHN (US)
JORGENSEN INER (US)
HARDAS SHWETA (US)
Application Number:
PCT/US2023/033344
Publication Date:
March 28, 2024
Filing Date:
September 21, 2023
Export Citation:
Assignee:
OHIO STATE INNOVATION FOUNDATION (US)
PACCAR INC (US)
PACCAR INC (US)
International Classes:
G06N3/08; B60L53/53; B60L53/66; B60L58/12; G05B23/02; G06Q10/04; G06N3/0442
Foreign References:
CN112819203A | 2021-05-18 | |||
US20100280698A1 | 2010-11-04 | |||
US20210221247A1 | 2021-07-22 | |||
US20160046292A1 | 2016-02-18 |
Attorney, Agent or Firm:
HAMILTON, Lee G. et al. (US)
Download PDF:
Claims:
MCC Ref. No.: 103361‐369WO1 CLAIMS 1. A method of optimized energy management, the method comprising: creating a synthetic training dataset, wherein the synthetic training dataset comprises a plurality of activity profiles for a period of time; training a deep learning model using the synthetic training dataset; predicting, using the trained deep learning model, a power load for the period of time; determining a projected state of charge (SOC) of an energy storage device during the period of time based, at least in part, on the predicted power load; and controlling charging operations for the energy storage device based on the projected SOC. 2. The method of claim 1, wherein the deep learning model comprises a recurrent neural network. 3. The method of claim 1, wherein the deep learning model comprises a long short term memory (LTSM) model. 4. The method of any one of claims 1‐3, wherein controlling charging operations for the energy storage device based on the projected SOC comprises controlling a vehicle engine. 5. The method of any one of claims 1‐4, wherein creating a synthetic dataset comprises generating the plurality of activity profiles from a base dataset. 6. The method of any one of claims 1‐5, wherein each of the plurality of activity profiles comprises sleep activity data and energy usage data. MCC Ref. No.: 103361‐369WO1 7. The method of any one of claims 1‐6 wherein each of the plurality of activity profiles comprises a time allocation matrix (TAM), wherein the TAM comprises temporal activity information. 8. The method of any one of claims 1‐7, wherein each of the plurality of activity profiles comprises a transition matrix (TM), wherein the TM comprises relational activity information. 9. The method of any one of claims 1‐8, wherein each of the plurality of activity profiles comprises a power load profile. 10. The method of any one of claims 1‐9, wherein the energy storage device is one or more batteries. 11. The method of any one of claims 1‐ 10 wherein the period of time is a hotel period for a long‐haul vehicle driver. 12. The method of any one of claims 1‐11, further comprising predicting an HVAC load, and wherein the predicted power load is based at least in part on the HVAC load. 13. The method of claim 12, wherein the step of determining a projected SOC comprises using dynamic programming to determine the projected SOC using the HVAC load and the predicted power load. 14. A system for optimized energy management, the system comprising: a vehicle comprising an energy storage device, a vehicle controller, and an engine; an energy management controller operably coupled to the vehicle, the energy management controller comprising a processor and a memory, the memory having MCC Ref. No.: 103361‐369WO1 computer‐executable instructions stored thereon that, when executed by the processor, cause the processor to: create a synthetic training dataset, wherein the synthetic training dataset comprises a plurality of activity profiles for a period of time; train a deep learning model using the synthetic training dataset; predict, using the trained deep learning model, a power load for the period of time; determine a projected state of charge (SOC) of an energy storage device during the period of time based, at least in part, on the predicted power load; and transmit the projected SOC to the vehicle controller, wherein the vehicle controller is configured to control charging operations for the energy storage device based on the projected SOC. 15. The system of claim 14, wherein the deep learning model comprises a recurrent neural network. 16. The system of claim 14, wherein the deep learning model comprises a long short term memory (LTSM) model. 17. The system of any one of claims 14‐16, wherein the vehicle controller is configured to control charging operations for the energy storage device based on the projected SOC by controlling a vehicle engine. 18. The system of any one of claims 14‐17, wherein creating a synthetic dataset comprises generating the plurality of activity profiles from a base dataset. 19. The system of any one of claims 14‐18 , wherein each of the plurality of activity profiles comprises sleep activity data and energy usage data. MCC Ref. No.: 103361‐369WO1 20. The system of any one of claims 14‐19, wherein each of the plurality of activity profiles comprises a time allocation matrix (TAM), wherein the TAM comprises temporal activity information. 21. The system of any one of claims 14‐20, wherein each of the plurality of activity profiles comprises a transition matrix (TM), wherein the TM comprises relational activity information. 22. The system of any one of claims 14‐21, wherein each of the plurality of activity profiles comprises a power load profile. 23. The system of any one of claims 14‐22, wherein the energy storage device is one or more batteries. 24. The system of any one of claims 14‐23, wherein the period of time is a hotel period for a long‐haul vehicle driver. 25. The system of any one of claims 14‐24, wherein the energy management controller is operably coupled to the vehicle over a communication network. 26. The system of any one of claims 14‐25, further comprising predicting an HVAC load, and wherein the predicted power load is based at least in part on the HVAC load. 27. The system of claim 26, wherein the projected SOC is determined using dynamic programming based on the HVAC load and the predicted power load. |
Description:
MCC Ref. No.: 103361‐369WO1 SYSTEMS AND METHODS FOR OPTIMAL ENERGY MANAGEMENT BAS
ED ON TIME SERIES FORECASTING OF POWER LOAD CROSS‐REFERENCE TO RELATED APPLICATIONS [0001] This application claims the benefit of U.S. provision
al patent application No. 63/408,626, filed on September 21, 2022, and titled
“CABIN LOAD PREDICTION USING TIME SERIES FORECASTING FOR LONG HAUL TRUCKS FOR OPTIMAL
ENERGY MANAGEMENT,” the disclosure of which is expressly incorporated herein
by reference in its entirety. STATEMENT REGARDING FEDERALLY FUNDED RESEARCH [0002] This invention was made with government support under
DE‐EE0008265 awarded by the Department of Energy. The government
has certain rights in the invention. BACKGROUND [0003] Long‐haul trucks are a major form of transportation
, and consume significant amounts of energy. A long‐haul truck can be used
in trips that last multiple days, and a driver may live aboard the truck during the trip. When a
driver lives aboard a truck, it can be referred to as “hoteling,” and the electric usage during
hoteling can be referred to as “cabin loads.”
Currently, many long haul trucks run on diesel fuel,
which can be a significant emitter of CO2 and other types of pollution. Hoteling can add to t
hese emissions because drivers can use appliances, HVAC (heating, ventilation and air conditi
oning) systems, entertainment devices, and other electrical devices. Drivers may need to ru
n the diesel truck engine to supply these loads, which generates additional emissions. MCC Ref. No.: 103361‐369WO1 [0004] Hybrid and electric power trains can be used to rep
lace and/or supplement existing diesel power trains on long haul trucks. Th
ese hybrid and electric power trains include energy storage systems (e.g., batteries) that can be
used to propel the truck using the stored energy. These energy storage systems can also be ade
quate to supply power for cabin loads during hoteling periods. To rely on the energy stora
ge systems during hoteling periods, the energy storage systems must be charged sufficiently t
o supply the cabin loads. SUMMARY [0005] Systems and methods for performing optimized energy m
anagement are described herein. In some implementations described he
rein, the systems and methods can be used to optimize energy management in a long‐haul
truck so that an energy source (e.g., a battery) has enough energy to supply the cabin loads
of the long‐haul truck during a hoteling period. [0006] In some aspects, the techniques described herein rela
te to a method of optimized energy management, the method including: cre
ating a synthetic training dataset, wherein the synthetic training dataset includes a plu
rality of activity profiles for a period of time; training a deep learning model using the synth
etic training dataset; predicting, using the trained deep learning model, a power load for the p
eriod of time; determining a projected state of charge (SOC) of an energy storage device during
the period of time based, at least in part, on the predicted power load; and controlling charging op
erations for the energy storage device based on the projected SOC. MCC Ref. No.: 103361‐369WO1 [0007] In some aspects, the techniques described herein rela
te to a method, wherein the deep learning model includes a recurrent neural
network. [0008] In some aspects, the techniques described herein rela
te to a method, wherein the deep learning model includes a long short term
memory (LTSM) model. [0009] In some aspects, the techniques described herein rela
te to a method, wherein controlling charging operations for the energy storage
device based on the projected SOC includes controlling a vehicle engine. [0010] In some aspects, the techniques described herein rela
te to a method, wherein creating a synthetic dataset includes generating the
plurality of activity profiles from a base dataset. [0011] In some aspects, the techniques described herein rela
te to a method, wherein each of the plurality of activity profiles includes
sleep activity data and energy usage data. [0012] In some aspects, the techniques described herein rela
te to a method wherein each of the plurality of activity profiles includes
a time allocation matrix (TAM), wherein the TAM includes temporal activity information. [0013] In some aspects, the techniques described herein rela
te to a method, wherein each of the plurality of activity profiles includes
a transition matrix (TM), wherein the TM includes relational activity information. [0014] In some aspects, the techniques described herein rela
te to a method, wherein each of the plurality of activity profiles includes
a power load profile. [0015] In some aspects, the techniques described herein rela
te to a method, wherein the energy storage device is one or more batteries.
MCC Ref. No.: 103361‐369WO1 [0016] In some aspects, the techniques described herein rela
te to a method‐ 10 wherein the period of time is a hotel period for a
long‐haul vehicle driver. [0017] In some aspects, the techniques described herein rela
te to a method, further including predicting an HVAC load, and wherein the p
redicted power load is based at least in part on the HVAC load. [0018] In some aspects, the techniques described herein rela
te to a method, wherein the step of determining a projected SOC includes usi
ng dynamic programming to determine the projected SOC using the HVAC load and the predicted
power load. [0019] In some aspects, the techniques described herein rela
te to a system for optimized energy management, the system including: a
vehicle including an energy storage device, a vehicle controller, and an engine; an ener
gy management controller operably coupled to the vehicle, the energy management controller incl
uding a processor and a memory, the memory having computer‐executable instructions stored
thereon that, when executed by the processor, cause the processor to: create a synthetic
training dataset, wherein the synthetic training dataset includes a plurality of activity pro
files for a period of time; train a deep learning
model using the synthetic training dataset; predict,
using the trained deep learning model, a power load for the period of time; determine a proj
ected state of charge (SOC) of an energy storage device during the period of time based, at
least in part, on the predicted power load; and transmit the projected SOC to the vehicle contro
ller, wherein the vehicle controller is configured to control charging operations for the ene
rgy storage device based on the projected SOC. MCC Ref. No.: 103361‐369WO1 [0020] In some aspects, the techniques described herein rela
te to a system, wherein the deep learning model includes a recurrent neural
network. [0021] In some aspects, the techniques described herein rela
te to a system, wherein the deep learning model includes a long short term
memory (LTSM) model. [0022] In some aspects, the techniques described herein rela
te to a system, wherein the vehicle controller is configured to control charg
ing operations for the energy storage device based on the projected SOC by controlling a vehicle
engine. [0023] In some aspects, the techniques described herein rela
te to a system, wherein creating a synthetic dataset includes generating the
plurality of activity profiles from a base dataset. [0024] In some aspects, the techniques described herein rela
te to a system , wherein each of the plurality of activity profiles includes
sleep activity data and energy usage data. [0025] In some aspects, the techniques described herein rela
te to a system, wherein each of the plurality of activity profiles includes
a time allocation matrix (TAM), wherein the TAM includes temporal activity information. [0026] In some aspects, the techniques described herein rela
te to a system, wherein each of the plurality of activity profiles includes
a transition matrix (TM), wherein the TM includes relational activity information. [0027] In some aspects, the techniques described herein rela
te to a system, wherein each of the plurality of activity profiles includes
a power load profile. [0028] In some aspects, the techniques described herein rela
te to a system, wherein the energy storage device is one or more batteries.
MCC Ref. No.: 103361‐369WO1 [0029] In some aspects, the techniques described herein rela
te to a system, wherein the period of time is a hotel period for a long‐
haul vehicle driver. [0030] In some aspects, the techniques described herein rela
te to a system, wherein the energy management controller is operably coupled
to the vehicle over a communication network. [0031] In some aspects, the techniques described herein rela
te to a system, further including predicting an HVAC load, and wherein the p
redicted power load is based at least in part on the HVAC load. [0032] In some aspects, the techniques described herein rela
te to a system, wherein the projected SOC is determined using dynamic program
ming based on the HVAC load and the predicted power load. [0033] It should be understood that the above‐described su
bject matter may also be implemented as a computer‐controlled apparatus, a co
mputer process, a computing system, or an article of manufacture, such as a computer‐reada
ble storage medium. [0034] Other systems, methods, features and/or advantages wil
l be or may become apparent to one with skill in the art upon examinat
ion of the following drawings and detailed description. It is intended that all such additional
systems, methods, features and/or advantages be included within this description and be
protected by the accompanying claims. BRIEF DESCRIPTION OF THE DRAWINGS [0035] The components in the drawings are not necessarily t
o scale relative to each other. Like reference numerals designate corresponding
parts throughout the several views. MCC Ref. No.: 103361‐369WO1 [0036] FIG. 1A illustrates an example method of optimized e
nergy management, according to implementations of the present disclosure
. [0037] FIG. 1B illustrates an example method of optimized e
nergy management, according to implementations of the present disclosure
. [0038] FIG. 2 illustrates a system for optimized energy man
agement, according to implementations of the present disclosure. [0039] FIG. 3 illustrates an example computing device. [0040] FIG. 4 illustrates an example load profile for a dr
iver during a hotel period. [0041] FIG. 5 illustrates an example power load profile for
an example 10 hour hotel load period. [0042] FIG. 6 illustrates an example plot of sleep duration
distributions for drivers on long‐haul truck trips. [0043] FIG. 7 illustrates an example method of performing p
rediction. [0044] FIG. 8 illustrates an example time allocation matrix
for a hotel period. [0045] FIG. 9 illustrates an example Euclidian distance comp
arison for an algorithm trained over test periods and hidden units. [0046] FIG. 10A illustrates an example predicted power load
profile. [0047] FIG. 10B illustrates an example test day load profil
e. [0048] FIG. 11 illustrates an example schematic of load pre
diction and energy estimation for a truck. [0049] FIG. 12 illustrates an example schematic for a cabin
HVAC system in a truck. MCC Ref. No.: 103361‐369WO1 [0050] FIG. 13 illustrates an example plot of temperature f
or various components of an example truck as a function of time. [0051] FIG. 14A illustrates an example of cabin temperature
error in an experiment compared to a simulation. [0052] FIG. 14B illustrates an example of temperature differ
ence for the experiment shown in FIG. 14A. [0053] FIG. 14C illustrates an error histogram for the expe
riment shown in FIGS. 14A and 14B. [0054] FIG. 14D illustrate an example of window temperature
error in an experiment compared to a simulation. [0055] FIG. 14E illustrates an example of temperature differ
ence for the experiment shown in FIG. 14D. [0056] FIG. 14F illustrates an error histogram for the expe
riment shown in FIGS. 14D‐ 14E. [0057] FIG. 15A illustrates an example of cabin temperature
error in an experiment compared to a simulation. [0058] FIG. 15B illustrates an example of temperature differ
ence for the experiment shown in FIG. 15A. [0059] FIG. 15C illustrates an error histogram for the expe
riment shown in FIGS. 15A and 15B. [0060] FIG. 15D illustrate an example of window temperature
error in an experiment compared to a simulation. MCC Ref. No.: 103361‐369WO1 [0061] FIG. 15E illustrates an example of temperature differ
ence for the experiment shown in FIG. 15D. [0062] FIG. 15F illustrates an example of roof temperature
error for a simulation compared to an experiment. [0063] FIG. 15G illustrates an example of temperature differ
ence for the experiment shown in FIG. 15F. [0064] FIG. 15H illustrates an example error histogram for
the results shown in FIGS. 14F and 15G. [0065] FIG. 16 illustrates a comparison of an example 2‐n
ode and an example 3‐node model. [0066] FIG. 17 illustrates a comparison of an example 2‐n
ode and an example 3‐node model with different data used. [0067] FIG. 18 illustrates example temperature profiles with
simulated vs. experimental data. [0068] FIG. 19 illustrates an example vapor compression cycl
e. [0069] FIG. 20 illustrates an example vapor compression cycl
e plotted on a P‐h graph. [0070] FIG. 21 illustrates example plots of test data tempe
rature as a function of time. [0071] FIG. 22A illustrates example evaporator flow rates. [0072] FIG. 22B illustrates example condenser flow rates.
[0073] FIG. 22C illustrates example coolant flow rates. MCC Ref. No.: 103361‐369WO1 [0074] FIG. 23A illustrates a calibrated evaporator model co
mparison on initial no‐ flow data. [0075] FIG. 23B illustrates a calibrated evaporator model co
mparison on transient data. [0076] FIG. 23C illustrates a calibrated evaporator model co
mparison on flow rate oscillating data. [0077] FIG. 23D illustrates a calibrated evaporator model co
mparison on steady state data. [0078] FIG. 24A illustrates a calibrated evaporator model co
mparison on initial no‐ flow data. [0079] FIG. 24B illustrates a calibrated evaporator model co
mparison on transient data. [0080] FIG. 24C illustrates a calibrated evaporator model co
mparison on flow rate oscillating data. [0081] FIG. 24D illustrates a calibrated evaporator model co
mparison on steady state data. [0082] FIG. 25 illustrates an RMSE error analysis of exampl
e data using two different calibrations. [0083] FIG. 26A illustrates experimental data compared to an
example model for a first data segment. [0084] FIG. 26B illustrates an error histogram for the comp
arison illustrated in FIG. 26A. MCC Ref. No.: 103361‐369WO1 [0085] FIG. 26C illustrates experimental data compared to an
example model for a second data segment. [0086] FIG. 26D illustrates an error histogram for the comp
arison illustrated in FIG. 26C. [0087] FIG. 26E illustrates experimental data compared to an
example model for a third data segment. [0088] FIG. 26F illustrates an error histogram for the comp
arison illustrated in FIG. 26E. [0089] FIG. 27 illustrates a comparison of RMSE error for
models of example condenser pressures. [0090] FIG. 28A illustrates experimental data compared to an
example model for a first data segment. [0091] FIG. 28B illustrates an error histogram for the comp
arison illustrated in FIG. 26A. [0092] FIG. 28C illustrates experimental data compared to an
example model for a second data segment. [0093] FIG. 28D illustrates an error histogram for the comp
arison illustrated in FIG. 26C. [0094] FIG. 28E illustrates experimental data compared to an
example model for a third data segment. [0095] FIG. 28F illustrates an error histogram for the comp
arison illustrated in FIG. 26E. MCC Ref. No.: 103361‐369WO1 [0096] FIG. 29A illustrates an example model compared to an
experiment. [0097] FIG. 29B illustrates an example model compared to an
experiment, using a different model calibration from the model shown in
FIG. 29A. [0098] FIG. 30 illustrates example efficiency values at a c
ompressor. [0099] FIG. 31A illustrates an example map of mechanical an
d electrical efficiency for a compressor. [00100] FIG. 31B illustrates an example map of volumetric ef
ficiency for a compressor. [00101] FIG. 31C illustrates an example of isotropic efficien
cy for a compressor. [00102] FIG. 32 illustrates an example block diagram of a s
imulator for an cabin HVAC in a truck. [00103] FIG. 33A illustrates an example of heat exchanger pr
essure over time. [00104] FIG. 33B illustrates an example of cabin temperature
profiles over time. [00105] FIG. 33C illustrates an example of battery power dem
anded over time. [00106] FIG. 34 illustrates an example optimal state of char
ge trajectory for a one day driving and hoteling cycle. [00107] FIG. 35 illustrates an example system configured to
estimate a load cycle based on user activity prediction and estimate HVAC
load cycle information, according to implementations of the present disclosure. DETAILED DESCRIPTION [00108] Unless defined otherwise, all technical and scientific
terms used herein have the same meaning as commonly understood by one of o
rdinary skill in the art. Methods and MCC Ref. No.: 103361‐369WO1 materials similar or equivalent to those described he
rein can be used in the practice or testing of the present disclosure. As used in the specificat
ion, and in the appended claims, the singular forms “a,” “an,” “the” include plural refe
rents unless the context clearly dictates otherwise.
The term “comprising” and variations thereof as used
herein is used synonymously with the term “including” and variations thereof and are open,
non‐limiting terms. The terms “optional” or “optionally” used herein mean that the subsequentl
y described feature, event or circumstance may or may not occur, and that the description incl
udes instances where said feature, event or circumstance occurs and instances where it does not.
Ranges may be expressed herein as from "about" one particular value, and/or to "about" anoth
er particular value. When such a range is expressed, an aspect includes from the one particular
value and/or to the other particular value. Similarly, when values are expressed as approx
imations, by use of the antecedent "about," it will be understood that the particular v
alue forms another aspect. It will be further understood that the endpoints of each of the ranges
are significant both in relation to the other endpoint, and independently of the other endpoint. Wh
ile implementations will be described for predicting energy consumption in vehicles, it wil
l become evident to those skilled in the art that the implementations are not limited thereto, but
are applicable for predicting energy use in different scenarios and contexts. [00109] The term “artificial intelligence” is defined here
in to include any technique that enables one or more computing devices or compin
g systems (i.e., a machine) to mimic human intelligence. Artificial intelligence (AI) includ
es, but is not limited to, knowledge bases, machine learning, representation learning, and deep le
arning. The term “machine learning” is defined herein to be a subset of AI that enables a
machine to acquire knowledge by extracting MCC Ref. No.: 103361‐369WO1 patterns from raw data. Machine learning techniques i
nclude, but are not limited to, logistic regression, support vector machines (SVMs), decision t
rees, Naïve Bayes classifiers, and artificial neural networks. The term “representation
learning” is defined herein to be a subset of machine learning that enables a machine to automa
tically discover representations needed for feature detection, prediction, or classification f
rom raw data. Representation learning techniques include, but are not limited to, autoencod
ers. The term “deep learning” is defined herein to be a subset of machine learning that that
enables a machine to automatically discover representations needed for feature detection, predictio
n, classification, etc. using layers of processing. Deep learning techniques include, but are
not limited to, artificial neural network or multilayer perceptron (MLP). [00110] Machine learning models include supervised, semi‐supe
rvised, and unsupervised learning models. In a supervised learning
model, the model learns a function that maps an input (also known as feature or features) t
o an output (also known as target or targets) during training with a labeled data set (or dataset)
. In an unsupervised learning model, the model learns patterns (e.g., structure, distribution,
etc.) within an unlabeled data set. In a semi‐ supervised model, the model learns a function that m
aps an input (also known as feature or features) to an output (also known as target or tar
get) during training with both labeled and unlabeled data. [00111] Long haul journeys in trucks can take up to a few
days and the driver spends 10 hours resting inside the cabin after every 11 ho
urs driving. This 10 hour rest period is referred to herein as the hoteling period, although
it should be understood that 10 hours is only an example, and that the systems and methods describ
ed herein can be used with any length of MCC Ref. No.: 103361‐369WO1 hoteling period. Sleeper cabs are used for this ap
plication and are equipped with devices like microwaves, coffee makers, lights, etc. for the drive
r to have a comfortable rest inside the truck anywhere along the route. The driver uses some of t
hese devices based on the need, for example, the driver uses a microwave when they want
to cook/heat food, and a coffee maker when they want to drink coffee. These devices have
a rated power that they draw when being used. This power draw is highly subjective to driver
behavior (what devices does the driver use and how many times). With the advent of hybrid elec
tric and battery electric trucks, it has become immensely important to have enough battery ene
rgy stored in the battery pack before the hoteling period since battery packs are the only
source of power unlike internal combustion engines and the drivers cannot be left with no powe
r in the battery when they are hoteling. Since federal laws mandate the drivers to rest for
10 hours after every 11 hours of driving, sometimes the drivers may not be near a truck stop
(i.e., near a grid‐based power source) and would need to start their hoteling period. This gene
rates a need for the driver to have sufficient energy stored in the trucks’ battery to supply the
cabin loads during the hotel period. [00112] Implementations of the present disclosure include meth
ods of optimized energy management. The methods described herein can b
e used to train deep learning models to perform optimized energy management using synthetic
data. Using synthetic data can overcome the limitations of existing training methods,
which can require large amounts of real‐ world data. Real‐world systems, however, may not be
configured to generate real‐world data (for example, they may lack sensors, data storage, a
nd/or networking capabilities). Thus, the use of synthetic data for training allows for the t
raining of deep learning models for optimized energy management in situations where real‐world dat
a is not available. For example, there is MCC Ref. No.: 103361‐369WO1 no existing dataset for cabin loads during hoteling
periods that can be used to train a deep learning model. Real‐world data, even if it exists,
would not include a sufficiently large number of samples, and much less data would lack diversity.
The present disclosure describes techniques for generating synthetic data, which includ
es but is not limited to generating activity profiles. Such activity profiles include temporal and
relational energy usage data as well as sleep activity data during hoteling periods. The synt
hetic data created according to this disclosure addresses challenges unique to the cabin l
oad during hoteling application (e.g., time series forecasting of load) and results in a dataset
representative of real‐world data. Thus, the present disclosure contemplates that deep learning mod
els trained with such a synthetic training dataset will have better accuracy, have less
tendency to overfit, and better generalize to unseen data. [00113] With reference to FIG. 1A, a method 100 of optimize
d energy management is shown according to an implementation of the present
disclosure. [00114] At step 110, the method 100 can include creating a
synthetic training dataset. The synthetic training dataset can include a pluralit
y of activity profiles for a period of time. [00115] Optionally, the period of time is the hotel period
for a long‐haul driver of a vehicle. As used herein, the hotel period can be an
y length of time that cabin loads are being used in the vehicle without the vehicle being in tr
ansit. [00116] The synthetic dataset created at step 110 can option
ally include a plurality of activity profiles from a base dataset. Activity profi
le creation is described in detail, for example, in Example 1 below. Alternatively or additionally, ea
ch of the plurality of activity profiles can include a time allocation matrix (TAM), where the TA
M comprises temporal activity MCC Ref. No.: 103361‐369WO1 information. The temporal activity information can inc
lude probability density functions of activities over time. An example time allocation matr
ix is shown in FIG. 8. [00117] Optionally, the plurality of activity profiles can fu
rther include sleep activity data and energy usage data. Alternatively or addition
ally, each of the plurality of activity profiles can include a power load profile. [00118] Alternatively or additionally, each of the plurality
of activity profiles an optionally further include a transition matrix (TM),
wherein the TM comprises relational activity information. Relational activity information can be th
e probability of a next activity in the sequence being completed. Optionally, relational activi
ty information can be modeled using a Markov chain. [00119] At step 120, the method 100 can include training a
deep learning model using the synthetic training dataset. According to the pres
ent disclosure, the deep learning model is trained to “learn” a function that maps an input
(also known as feature or features) to an output (also known as target or targets) during trai
ning with the synthetic training dataset. For example, the features may include, but are not limit
ed to, energy usage information such as the TAM and/or TM and sleep activity data, and the targ
et may be a power load profile. The deep learning model is trained with the synthetic training
dataset to maximize or minimize an objective function. This disclosure contemplates traini
ng the deep learning model using techniques known in the art. [00120] Optionally, the deep learning model includes an artif
icial neural network (ANN). An artificial neural network (ANN) is a compu
ting system including a plurality of interconnected neurons (e.g., also referred to as “
nodes”). This disclosure contemplates that MCC Ref. No.: 103361‐369WO1 the nodes can be implemented using a computing devic
e (e.g., a processing unit and memory as described herein). The nodes can be arranged in
a plurality of layers such as input layer, output layer, and optionally one or more hidden laye
rs. An ANN having hidden layers can be referred to as deep neural network or multilayer per
ceptron (MLP). Each node is connected to one or more other nodes in the ANN. For example, e
ach layer is made of a plurality of nodes, where each node is connected to all nodes in the p
revious layer. The nodes in a given layer are not interconnected with one another, i.e., the nodes
in a given layer function independently of one another. As used herein, nodes in the input lay
er receive data from outside of the ANN, nodes in the hidden layer(s) modify the data between
the input and output layers, and nodes in the output layer provide the results. Each node is
configured to receive an input, implement an activation function (e.g., binary step, linear, sigmoi
d, tanH, or rectified linear unit (ReLU) function), and provide an output in accordance with
the activation function. Additionally, each node is associated with a respective weight. ANNs ar
e trained with a dataset to maximize or minimize an objective function. In some implementation
s, the objective function is a cost function, which is a measure of the ANN’s performa
nce (e.g., error such as L1 or L2 loss) during training, and the training algorithm tunes the node
weights and/or bias to minimize the cost function. This disclosure contemplates that any algori
thm that finds the maximum or minimum of the objective function can be used for training
the ANN. Training algorithms for ANNs include, but are not limited to, backpropagation. ANN
s are known in the art and are therefore not described in further detail herein. [00121] Optionally, the deep learning model can be a recurre
nt neural network (RNN). An RNN is a class of artificial neural network wher
e connections between nodes can create a MCC Ref. No.: 103361‐369WO1 cycle, allowing output from some nodes to affect sub
sequent input to the same nodes. The RNN has internal memory and can be used to analyze
sequential or time series data. A non‐ limiting example RNN architecture is a long short te
rm memory (LTSM). LSTM models are designed to handle sequential data, such as time ser
ies data. [00122] It should be understood that RNN and LSTM are provi
ded only as example deep learning models. This disclosure contemplates tha
t the deep learning model can be another type of deep learning model. [00123] At step 130, the method 100 can include predicting,
using the trained deep learning model, a power load for the period of time
. At step 130, the deep learning model is operating in inference mode. The deep learning model
has therefore been trained (i.e. at step 120) and is configured to make predictions based on
new input data. Accordingly, such a model is referred to herein as the “trained deep learnin
g model.” The input to the trained deep learning model can include measurements of energy usa
ge and/or power consumption taken during a first time period (e.g., a first hotel per
iod). Alternatively or additionally, the input to the trained deep learning model can include energy c
onsumption during the start of the hotel period. Alternatively or additionally, the input to t
he trained deep learning model can include power consumption during any part of the hotel perio
d. Additional non‐limiting examples of the outputs of the trained deep learning model inclu
de the power load for an entire day, an entire hotel period, or a period of time within a
day or hotel period (e.g., a certain number of minutes or hours). As yet another non‐limiting exam
ple, input can be the power load during one or more time periods (e.g., hotel periods). The
output can be the predicted power load for a future time period (e.g., the next hotel period).
MCC Ref. No.: 103361‐369WO1 [00124] At step 140, the method 100 can include determining
a projected state of charge (SOC) of an energy storage device during the
period of time based, at least in part, on the predicted power load. As a non‐limiting example
, the energy storage device can include one or more batteries or battery packs (e.g., packs of
lithium or lead‐acid batteries). Optionally, the batteries or battery packs can be part of an electr
ic and/or hybrid power train for a vehicle (e.g., a truck). Optionally, the projected SOC can
be based on both the predicted power load and a predicted HVAC load. An example HVAC model th
at can be used to predict the HVAC load is described in Example 2. In some implementations,
dynamic programming can be used to obtain the projected SOC, and the dynamic programming
inputs can include the predicted HVAC load and the predicted power load. [00125] At step 150, the method 100 can include controlling
charging operations for the energy storage device based on the projected SOC
. Optionally, controlling charging operations for the energy storage device based on th
e projected SOC can include controlling a vehicle engine. Alternatively or additionally, controll
ing charging operations for the energy storage device based on the projected SOC can includ
e controlling a vehicle engine. Non‐ limiting examples of controlling the engine can inclu
de starting the vehicle engine to charge the battery, and/or turning off the engine to stop charg
ing the battery. [00126] With reference to FIG. 2, implementations of the pre
sent disclosure include systems for optimized energy management. The system 2
00 shown in FIG. 2 can include an energy management controller 210 and a vehicle 250.
The vehicle 250 can include an engine 260 and an energy storage device 270. MCC Ref. No.: 103361‐369WO1 [00127] The energy management controller 210 can include a c
omputing device (e.g., the computing device 300 shown in FIG. 3), including
a processor and a memory. The energy management controller 210 can be operably coupled to
the vehicle 250. In some implementations, the energy management controller 210
can be connected to the vehicle 250 by a network (e.g., the network connections 316 illu
strated in FIG. 3). The energy management controller 210 can include a synthetic training datas
et 220 and a deep learning model 230. [00128] The energy management controller 210 can be configure
d to perform any one or more of the steps of the methods described with
reference to FIGS. 1A‐1B. [00129] Optionally, the energy management controller 210 can
be configured to create a synthetic training dataset, where the synthe
tic training dataset can include a plurality of activity profiles for a period of time. Optionall
y, the synthetic dataset can be created using a number of activity profiles from a base dataset. It
should be understood that the energy management controller 210 can be configured to run a
trained deep learning model, and that the deep learning model 230 can be a trained deep
learning model. [00130] In some implementations, each of the plurality of ac
tivity profiles comprises sleep activity data and energy usage data. Optionally
, each of the plurality of activity profiles comprises a time allocation matrix (TAM), wherein the
TAM comprises temporal activity information. Alternatively or additionally, each of th
e plurality of activity profiles can include a transition matrix (TM), where the TM can include rel
ational activity information. Alternatively or additionally, each of the plurality of activity p
rofiles comprises a power load profile. [00131] The energy management controller 210 can also include
a deep learning model 230. The deep learning model 230 can be train
ed using a synthetic training dataset as MCC Ref. No.: 103361‐369WO1 described with reference FIG. 1A‐1B. Optionally, the
deep learning model can include a recurrent neural network. Alternatively or additionally
, the deep learning model can include a long short term memory (LTSM) model. [00132] The energy management controller 210 can be configure
d to predict a power load for the vehicle 250 for a period of time. As
a non‐limiting example, the period of time can be a hotel period for a long‐haul vehicle driver,
but it should be understood that any length of time, including any length of hotel period, can be
used. [00133] Using the prediction, the energy management controller
can determine a projected state of charge (SOC) of an energy storage
device 270 during the period of time based, at least in part, on the predicted power loa
d. The energy management controller 210 can control charging operations for the energy storag
e device 270 based on the predicted power load and the projected state of charge. Option
ally, the engine 260 can be used to charge the energy storage device 270, and the engine 260 c
an be controlled by the energy management controller 210 based on the projected SOC
(for example, to charge the energy storage device 270 to the projected state of charge)
. As a non‐limiting example, the energy management controller 210 can turn the engine 260 on
to charge the energy storage device 270 and turn the engine 260 off to stop charging t
he energy storage device 270. By turning the engine 260 on and off, the energy management control
ler 210 can control the SOC. [00134] In some implementations, the energy management control
ler 210 can be located on the vehicle 250. As a non‐limiting ex
ample, the energy management controller 210 can be part of any computing device located on the
vehicle 250. In some implementations, the energy management controller 210 can be located separ
ately from the vehicle 250 (e.g., on MCC Ref. No.: 103361‐369WO1 another computing device or server). When the energy
management controller 210 is separate from the vehicle 250, the energy management controlle
r 210 can operably coupled to the vehicle 250 by a communication link (e.g., a cellula
r network) such that the energy management controller 210 can communicate with the ve
hicle 250. Optionally, the vehicle 250 can include a vehicle controller 275 configured to r
eceive instructions from the energy management controller 210 and/or transmit information
about the power consumption of the vehicle 250 to the energy management controller 210.
Optionally, the synthetic dataset step 110 of the method 100 shown in FIG. 1A can be per
formed on the energy management controller 210, which can reduce the amount of memor
y and processing power required by the vehicle. [00135] In some implementations, the vehicle controller 275 c
an include a lightweight machine learning model. As used herein, the term “
lightweight” model refers to models that can require fewer computational resources and/or less
memory to run in inference mode. The lightweight machine learning model can be based on t
he deep learning model 230. The lightweight machine learning model can be a version
of the deep learning model 230 that is optimized to efficiently operate in inference mode. T
he lightweight machine learning model of the vehicle can optionally be incrementally trained b
ased on new data. [00136] Alternatively or additionally, it should be understood
that in some implementations, the deep learning model 230 trained
by the energy management controller 210 can be used to generate lightweight machine lear
ning models for any number of vehicles 250. Optionally, the vehicles 250 can update the res
pective lightweight machine learning models of each vehicle 250 based on the actual ener
gy usage of each vehicle 250. This MCC Ref. No.: 103361‐369WO1 approach can allow for efficient training of the dee
p learning model 230 using synthetic data (e.g., according to the methods 100, 150 of FIG. 1A
and FIG. 1B, described herein); as well as efficient deployment and customization of the deep le
arning model 230 for any number of vehicles using the lightweight machine learning models
. [00137] It should also be understood that performance optimiz
ations can be used for the HVAC models described herein. The equations of e
xample 2 can optionally be discretized and/or mapped for faster computation. [00138] FIG. 1B illustrates a method 160 for modeling energy
consumption, according to implementations of the present disclosure. [00139] At step 162, the method 160 includes collecting info
rmation (e.g., survey information). For example, step 162 can include colle
cting information and drawing observations from surveys in literature about driver
sleeping & driving behavior. Alternatively or additionally, the information can include compilati
on of driver schedules during hoteling and/or surveys from online available forums/videos/blog
s from truck drivers. [00140] At step 164, the method 160 can include creating tr
aining data. The training data can be created using the derived observations t
o create rules which can be used in duplicating the data into 1000’s of data points fo
r machine learning model training. [00141] At step 166, the method 160 can include performing
exploratory data analysis. Exploratory data analysis can include extrac
ting additional information from the data using data analysis to produce additional features. A
dditional features can support machine learning model training. MCC Ref. No.: 103361‐369WO1 [00142] At step 168, the method can include data augmentatio
n. The data augmentation can be performed using the additional fe
atures created in step 166. [00143] At step 170, the method can include using an LSTM
algorithm based on the data augmented at step 168. [00144] At step 172, the method can include result analysis.
The result analysis can include analyzing and processing the output of the L
STM algorithm to establish a uniform comparison metric. [00145] At step 174, the method can include hyperparameter o
ptimization. Hyperparameter optimization can include tuning algorith
m parameters to find an optimal compromise between the accuracy of the algorithms and
the computational power required for the algoirhtms. [00146] At step 176, the method can include prediction perfo
rmance evaluation. Prediction performance evaluation can include checking
algorithm accuracy for the predicted overall energy consumption. [00147] It should be appreciated that the logical operations
described herein with respect to the various figures may be implemented (1
) as a sequence of computer implemented acts or program modules (i.e., software) running on
a computing device (e.g., the computing device described in FIG. 3), (2) as interconnected m
achine logic circuits or circuit modules (i.e., hardware) within the computing device and/or (3) a c
ombination of software and hardware of the computing device. Thus, the logical operations di
scussed herein are not limited to any specific combination of hardware and software. The im
plementation is a matter of choice dependent on the performance and other requirements o
f the computing device. Accordingly, MCC Ref. No.: 103361‐369WO1 the logical operations described herein are referred
to variously as operations, structural devices, acts, or modules. These operations, structura
l devices, acts and modules may be implemented in software, in firmware, in special purp
ose digital logic, and any combination thereof. It should also be appreciated that more or
fewer operations may be performed than shown in the figures and described herein. These ope
rations may also be performed in a different order than those described herein. [00148] Referring to FIG. 3, an example computing device 300
upon which the methods described herein may be implemented is illust
rated. It should be understood that the example computing device 300 is only one example of
a suitable computing environment upon which the methods described herein may be implemented
. Optionally, the computing device 300 can be a well‐known computing system including,
but not limited to, personal computers, servers, handheld or laptop devices, multiprocessor sy
stems, microprocessor‐based systems, network personal computers (PCs), minicomputers, mainfr
ame computers, embedded systems, and/or distributed computing environments including a
plurality of any of the above systems or devices. Distributed computing environments enable remo
te computing devices, which are connected to a communication network or other data t
ransmission medium, to perform various tasks. In the distributed computing environment, the
program modules, applications, and other data may be stored on local and/or remote computer
storage media. [00149] In its most basic configuration, computing device 300
typically includes at least one processing unit 306 and system memory 304.
Depending on the exact configuration and type of computing device, system memory 304 may
be volatile (such as random access memory (RAM)), non‐volatile (such as read‐only mem
ory (ROM), flash memory, etc.), or some MCC Ref. No.: 103361‐369WO1 combination of the two. This most basic configuration
is illustrated in FIG. 3 by dashed line 302. The processing unit 306 may be a standard programmab
le processor that performs arithmetic and logic operations necessary for operation of the
computing device 300. The computing device 300 may also include a bus or other communic
ation mechanism for communicating information among various components of the computing
device 300. [00150] Computing device 300 may have additional features/func
tionality. For example, computing device 300 may include additional
storage such as removable storage 308 and non‐removable storage 310 including, but not li
mited to, magnetic or optical disks or tapes. Computing device 300 may also contain network connect
ion(s) 316 that allow the device to communicate with other devices. Computing device 300
may also have input device(s) 314 such as a keyboard, mouse, touch screen, etc. Output devi
ce(s) 312 such as a display, speakers, printer, etc. may also be included. The additional d
evices may be connected to the bus in order to facilitate communication of data among the compone
nts of the computing device 300. All these devices are well known in the art and need n
ot be discussed at length here. [00151] The processing unit 306 may be configured to execute
program code encoded in tangible, computer‐readable media. Tangible, compu
ter‐readable media refers to any media that is capable of providing data that causes the c
omputing device 300 (i.e., a machine) to operate in a particular fashion. Various computer‐re
adable media may be utilized to provide instructions to the processing unit 306 for execution
. Example tangible, computer‐readable media may include, but is not limited to, volatile
media, non‐volatile media, removable media and non‐removable media implemented in any method o
r technology for storage of information such as computer readable instructions, da
ta structures, program modules or other MCC Ref. No.: 103361‐369WO1 data. System memory 304, removable storage 308, and
non‐removable storage 310 are all examples of tangible, computer storage media. Example
tangible, computer‐readable recording media include, but are not limited to, an integrated
circuit (e.g., field‐programmable gate array or application‐specific IC), a hard disk, an optica
l disk, a magneto‐optical disk, a floppy disk, a
magnetic tape, a holographic storage medium, a solid
state device, RAM, ROM, electrically erasable program read‐only memory (EEPROM), flash me
mory or other memory technology, CD‐ROM, digital versatile disks (DVD) or other opti
cal storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage
devices. [00152] In an example implementation, the processing unit 306
may execute program code stored in the system memory 304. For example,
the bus may carry data to the system memory 304, from which the processing unit 306 recei
ves and executes instructions. The data received by the system memory 304 may optionally be
stored on the removable storage 308 or the non‐removable storage 310 before or after execu
tion by the processing unit 306. [00153] It should be understood that the various techniques
described herein may be implemented in connection with hardware or software o
r, where appropriate, with a combination thereof. Thus, the methods and apparatuses
of the presently disclosed subject matter, or certain aspects or portions thereof, may
take the form of program code (i.e., instructions) embodied in tangible media, such as flo
ppy diskettes, CD‐ROMs, hard drives, or any other machine‐readable storage medium wherein, w
hen the program code is loaded into and executed by a machine, such as a computing devi
ce, the machine becomes an apparatus for practicing the presently disclosed subject matter.
In the case of program code execution on programmable computers, the computing device generally
includes a processor, a storage MCC Ref. No.: 103361‐369WO1 medium readable by the processor (including volatile
and non‐volatile memory and/or storage elements), at least one input device, and at least
one output device. One or more programs may implement or utilize the processes described in
connection with the presently disclosed subject matter, e.g., through the use of an applicat
ion programming interface (API), reusable controls, or the like. Such programs may be implemen
ted in a high level procedural or object‐ oriented programming language to communicate with a c
omputer system. However, the program(s) can be implemented in assembly or machine
language, if desired. In any case, the language may be a compiled or interpreted language a
nd it may be combined with hardware implementations. [00154] Examples [00155] The following examples are put forth so as to provi
de those of ordinary skill in the art with a complete disclosure and description o
f how the compounds, compositions, articles, devices and/or methods claimed herein are m
ade and evaluated, and are intended to be purely exemplary and are not intended to limit t
he disclosure. Efforts have been made to ensure accuracy with respect to numbers (e.g., amount
s, temperature, etc.), but some errors and deviations should be accounted for. Unless indica
ted otherwise, parts are parts by weight, temperature is in ^C or is at ambient temperature, and pressure is at
or near atmospheric. [00156] Example 1: [00157] An example implementation of the present disclosure i
ncludes methods of predicting cabin load in long haul trucks. Predicting
cabin load in long haul trucks can be important to operating trucks efficiently. For example
, long haul trucks can include batteries that are charged from an engine or generator while
the truck is operating, or from an electric MCC Ref. No.: 103361‐369WO1 charger (e.g., a charger connected to the electric g
rid). When the engine or generator is not running, the battery is responsible for providing ene
rgy to the truck, including the truck cabin. If the battery is over charged, it can result in waste
d energy (for example, wasted fuel used to charge the battery). On the other hand, if the batt
ery is under charged, the truck can run out of energy before the engine/generator is started again.
Thus, it is desirable to accurately predict how much energy will be used when the engine/generat
or is not running, so that the batteries can be sufficiently charged, but not over charged.
[00158] Implementations of the present disclosure can be used
for hybrid or electric trucks where energy management of the batteries can
be especially important. Alternatively, implementations of the present disclosure can be used
for conventional internal combustion trucks where idling the engine can be harmful. [00159] Hybridization of the vehicle allows for the substitut
ion of this idling (to power the auxiliaries) with any Energy Storage System (ESS)
like battery packs. Moreover, it is important to ensure the battery pack has sufficient
State‐Of‐Charge (SOC) at all times. This is not entirely possible because of the limitation in t
he battery sizing, hence there might be instances where some idling may be required to charg
e the battery back up. In this situation, it is helpful to know how much instantaneous power or
the total electrical energy would be required in the hotel period so that the battery pa
ck is only sufficiently charged and only charged at required times eliminating unnecessary idli
ng. [00160] The present disclosure includes methods to predict th
ese electrical loads in the sleeper cabs of the trucks using machine learnin
g methods. In the present example, a neural network referred to herein as Long and Short
Term Memory (LSTMs) can be used to MCC Ref. No.: 103361‐369WO1 predict the power demand as a time series. LSTMs ca
n be very capable of capturing temporal dependencies quite well and hence is the algorithm o
f choice. The algorithm is trained for 20 days and makes predictions for the 21st day. Data i
s in the form of on/off timing of each device in the 10‐ hour hoteling period. At each instant,
the rated power of all the devices that are switched on is added together to form 1 time series
capturing the total power. This can be fed into the LSTM network along with the temporal and r
elational information about the different devices. Because of limitations in finding data about
driver behavior during hoteling periods in the industry, some survey information made available
by the industry partner and some surveys reported in the literature about driver sleep
ing behavior were studied. This study is used to derive some observations about driver behavio
r and this information is replicated into 1000 synthetically generated datasets which were used
to train and make predictions. After generating predictions, they can be validated visually
and numerically. [00161] For visual validation, dynamic time warping can be u
sed to map the prediction to a test day and for numerical validation, the err
or in total energy consumption is used and the accuracy attained by the example implementation was 9
0%. Optionally, the dynamic time warping can “level the field” of running standard
error comparison matrices. For example, dynamic time warping can allow for more accurate ali
gnment and/or comparison of sequences of data. Dynamic time warping can be used in implem
entations of the present disclosure to handle temporal distortions or variations in speed or
timing of sequences of data. Alternatively or additionally, dynamic time warping can be used fo
r hyperparameter optimization. [00162] In some implementations, the types of power loads th
at are expected in the hotel period can be predicted. FIG. 4 illustrates ex
amples of typical cabin electrical loads on a MCC Ref. No.: 103361‐369WO1 long‐haul truck. These include the use of a lamp,
TV, radio etc. Predicting these activities can be used to predict instantaneous power expected in the
next time horizon (e.g., until the end of the hotel period). In a situation where the battery
reaches the minimum allowable SOC, this prediction can be used to idle the vehicle to charg
e the battery just enough. In the example implementation of SuperTruck II, the HVAC is also po
wered by the battery pack hence also comes within the cabin loads. However, the HVAC powe
r requirement estimation is not a part of the this predictive algorithm because of its dyna
mic nature. A physics‐based model can be developed instead for the e‐HVAC power load estimat
ion modelled separately on MATLAB/Simulink [ Khuntia et al. (2022)]. [00163] User activity prediction can include Sequence Predicti
on via Enhanced Episode Discovery (SPEED). SPEED can be superior to other pr
ediction algorithms like Active LeZi (ALZ) algorithm with temporal rule, or Patterns of User Be
havior System (PUBS). User activity prediction can include using the temporal pattern of
humans for the prediction [ Aztiria et al. (2012)]. For example, a modified‐SPEED algorithm can
achieve 96.8% accuracy. [Marufuzzaman et al. (2015)], the authors introduce This prediction
algorithm is also adopted for the prediction of activities in smart homes for gird power‐cost m
anipulation. As another example an application of a multi layer Long and Short Term Me
mory (LSTM) algorithm to predict the time and duration of activities in a 24 hour period can
include a two layer LSTM algorithm with 50 hidden units each can give good predictions with a
learning on a moving 75 day window. [Goutham (2020)]. [00164] Time series prediction can be done using Recurrent N
eural Networks (RNN) as the ability of having a "memory" which makes them g
ood for long sequence prediction tasks. MCC Ref. No.: 103361‐369WO1 The RNN can remember contextual information through t
he hidden layer activations that are passed from one step in time to another. Different
variants of RNN have been useful when temporal dependency of the data are important [ Grav
es (2013)]. A popular variant of RNN is the Long and Short Term Memory (LSTM). LSTMs work i
n an iterative fashion like RNNs with the addition of a gating mechanism. It can include three
gates named as: (i) Forget gate (ii) Update gate and (iii) Output gate which regulate the flow
of information from input to activation, activation to activation and activation to output. Th
is makes LSTMs robust against outliers in the data and learn long term dependencies and patter
ns in the time series. Deeper version of neural network, that is multiple (aka stacked) LSTM
layers can be used to improve the performance [ Goutham et al. (2021)]. [00165] Implementations of the present disclosure include syst
ems and methods that can be used to implement energy prediction in vehicl
es, for example long haul trucks and/ or hybrid trucks. The study described herein includes a
method to synthesize data from a single available data set generated from a survey conducted
by PACCAR Inc. Using the synthesized data, Neural Networks can be used to predict the fu
ture power load estimate (e.g., implemented using MATLAB). Moreover, implementations of
the present disclosure can include leveraging the properties of Markov chain and an ada
ptation of a transition matrix. The study shows that the systems and methods disclosed can be
used to output a prediction for the overall power demand. This power estimate can be use
d in the estimation of the projected SOC for the duration of the hotel period, which in turn
is used for the idle/ Engine On‐Off control strategies. MCC Ref. No.: 103361‐369WO1 [00166] The present disclosure includes different metrics for
evaluating the predictions of a time series. For the activity predi
ction as a classification problem, classifier based methods can be used. In this work, the proble
m is formulated as a regression problem. Alternatively or additionally, a point‐wise numeric
distance between the prediction and the original values can be used. [00167] Some of these metrics are: [00168] Root mean squared error (RMSE): provides an average
error throughout the predicted time series in real units. 0 169] ^^ ^^ ^^ ^^ ൌ ∑ ∗ మ [ 0 ^ ൫ ˆ^ି௬^ ൯ ் : a method that uses a normalized error over the total range of values of the test set. This is par
ticularly useful in the case when the prediction of
the time of the activity is not critical. [00171] ^^ ^^ ^^ ^^ ^^ ൌ ோெௌா ୫ ୟ^൫௬∗ ∗ ^ ൯ି୫୧୬൫௬^ ൯ (MAPE): another metric to evaluate the predictions where the error is normalized over the a
ctual value. This error is a percentage of the true value. ห ∗ ∑ ^ ^ˆ^ష^ ห ∗ ^ ^ ^ (ETF). [Minor et al. (2015)]. In this method, the error function, ^^^ ^ˆ^ ^ , ^^ ^ ∗ ^ ൌ 1 if | ^ˆ^ ^ െ ^^ ^ ∗ | ^ ^^, ^^ being a non negative threshold, i.e., error in a particular predi
ction contributes to the total error only if it is
significant enough. MCC Ref. No.: 103361‐369WO1 75] ^^ ^^ ^^ ൌ ∑ ൫ ∗ [001 ூ ˆ^,௬^ ൯ ் another signal. Using Dynamic Time Warping (DTW), the
example implementation can determine the similarity between the two signals. DTW
warps the x‐axis (in this case, time axis) between the two signals to match the best y axis v
alues irrespective of the lengths of the two signals [ Müller (2007)]. For two inputs ^^ ^ ∈ ℝ ே and ^^ ଶ ∈ ℝ ே DTW computes a cost matrix ^^ ∈ ℝ ^ேା^^ൈ^ெା^^ such that, ^ ^^ି^,^ି^ ൌ ^^൫ ^ min^ ^^ ^ and ^^ ଶ at time steps ^^ and ^^ respectively. It can be defined in any method like
euclidean, absolute or squared. This cost matrix ' ^^ ' is then used to trace back from ^^ ே,ெ to ^^ ^,^ which gives the best mapping of y values for the two‐time series. [00179] In the study described in the present example, DTW
was used as the performance metric. Optionally, the present disclosure
is configured to predict loads happening in a broad time horizon rather than focusing on the
exact time in which the load is happening. As a non‐limiting example, it may be more importan
t to predict the load caused by a microwave being used in a long‐haul truck, than it is to p
redict the exact point in time when the microwave is used. Using DTW the study can warp the time axi
s such that the similarity between the power load predicted and the power load profile of any te
st day can be checked and they can be compared. MCC Ref. No.: 103361‐369WO1 [00180] The data described herein regarding energy usage can
be collected using on board data loggers or OBD scanners. Optionally, the
data can be predicted using estimations based on known literature related to energy usage. L
STM can be a data hungry algorithm and hence there is a need to generate a lot of data s
ets in order to give enough input to the algorithm to learn from it. In the example study, 1
000 day’s of data is synthesized from the base data. [00181] A survey was conducted on the various drivers about
their usual activities in a 10 hour hotel period and data is recorded. An auxil
iary power usage is then generated. FIG. 4 shows the activity profile for each device used in
the cabin of long‐haul truck. [00182] It can be seen from FIG. 4 that Activity 5 and th
e Activity 2 are on throughout the 10 hours hotel period and the activity 12 (slee
p) is also done for a significant period of time.
The example driver used the microwave and coffee mak
er for very short intervals (before and once after sleeping). It was seen in the example th
at the load requirement for these two Activities are relatively higher than the rest. [00183] FIG. 5 illustrates the normalized electric power for
the 10 hour hotel period. It can be seen that the load requirement increases to
high values once in either half of the hotel period. This was likely because of the microwave and
coffee maker. [00184] Additional data sets were generated. First, the activ
ities are divided into four groups, that are, i) sleep, ii) Food, iii) Coffee a
nd iv) miscellaneous. As mentioned before, the miscellaneous activities do not have very high load
contribution as compared to Food and coffee, hence individually do not influence the total
power load profile unlike the usage of coffee maker and microwave which are rated relatively
high. Hence the combined total load MCC Ref. No.: 103361‐369WO1 requirement from the miscellaneous activities is kept
as same and is not varied throughout the 1000 different activity profile generated. Variability
in the data is introduced using the food and coffee consumption and sleeping behavior and is discu
ssed in the paragraphs below. [00185] Sleep studies of 80 long‐haul truck drivers with a
total of 400 principal sleep periods have been conducted. [Mitler et al. (1997)].
Such studies have found that though the drivers desired an average േSD sleep of 7.2 േ 1.2 hours, in reality they may average 5.34 hours. [00186] In the above study the average time off duty was l
ess than 8 hours (7.4 hours), however in the example application, the numbe
r of OFF duty hours for the driver is 10 hours. In order to accommodate for the 2 hour incre
ase, the sleep behavior of the driver is randomly generated using a Gaussian distribution ^^ ∼ ^^^ ^^, ^^^ ∼ ^^^5.5,0.5^, that is, 5.5 hours of average sleep with a standard deviation of 0.5 h
ours. FIG. 6 shows the random distribution of sleep hours for the 1000 days data set. It can be
seen that the minimum can go up to 3.8 hours while the maximum can go up to 7.4 hours. [00187] Usage of coffee makers can be variable between diffe
rent drivers, and implementations of the present disclosure can be conf
igured to consider variable usage patterns of coffee makers (and other devices). Option
ally, it can be assumed that usage of coffee makers can be performed zero number of times
to a maximum of one per hour of the awake time. Optionally, it can be further assumed th
at coffee is made in fixed intervals of 10 minutes each. It is assumed that the driver is very
likely to make coffee in the last 10 minutes of
the hotel period as well. MCC Ref. No.: 103361‐369WO1 [00188] As another example, it can be assumed that the micr
owave is used twice during the hotel period. It can further be assumed
that the microwave is used once at the start of the hotel period and again at the end of the h
otel period. The total time for this activity is done is for a fixed interval of five minutes. It c
an also be assumed that the driver does not do this activity in the last 30 minutes of the hotel
period just to account for the fact that they would be preparing for the journey and performing th
e last minute checks. [00189] To avoid the redundancy in the synthesized data for
training purpose, permutation and combination is used to calculate how
many combinations of such activities are possible with the above discussed variability in the
data. The total number of combination is equal to the number of possible unique data sets. F
or this calculation, the problem is conservatively simplified such that the sleeping activ
ity can take 5 values between 4 to 8 hours. This introduces 5 different cases where different com
binations of using microwave and coffee maker calculated. Adding up the number of combination
s of these 5 cases, it is concluded that it is possible to generate 3.8 ^^ ^ 10 unique data sets. [00190] It should be understood that the different activities
described herein, as well as the combinations of activities and assumptions are
intended only as non‐limiting examples, and that implementations of the present disclosure ca
n include different activities, combinations of activities, and assumptions. [00191] In the example implementation, the LSTM algorithm is
set up as a regression problem with the input as a sequence and the output
as the next value in the series (many‐to‐ one) in a predictor response format where the respon
se for a time step is added to the predictor and the combined becomes the predictor for
the response of the next time step. MCC Ref. No.: 103361‐369WO1 Activity prediction can be a multi‐variate problem,
which can be converted to univariate by adding the load ratings of all the active devices a
t a particular time and predicting this total power load. This 1‐D matrix would capture the info
rmation of multiple activities happening together and also would eliminate the need to create
new categories in a classification problem. Time is discretized at 1 min to produce a
1‐by‐600 matrix containing the total power load profile for a 10 hours of hotel period. The i
nput to the algorithm for training is a 25‐by‐59
9 matrix representing one day. Out of the 25 rows, 12
rows store the Time Allocation Matrix [TAM] as a 12‐by‐599 matrix, 12 rows store the
Transition Matrix [TM] in a 12‐by‐599 matrix and 1 row stores the power load profile, P is a 1
‐by‐599 matrix, hence it is fed to the algorithm
as [TAM TM P ^ ^ . The response is the power load at the next
time step. For the prediction, the activities at the first minute of hotel period of t
he test day along with the relational information is provided as a 25‐by‐1 matrix and the algorith
m predicts the power load. This value is appended to the power load prediction series and fed
to the algorithm again to predict the power load value next in line until the end of the
10 hours. [00192] The temporal information can be given in the form o
f a time allocation matrix which can store information about the probability dis
tribution of each of the activities. FIG. 8 illustrates example probability distributions of three
activities (sleep, Microwave, and Coffee maker). It can be seen that it is highly probable
that the driver is sleeping at the middle of the hotel period. Since the mean of the sleep distributi
on is 5.5 , hence, no matter when the driver starts to sleep, the chance of them being asleep at
the 5^"th " hour is very high. Also, the chances of microwave and coffee maker is high after
sleeping. This behavior is also expected as the they used when the driver is not sleeping. Also
, the probability of making coffee at the end MCC Ref. No.: 103361‐369WO1 of the hotel period is very high, this is because
in the rule based data generation (discussed in example 1), very high weightage is given to the usa
ge of coffee machine before the journey is resumed, i.e, end of the 10hr hotelling period. [00193] The transition matrix stores the relational informatio
n of the activities. It can predict the probability of the next activity in line
[Gagniuc (2017)]. Using the Markov property , an ^^‐by‐ ^^ matrix for ^^‐activities can be generated. In this ^^‐by‐n matrix, the element ^^‐ by ^^^ ^^, ^^ ∈ ^^^ would store the probability of going from activi
ty ^^ to activity ^^ at any instant. Hence, the rows of the transition matrix add up to
1. [00194] The Markov property, however, is valid only when 1
activity is being performed at a given time instance. In the hotel pe
riod of long‐haul truck, there can be multiple activities happening at the same time. This can be
handled by setting breakpoints on the time stamps and calculate the conditional probability of a
ll the activity next in line. The fundamental difference in this method is that the sum of probab
ilities of all the activities following a particular activity will be more than 1. Optionally,
normalization can be used to force the sum of the rows to be equal to 1. In the example stud
y, there are 12 activities, hence a 12‐by‐12 transition matrix was generated. As discussed herein,
in transition matrix, the rows do not add up to 1 . Such as the device corresponding to Acti
vity 2 is switched "ON" for the 10 hours hotel period, the probability it will be switched on after
a given activity is always 1. Moreover, the second column of the transition matrix is always 1.
[00195] To feed this information to the learning algorithm,
the activities that are currently being done are recognized, and the transiti
on probabilities from the activities are added together. For instance, if Activity 5, Activity
10 and Activity 11 are being done at time 't', MCC Ref. No.: 103361‐369WO1 the rows of the transition matrix corresponding to t
hese activities are added and this 12‐by‐1 matrix is transposed and appended to the input. Henc
e, for all 600 time steps, a 12‐by‐600 matrix represents the probabilities of the activities
using relational information. [00196] Optionally, implementations of the present disclsoure
can include the following steps: [00197] (1) Split the data into training and test sets in
a 9: 1 ratio [00198] (2) Create predictor ^^ ழ^:^ షభவ and response ^^ ழଶ:^ வ for the training sequences where ^^ ௫ is the length of the series. [00199] (3) Row wise append the predictor with TAM and TM,
i.e., create the input matrix. and train the network. [00200] (4) Use the trained network to predict the test day
. An initial guess of ^^ ழ^வ is 0 and a prediction ^^ ழ^வ is made. This prediction is column wise appen
ded to ^^ ழ^வ as ^^ ழଶவ and ^^ ழଶவ is made until ^^ ழ ^வ [00201] was observed to have improved the performance. The M
ATLAB neural network toolbox was used. The architecture of the LSTM is: [00202] (1) Input Layer: With 25 features [00203] (2) LSTM Layer: With 50 hidden units [00204] (3) LSTM Layer: With 50 hidden units [00205] (4) Fully Connected Layer: Multiplies the input weigh
t matrix and adds the bias vector [00206] (5) Regression Layer: Compares the mean squared error
MCC Ref. No.: 103361‐369WO1 [00207] It was seen that having a multi‐layer LSTM worked
better in capturing the minor trends in the data however having lot of LSTM
layers increased the computation time exponentially. Two LSTM layers seemed to work perfect
ly well without compromise on the computation time. For training the algorithm, optional
ly the adam solver can be used with initial learning rate of 0.005 and gradient threshold
of 1 for 250 epochs. [00208] As discussed herein, DTW can be used to characterize
the error. The number of hidden units and the period of training are chos
en as the training parameters. The number of hidden units chosen are [20 50100150] while the trai
ning period is [10 204070100140 200]. With large number of hidden units the model is able
to learn more relations between the events of the time series. While this is a desirabl
e feature, it compromises on the computation speed and the risks over fitting. Similar is the ca
se with the length of training set. FIG. 9 shows a
surface plot of the error matrix, that is the eucli
dean distance measured point wise on the warped time axis, created using dynamic time warping
of the test day power load profile and the predicted power load profile. [00209] It was seen that increasing the number of hidden un
its or number of training days alone did not improve the accuracy. Having a s
mall number of hidden units and training days did just as good of a job in the predicting
the series as very high number of hidden units and training days. The global minimum error was foun
d to be with 50 hidden units and a training period of 20 days. FIG. 10A shows the fore
cast and the prediction, while FIG. 10B shows the warped version of FIG. 10A. The warped version
is a manipulation of the time axis such that DTW algorithm finds the minimum Euclidean distance be
tween the two signals. Though in some cases warping the time axis can be undesirable to c
heck the accuracy of a prediction, however MCC Ref. No.: 103361‐369WO1 in this case however it is an acceptable metric. Th
is is because it can be more important for the algorithm to predict an event (in this case the tot
al power load value) and in more general sense the pattern, rather than the exact time it is
supposed to happen. In other words, it can be important for the supervisory control on the truck's
ECU to know if there is going to be a surge in the power demand and prepare the battery for it.
[00210] The present disclosure includes a multivariate time s
eries prediction problem is discussed and reduced to a univariate prediction
problem. A combined effect of all activities as the total power load is predicted at a given ti
me instead if predicting the individual power load ratings of the various devices inside the cabin
of a long‐haul truck. This algorithm is trained
on synthetic data generated using observations and ju
dgements from a baseline profile created from survey data. A multi‐layer LSTM with each lay
er with 50 hidden units is trained on total of 20 days. The total time required for this training
is about 4 minutes on a 32 GB RAM, 2.2Ghz clock rate and 64‐bit processor. Preference can giv
en to predicting an event (an event being a particular load value) over the exact time of the e
vent happening. [00211] Example 2: [00212] Implementations of the present disclosure include syst
ems and methods that can be used to model and predict the performance of
vehicles, for example long haul trucks. A study was performed using an example implementation o
f the present disclosure. [00213] Optionally, the example implementation can include: a
control‐oriented two‐ node and three‐node cabin model to estimate the ca
bin average air temperature at 93% accuracy for a heavy‐duty truck. MCC Ref. No.: 103361‐369WO1 [00214] Alternatively or additionally, the example implementati
on can include a vapor compression model including heat exchanger models
(using moving interface method) with 95% accuracy and compressor model (using empirica
l relations) for power estimation for the compressor work with an accuracy of 98.4% in the HV
AC system of a heavy‐duty truck. [00215] Alternatively or additionally, the example implementati
on includes a machine learning‐based model to estimate the load from the
driver behavior during the hotel period of a heavy‐duty truck. [00216] Alternatively or additionally, the example implementati
on can establish an optimal state of charge trajectory for a custom full
‐day route for a long haul application to reduce/eliminate the idling during hoteling saving up
to $40 per day in fuel cost to the owner‐ operators. [00217] The study described herein shows a control‐oriented
eHVAC and cabin models that can validated using the experimental data to me
asure their efficacy in a heavy‐duty truck. [00218] In some jurisdictions, long‐haul truck drivers are
mandated to take off‐duty time of 10 hours (referred to herein as “hoteling
) before driving. Maximum driving time is 11 hours after 10 consecutive hours off duty. Driving i
s not allowed after being on duty for 14 hours. During the hotel phase, drivers spend time in
side their trucks and idle the internal combustion engine for comfort by utilizing the heatin
g ventilation, air‐conditioning (HVAC), and other onboard appliances. A 13L engine on average co
nsumes about 0.8 gal/hour of diesel for idling while the auxiliary power unit consumes 0.5 g
al/hour. For one 10‐hour period, the average cost is about $40, which can be significant
spread across the approximately one‐million MCC Ref. No.: 103361‐369WO1 truck drivers idling overnight. An example truck, Sup
er Truck II, is a 48 V mild‐hybrid heavy‐duty truck with auxiliary loads powered by an onboard bat
tery pack. [00219] An optimal control algorithm is described herein to
charge the battery pack during the drive phase up to a certain state‐of‐
charge (SOC) level, sufficient to meet the power demands of the auxiliary load during the hotel phase
. The study described herein includes systems and methods to predict the energy consumption
in a mild‐hybrid heavy‐duty sleeper truck during the hotel period. Physics‐based grey‐
box models are developed to estimate the e‐ HVAC power consumption. E‐HVAC refers to an electro
nically controlled compressor as compared to the conventional engine‐run compressor.
For the other auxiliary loads, a machine learning algorithm was developed to predict the power
as a time series by tracking the user activity. The developed physics‐based and data‐driv
en models are validated for the experimental data of class 8 heavy‐duty truck to s
how their efficacy. These implementations of the present disclosure also generate precise load pro
files which are fed to the developed dynamic programming (DP) framework to generate the op
timal SOC trajectories. These models ultimately help the vehicle's battery pack charge onl
y up to the SOC necessary for the hotel phase during the drive time. When the vehicle is ou
t of charge during the hotel phase, these models also help in estimating the amount of idling
required to charge the battery enough to support the rest of the hotel period. This save
s unnecessary idling. As a result a cost savings of
$40 and COଶ reduction of 175lb to the environ
ment is achieved for a single heavy‐duty truck. [00220] For the past 20 years, trucks have been a prime mo
de of freight transportation which has grown from nearly 25% of th
e total ton‐mile in the US freight industry in 1980 to nearly 50% in 2020 (Margreta et
al. (2014)). MCC Ref. No.: 103361‐369WO1 [00221] Vehicles weighing over 33,000 pounds are classified a
s class 8 vehicles and semi‐trailers come under this category. These class
8 vehicles are nearly 2.5% of the total commercial vehicles. [00222] Nearly 2.5 million trucks travel long distances of 6
6,000 miles per year [Davis and Boundy (2021)]. According to the DOE SuperTruck
report, class 8 tractor‐trailers consume about 22% of the total transportation energy, which
is nearly 28 billion gallons of fuel per year [Delgado and Lutsey (2014)]. Nearly 8% of this total
fuel cost comes from the overnight idling of trucks for hoteling. [00223] For the long haul journeys (journeys more than 650
miles per trip), Federal Motor Carrier Safety Administration (FMCSA), a federal
agency with a mission to reduce commercial motor vehicle‐related fatalities and injur
ies, mandates that drivers rest for 10 hours in every 14 hour of on‐duty time [fmc (2022)]. Th
ese 10 hours in a 24 hour period are called hoteling. Nearly 1 million drivers practice hoteling
overnight [ANL (2012)]. Studies from ANL also suggest that the drivers are idling for an ave
rage of 6 hours a day [ANL (2012)][Gains (2017)]. Studies in California, on average a truck,
is idled for nearly 30 hours in a week [Brodrick et al. (2001)]. [00224] Studies in National Renewable Energy Lab (NREL) in S
todolsky et al. (2000) show that class 8 sleeper trucks idle for nearly 1,
800 hours in a year and consume nearly 838 million gallons of diesel fuel. [00225] In efforts for electrification the usage of Auxiliary
Power Units (APU) for auxiliary load is a well‐established area for freig
ht efficiency improvement. Kshirsagar (2015) demonstrated 50% energy consumption reduction in a fu
el cell‐powered APU as compared to a MCC Ref. No.: 103361‐369WO1 diesel engine. Surampudi et al. (2005) used 2.4 kW
APU to power the 42 V accessories and found a reduction in energy consumption from 407MJ t
o 78MJ for the water pump and from 400 MJ to 176.1 MJ for air conditioning (AC) system
respectively [Surampudi et al. (2004)][Surampudi et al. (2005)] . Similar methods we
re used in Redfield et al. (2006) and even better improvements were reported with a fuel cell A
PU. [00226] Controls also have played an important part in effic
iency improvements. Using optimal control Surampudi et al. (2006) explored thre
e control strategies for the Heating Ventilation and Air Conditioning (HVAC) system; (1) e
vaporator temperature control, (2) evaporator pressure control, and (3) cabin temperature
control. For a 9000 second simulation, they observed an energy consumption of 932 kJ, 1228 kJ, and 975 kJ. They concluded that with electrified auxiliaries and a control strategy i
n place, the worst‐case scenario would be nearly 1.3 kW of average power. Similar trends were
observed in an electrified bus powertrain. Campbell et al. (2012) running a hybrid electric cit
y bus for 145 hours in 11 days, an average power from ACOFF and ACON can be 11 kW and 19.3 k
W. If the mechanical components are replaced with electrical components, there is a
reduction in energy consumption reflected in 34% and 31% for AC OFF and AC ON respectively
in fuel consumption benefit. [00227] SuperTruck II is a parallel hybrid truck with an el
ectric motor and a battery pack only big enough to support the engine off hote
ling and parking lot maneuvers. While the hybrid drive doesn't support the torque split while
driving, it is big enough to support all the auxiliary loads when the driver is resting inside th
e sleeper cabins of their trucks for 10 hours. If
the battery runs out of energy during hoteling, the
engine can be idled to charge the battery back up via the electric motor. MCC Ref. No.: 103361‐369WO1 [00228] While there have been studies in model development f
or HVAC systems, there is not a complete detailed model coupled with
a cabin temperature estimator. The models in the literature are single‐node models for
the HVAC load at the cabin which do not give the ability to develop low‐level controls for
the HVAC system. Apart from that there is no model currently predicting the driver's behavior durin
g the hotel period and thereby no model exists to estimate the corresponding load. [00229] The present disclosure includes advancements in the f
ield of electrification of heavy‐duty vehicles and its impact on the auxiliary
loads. The present disclosure also includes mathematical modeling for the HVAC system driver beha
vior. The present disclosure further includes a simple control strategy to simulate the H
VAC load for a 10‐hour hotel period to estimate the instantaneous power. The present disclosu
re further includes one full day of on‐ duty/off‐duty simulation for a driver doing a long
haul trip. [00230] Hotel loads are the electrical loads that are seen
on the battery pack during the 10 hours of hotel phase of a long haul truck.
Although HVAC can contribute to nearly 30% of the total load during the drive time, it can go
up to nearly 80% when the vehicle is in the hotel phase. This is because since the engine is tu
rned off, all powertrain cooling components are also turned off. HVAC becomes a major contributo
r to this load and hence predicting it can help the supervisory controller to prepare for the l
oad ahead of time. [00231] For one day of travel on a long‐haul transit. The
driver is allowed 14 hours of on‐duty time in which he/she takes two 30 minutes
of rest and after the on‐duty time, takes 10 hours of rest. To save fuel by eliminating idling,
the battery pack needs to support the hotel loads. For this, the battery should be (1) big enou
gh to take the highest possible load, and (2) MCC Ref. No.: 103361‐369WO1 charged only enough to support the load when it is
not expected to be the highest. Regeneration from braking is not always sufficient to
charge the battery during the 14 hour on‐ duty driving period, hence the engine needs to kick
in at times. This can be leveraged to operate the engine at optimal operating points to ma
ximize the engine and electric motor/generator efficiency. An optimal control algorith
m developed in Singh et al. (2022) provides an optimal state of charge trajectory for a
defined time horizon. [00232] The challenge for deploying said algorithm is to pre
‐define all the inputs in the time horizon which includes the auxiliary load expect
ed during the hotel period. Eliminating the engine idling can ultimately result in savings of ne
arly $40 per hotel period assuming 10 hours of hotel phase in a class 8 sleeper cab with a 13
L diesel engine at $5/ gal cost of diesel fuel a
nd saving nearly 175lbs of CO ଶ from being released into the atmosphere from
a single truck per day. This saving will be significant when the fleet
of heavy‐duty vehicles is being considered. [00233] Optionally, the present disclosure can include energy
modeling. Khuntia et al. (2022a) and Khuntia et al. (2022b). The energy estim
ated can be provided as an input to the optimal control algorithm which enables the battery t
o be charged to the SOC required at the battery pack before the hotel period starts while th
e vehicle is in motion. FIG. 11 presents an overall modeling approach, according to an implementat
ion of the present disclosure. Optionally, the present disclosure includes a physics
based modeling approach for the modeling of HVAC components like heat exchangers, i.e., the c
ondenser and the evaporator, compressor, and cabin of a sleeper cab, and using it in tandem
with a machine learning model that predicts the loads from the activities of a driver when the
vehicle is in hotel phase. When the load MCC Ref. No.: 103361‐369WO1 required exceeds the battery capacity, the optimal co
ntrol also provides the instance the engine should idle to charge the battery and by how
much. [00234] Hotel loads are electrical loads that are seen on t
he onboard battery pack (e.g., during the 10 hours of hotel period of a lo
ng haul truck). Although HVAC can contribute to nearly 30% of the total load during the drive time,
it can go up to nearly 80% when the vehicle is in a hotel period. This is because since the en
gine is turned off, all powertrain cooling components are also turned off. HVAC becomes a major
contributor to this load and hence predicting it can help the supervisory controller to
prepare for the load demands ahead of time. [00235] As used herein, the following terms are defined: [00236] ^^ ^ Heat transfer coefficient between tube wall and ref
rigerant per unit area ^W/m ଶ /K^ [00237] ^^ Friction Coefficient; [ 00238] ^^ Density of the refrigerant ^ kg/m ଷ^ [00239] ^^ ^ Inner diameter of the tube ^m^ [00240] ℎ Specific enthalpy ^J/kg^ [00241] ^^ Pressure in HX^pa^ [00242] ^^ ^ Bulk temperature of refrigerant ^K^ [00243] ^^ ௪ Temperature of tube wall ^K^ [00244] ^^ Velocity of the refrigerant flowing along the tubes
^m/s^ [00245] As used herein, additional terms were defined for th
e cabin model: [00246] A Surface area for heat transfer ^m ଶ ^ [00247] ^^ ^ Heat capacity at constant pressure ^J/K^ MCC Ref. No.: 103361‐369WO1 [00248] ^^ Mass ^kg^ ^^ Heat transfer ^W^ [00249] ^^ Temperature ^K^ [00250] As used herein, additional terms were defined for mo
deling fluid dynamics: [00251] ^˙^ Mass flow rate ^kg/s^ [00252] ^^ kinematic viscosity ^m ଶ /s^ [ 00253] ℎ Convective heat transfer coefficient ^ W/m ଶ /K ^ [00254] ^^ Themal conductivity ^W/K^ [00255] ^^ ^^ Nusselt number [00256] Pr Prandtle number [00257] Re Reynolds number [00258] ^^ Flow speed ^m/s^ [00259] ^^ Surface width ^m^ [00260] Cabin model equations [00261] GHI Global Horizontal Irradiance ^W/m ଶ ^ [00262] h Convective heat transfer ^J/kg^ [00263] As used herein, additional terms were defined for mo
deling fans: [00264] Φ ^ Flow rate coefficient [00265] [00266] ^^ Number of Fans [00267] ^^ ^ ^ Mean fan radius ^m^ ൌ ^ ^ ^^ ଶ ^ ^^ ^ ଶ ^, ^^ ௧ is the fan tip radius, ^^ ^ is the hub radius. [00268] ^^ ᇱ Hub ratio, ൌ ^^ ^ / ^^ ௧ MCC Ref. No.: 103361‐369WO1 [00269] As used herein, additional terms were defined for he
at exchanger modeling: [00270] ^∗^ ^ Property at the condenser; ^∗^ ^ Property in the hot side; ^∗^ ^ Property of cabin air ^∗^ ^ Property at the evaporator; ^∗^ ^ property at saturated vapour line; ^∗^ ^ Property in the hot side ^∗^ ^ Property at saturated liquid line; ^∗^ ^ Properties averaged through HX; ^∗^ ^ property of the surface; ^∗^ ௩ Volumetric property; ^∗^ act Actual property; ^∗ ^ air Property of air; ^∗^ ^^^ Property of ambient; ^∗^ ^^^ Property of the fin; ^∗^ ^^^^ Isentropic property; ^∗^ ref Property of the refrigerant; ^∗^ win Property of the window; ^^ Heat transfer coefficient ^ W/m ଶ K ^ ; ^‾^ Void fraction; ^˙^ Heat transfer in/out of refrigerant from cross‐flo
wing fluid ^W^; ^˙^ disp Volumetric displacement ^m ଷ ^; ^^ Heat transfer effectiveness; ^^ Efficiency ^^ ^^ specific heat capacity at constant pressure ^J/kg/K^;
F Air‐Structure surface area ratio ℎ Specific enthalpy ^ kg/m ଷ^ ; ^^ Polytropic constant; ^^ ^^ ^^ Number of transfer units; ^^ Overall heat transfer coefficient ^W/m ଶ K^; ^^ Volume of heat exchanger ^m ଷ ^; ^^ Specific volume ^ m ଷ /kg ^ ; ^^ Vapor quality; ^^ Vapor quality. [00271] Additionally, as used herein, the following abbreviati
ons are defined: 21CTP (21st Century Truck Partnership); ^^ ^^ (Air Conditioning); ^^ ^^ ^^ (Auxiliary Power Unit); ^^ ^^ ^^ (Computation Fluid Dynamics); DOE (Department Of Energ
y); ^^ ^^ ^^ (Dynamic Time Warping); ^^ ^^ ^^ (Electronic Expansion Valve); FMCSA (Federal Motor
Carrier Safety Carrier); ^^ ^^ ^^ ^^ (Heating Ventilation and Air Conditioning); ^^ ^^ (Heat Exchanger); ^^ ^^ ^^ ^^ (Long and Short term memory) and MHDV (Medium Heavy Duty Vehicle). [00272] Additionally, the following terms will be understood
by those of skill in the art: NN Neural Network; ^^ ^^ ^^ Ordinary Differential Equation; PDE Partial Differe
ntial Equation; ph Pressure Enthalpy; ^^& ^^ Research and Development; RAM Random Access Memory
; ^^ ^^ ^^ ^^ MCC Ref. No.: 103361‐369WO1 Root Mean Squared Error; ^^ ^^ ^^ Recurrent Neural Network; ^^ ^^ ^^ Revolutions Per Minute; ^^ ^^ ^^ State of Charge; TAM Time Allocation Matrix; TM Tran
sition Matrix; ^^ ^^ ^^ Technology Readiness Level; and US United States. [00273] In the HVAC system, there are two HXs, a condenser
on the ambient side, and an evaporator interacting with the control volume of
interest (FIG. 19). For an AC application, a high‐temperature refrigerant enters the condenser as
a vapor. It enters a two‐phase region and exits the condenser as a sub‐cooled liquid. This s
ub‐cool liquid is then depressurized using an expansion valve before it enters the evaporator. This
depressurization helps the refrigerant to absorb higher heat content at the evaporator. The re
frigerant then goes into the compressor where a mass flow rate is is controlled by adjustin
g the compressor speed. [00274] Implementations of the present disclosure can include
a cabin model. A vapor compression cycle system interacts with the cabin's i
nternal air to cool or heat it up based on its application as AC or heat pump. It hence becomes im
portant to predict the temperature of the cabin. In this work, a grey‐box modeling approach
is adopted to develop a model that can work for different cabins (by considering their size) by
being calibrated against a universal test procedure. A literature survey helps to identify the
right level of complexity to balance the accuracy of the temperature prediction and the comple
xity of the model. [00275] For developing the cabin model, the quantities of in
terest are the average temperatures of the roof and the windows, the HVAC
air temperature from the interior vents of the cabin, and the temperature of the cabin's intern
al air with those are shown in FIG. 13. [00276] This grey box model formulation takes inspiration fro
m the heat transfer equations and is calibrated using the data. The obje
ctive of the model is to predict the cabin MCC Ref. No.: 103361‐369WO1 average air temperature using readily available weathe
r data, i.e., the solar GHI ^W/m ଶ ^, ambient temperature ^ ∘ C^, and additional wind/vehicle speeds. [00277] The elements chosen in this heat transfer prediction
are cabin air, windows, walls, and HVAC. While there are multiple sources of
external heat to influence the inside cabin temperature, describing all can be cumbersome, increas
ing the complexity of the model, but also hard to calibrate using data from standard test
procedures, and at the same time add little benefit to the prediction accuracy. [00278] Implementations of the present disclosure can include
a two‐node model. In a class 8 truck cabin, the windows take up the majori
ty of the area. A two‐node model is developed with cabin internal air as one element and
the window as the other. The heat transfer coefficients and the specific heats have bee
n chosen as calibration parameters. Some of these calibration parameters can be derived and s
implified using empirical relationships to help guide the calibration process and also can be
used to guess their initial values. [00279] In this example model, the effect from the roof has
been ignored. A study in Okaeme et al. (2021) suggests that insulating the ca
bin to reduce heat transfer from the walls would still have a massive impact. Hence a two‐nod
e model is proposed accounting for the temperature dynamics of the cabin's internal air and
the temperature dynamics of the windows. [00280] Concepts from fluid dynamics are used to capture the
heat exchanged from the effect of wind. The Nusselt number gives a rela
tionship between the heat transfer coefficient from the convection to that from conducti
on and is highlighted in Eqs.(1, 2, and 3). MCC Ref. No.: 103361‐369WO1 ℎ ^^^ ൌ ே௨ೌ^^^ೌ^^ ^ surface ^1^ ^^ ^^ ^^ ^/ଶ cross‐flowing wind is found by lumping all the con
stants in ' k ' as shown in Eq. (4) and Eq. (5).
Another term ℎ ^^^ is added that accounts for the natural heat
transfer coefficient in the absence of the wind. ^ /ଶ [ 00283] ℎ^^^ ൌ ^^ ^^^^^ ^4^ ^ ^^ ^.ହ captures the temperature dynamics of the window (abbr
eviated as 'win') and the cabin is shown in Eq (6). The term ^^ wind is the relative velocity of the wind and the
vehicle. The direction of the wind is not considered in the exam
ple study. The temperature of the air from the vents for the HVAC system in the cab is used
as ^^ ^௩^^ . ^ ^ ௗ் ^^^ ௪ ^^ ௗ௧ ൌ ^^^ ^^ ^^ ^^^ ^ ^^ ௪^^,^ ^ ^^ ^ െ ^^ ௪^^ ^ ^ ൫ ^^ ௪^^,^^^ ^ ^^ ௪^^ ^^ ^/ଶ ௪ ^^ௗ ൯^ ^^ ^^^ െ ^^ ௪^^ ^ calibrated using a gradient‐free optimization method
reducing a custom cost function (which is the root mean squared error between the simulated an
d the experimental values). The objective function in this dynamic optimization proble
m has been defined as the weighted sum of the two nodes. To bias the prediction accuracy t
owards the primary quantity, that is, cabin temperature, different weights were tried. However, it
was found that having an equal weight for all the parameters gave equal benefit in predict
ion as any other weighting combination. This MCC Ref. No.: 103361‐369WO1 is because while biasing the prediction to favor the
cabin temperature, authors consequently ignore the other node. Because of the coupling in t
he model equations, having a bad prediction for the other node would ultimately impact the accur
acy of the primary node too. [00287] It was observed that Root Mean Squared Error (RMSE)
values are within the allowable range and the window node is predicted as
desired. Although, cabin air temperature shows a deviation of 4 degrees which is a high val
ue. A positive error implies that the temperature of the cabin is predicted to be lower t
han the actual. This is suggesting that there is a source of heat in the cabin which is ignored.
This is designated as the wall of the truck. Hence, a three‐node model is proposed to capture t
he temperature of the roof and is described below. [00288] This estimation is useful when the vehicle is statio
nary, i.e., during the hotel phase. However, the effects of wind or a moving veh
icle on the temperature of the cabin are also vivid. Due to the disparity in the heat transf
er coefficients between the walls and the windows, the vehicle while moving might not affect t
he heat of the cabin through walls, but it might affect the windows. [00289] Although the window covers most of the heat convecti
on to the cabin's internal air. Because of the solar GHI, the roof on
a sunny day can get really hot. This might force the roof element to also have an impact on t
he temperature of the cabin. To study this effect a new node is added to the 2‐node model f
or the roof of the truck and the updated model is presented in Eqs ^7,8, and 9) and the imp
rovements from a 2‐node model are studied. MCC Ref. No.: 103361‐369WO1 ^^ ௗ் r oof roof ௗ ௧ ൌ ^^^ ^^ ^^ ^^^ ^ ^^ roof ,^ ^ ^^ ^ െ ^^ roof ^ ^ ൫ ^^ roof,amb ^ ^^ roof ^^ ௪ ^ ^ .ହ ^ ௗ ൯^ ^^ ^^^ െ ^^ ௪^^ ^^7^ ^ ^ ௗ்^^^ ^^ ^ ^^ ^^ ^^ ^ ^ ^ ^ model with an added vector of the roof temperature.
Each element is isolated to find the right parameters for the node as long as they can be ind
ependent of each other. Since the window and roof can be, the values found for ^^ ௪^^,^ and ^^ roof,c in separate calibration and reused while calibrating for the parameters for the cabin node. F
or fine‐tuning, after calibrating all the node parameters separately, a calibration was run again, w
ith all parameters combined in a single calibration. A biased‐based cost function was define
d just like the cabin model with two nodes and similar results were observed. [00292] Fig. 14A illustrates example cabin temperature error
for an example implementation of the present disclosure that was stu
died. FIG. 14B illustrates example temperature difference for an example implementation o
f the present disclouse. FIG. 14C illustrates an example error histogram for an example
implementation of the present disclosure. [00293] FIG. 14D illustrates example window temperature error
for an implementation of the present disclosure that was stu
died. FIG. 14E illustrates example temperature difference for the example implementation
of the present disclosure. FIG. 14F illustrates an example error histogram for an example
implementation of the present disclosure. MCC Ref. No.: 103361‐369WO1 [00294] Fig. 15A illustrates example cabin temperature error
for an example implementation of the present disclosure that was stu
died. FIG. 15B illustrates example temperature difference for an example implementation o
f the present disclosure. FIG. 15C illustrates an example error histogram for an example
implementation of the present disclosure. [00295] FIG. 15D illustrates example window temperature error
for an implementation of the present disclosure that was stu
died. FIG. 15E illustrates example window temperature difference for the example implementation
of the present disclosure. [00296] FIG. 15F illustrates example roof temperature error f
or an implementation of the present disclosure. FIG. 15G illustrates example
roof temperature difference for an implementation of the present disclosure. FIG. 15H il
lustrates an example error histogram for an implementation of the present disclosure. [00297] A comparison of the proposed models illustrated in F
IGS. 14A‐14F and FIGS. 15A‐15H is given in FIG. 16. FIG. 16 illustrates
RMSE comparison of 2‐node model and 3‐node model. It is observed that the 3‐node model illu
strated in FIGS. 15A‐15H has RMSE of 0.3 ∘ C as compared to the 2 ‐node model illustrated FIG. 14A
‐14F while adding five additional calibration parameters in the model. [00298] The study included model validation. To compare the
above given 2 ‐node and 3 ‐node models, a different vehicle was used in t
he test cell. The idea behind this validation is to
see which model holds more accuracy across different
vehicles. Unlike the previous calibration, the model may not require the majority of the data
set to calibrate the model parameters for this new vehicle since the parameters are already ca
librated for a similar vehicle. Different sizes MCC Ref. No.: 103361‐369WO1 of data from the data set are chosen to study how
quickly these two models can adapt. Three instances were evaluated where 10%, 30%, and 60% of the data for calibration, and the
results are reported in FIG. 17. [00299] As the amount of data used for calibration is incre
ased as expected the error reduces (FIG. 12). The temperature of the windows wa
s consistently accurate through all the calibrations and both the models. There is a massive
improvement in roof temperature tracking with increasing data. It can be seen that the error
is within േ5 ∘ C with the mean at nearly zero for all the nodes. The error at the cabin seems to
be increasing with time. This is due to the lack
of windshield temperature data. The windshield is dir
ectly impacted by the solar lamp (in the test cell) and hence the sun. Optionally, the side
window temperature can be used instead. Since the side windows are not impacted as much by
the heat of the sun as the windshield, they can represent a lower temperature. The cabin takes i
n heat directly from the windows too, a lower temperature from the windows indicated a lower
temperature for the cabin. Hence, adding the window temperature would fix the increasin
g cabin temperature error. [00300] When 60% data is used in calibration, both the 2‐
node model and 3‐node model perform about the same for the cabin temperatu
re estimation. However, when the data is reduced to 10% the 3‐node model performs better
. This implies that the 3 ‐node model, a smaller data set for a given vehicle will be requir
ed to obtain the same accuracy as the 2‐node model with larger data. The number of calibration pa
rameters, in this case, seems to have a smaller effect than previously assumed. FIG. 18 shows
the final prediction of the cabin temperature, window temperature, and roof temperatures
from the model using the new data set. MCC Ref. No.: 103361‐369WO1 [00301] Implementations of the present disclosure can include
a vapor compression cycle model. Air conditioners and heat pumps can wor
k on a principle described by the vapor compression refrigeration cycle or vapor compression c
ycle as shown in FIG. 19. In this method, cooling/heating is achieved by exchanging the heat of
a system with the environment via a refrigerant. Most vehicles and households use a refri
gerant called R134a and it is used in this application as well. [00302] FIG. 19 shows the components in a vapor compression
cycle and FIG. 20 shows its equivalent on a pressure‐enthalpy (ph) di
agram. The area between the two solid lines is the two‐phase region. For an ^^ ^^ application, the evaporator module interacts with
the space that needs to be temperature controlled. The refriger
ant enters the evaporator at point 3 as low‐pressure and low‐temperature two‐phase fluid
where it absorbs the heat to exit as a superheated gas ‐ point 4 ‐ from the relatively
hotter cross‐flowing air through the evaporator. The hot refrigerant then reaches the condenser at a
higher pressure and temperature as a superheated gas (point 1) and a relatively cooler cr
oss‐flowing fluid absorbs this heat changing the refrigerant to a two‐phase fluid and ultimately
to sub cool liquid as the refrigerant cools down. [00303] The Electric Expansion Valve (EXV) at the exit of t
he condenser is used to decrease the pressure of the refrigerant to lower it
s boiling point for effective heat transfer and achieved this effect at a constant enthalpy. The pre
ssure difference achieved by the compressor pushes the refrigerant to a higher pressur
e by commanding a flow rate. This is represented by the green line. The work done to ach
ieve this delta pressure is our point of focus. Compressor work is a function of the pressure
ratio ^ ^^ ^ / ^^ ^ ^ and compressor speed. A MCC Ref. No.: 103361‐369WO1 model is hence required that can represent the press
ure of the refrigerant at the condenser and the evaporator and this is done by the HX mode
l. [00304] Large‐scale Computational Fluid Dynamics (CFD) tools
have been useful in modeling the complicated state transitions in fluids
in small control volume flows also shown in Altwieb et al. (2020) and Fukuchi et al. (2019). Wh
ile modern computation resources help very large CFD problems to be solved faster and give ver
y accurate results (still slower than their 1D counterparts), they can be overkill for control objec
tives, and relatively large computation times for these problems are undesired. 1D lumped pa
rameter modeling also offers accurate solutions with faster computations. These models provi
de enough accuracy for any control development and validation while providing a reduction
in computation. One application in the performance evaluation of air‐cooled condensers is s
hown in Ge and Cropper (2005). In (Steinstraeter et al. (2022), Widmer et al. (2022)),
the authors proposed a technique to control the effect of cabin heating on the range and lifeti
me of electric vehicles and onboard battery packs. [00305] Dynamic equations can describe the flow of a phase
changing fluid in an HX. He (2005) and Ge and Cropper (2005). Both methods s
tart with the first principles ‐ using finite volume equations for the conservation of energy, mome
ntum, and mass as Partial Differential Equations (PDEs). To account for the changing phase
of the fluid, they divide the HX into zones representing different phases, i.e., liquid, two‐phas
e, and gas. [00306] He (2005) defines a moving interface lumped parameter
model in which the HX is divided into three sections/zones: two‐phase;
superheat; and sub cool zones with lengths L1, L2, and L3. They integrated the PDEs for the c
onservation of mass, energy, and momentum MCC Ref. No.: 103361‐369WO1 over the length of the HX to form a set of Ordina
ry Differential Equations (ODEs) in pressure and void fraction. Furthermore, Ge and Cropper (2005)
split the two‐phase region into two regions at a 3: 2 ratio because of the rapid rise in the heat tran
sfer. With this approach, the error in their models was േ10% of the experimental
value. [00307] The present disclosure includes HX models. Zhang et
al. (2015). The HX models developed in the present disclosure can suppor
t energy estimation during the hotel period of a class 8 truck using rule‐based control
. [00308] The present disclosure includes a fundamental approach
that can be incorporated in a controller [Lustbader et al. (2011)
, Richter (2008) Tummescheit et al. (2005), Tummescheit (2002)]. A grey‐box modeling approach is
disclosed hereinand some calibration parameters are included like the heat transfer coeffi
cient of an HX, which itself is calculated using the e‐NTU method [Browne and Bansal (1998)].
CoolProp, RefProp [McLinden (1998)], is used to calculate internal fluid properties as needed
. The present disclosure can include approaches to calculate the compressor work by assumi
ng polytropic work between two pressures at the HX with isentropic correction. [00309] The data that is used for calibration comes from th
e same data set as that used in cabin model calibration. Example quantities a
re shown in FIG. 21. The evaporator air in and out temperatures are the temperatures of the rec
irculating air fed to the evaporator from the cabin and the air blown from the AC vents on
the dashboard respectively and the corresponding air flow rate is shown in FIG. 22A al
ong with the condenser cross air flow rate (FIG. 22B) which is placed in front of the radiator
module and is air‐cooled through the ram air effect. The radiator fans control the cross‐air flo
w rate at the radiator face. While the study can MCC Ref. No.: 103361‐369WO1 use the air flow rate at the face of the radiator,
the metric may not be available from the test. Since the test captures the radiator fan RPM the ex
ample implementation can use that to calculate the air flow rate using the Eq. 10. It c
an be seen from Eq.10, the airflow rate depends on the density of air, RPM of the radiator fan, an
d some design parameters of the radiator which are not available. To bridge this gap, the st
udy used data Wang et al. (2014) for a similar size radiator and use appropriate scaling. మ ఘ య ᇲమ [00310] ^˙^ ସగ ೌ^^^^ ^^ே^ ^ି௩ ^ ൌ ^^ ^ ^ା௩మ ^ (10) radiators used in vehicles of similar size [Kenworth
and Peterbilt (2012)], and then ram air speed ^mph^ is converted into air flow rate ^kg/s^
considering the loss in flow rate at the grill of
the vehicle. Using the information provided in Wang
et al. (2014), a correlation is established between the air flow rate and the radiator RPM and
then RPM is converted into the flow rate for the condenser air flow rate and is shown in FI
G. 22A which shows the evaporator flow rate, and FIG. 22B which shows the condenser flow rate. T
otal coolant flow rate is illustrated in FIG. 22C. [00312] In Supertruck II, an example strategy is to maintain
a 5 ∘ C sub‐cool and a 10 ∘ C super‐heat. However, from the test data procedure,
the EXV and the compressor speed are manually controlled and sometimes they counteract each
other to produce inconsistent superheat and subcooling. Hence, for the right calibr
ation, this information is fed directly into the model. The fluctuations in the data from the ti
me at ^^ ൌ 20 minutes to ^^ ൌ 60 minutes are due to engaging and disengaging an AC clutch that r
egulates the compressor speed. The refrigerant flow rate is captured in FIG. 22B. MCC Ref. No.: 103361‐369WO1 [00313] In order to represent the pressures at the HXs, som
e assumptions can be used reduce the computation effort without significant comp
romise on accuracy. These assumptions pertain to the HX component and the complete vapor
compressor cycle as well. [00314] The refrigerant can be described completely with Pres
sure ^ ^^^ and average vapor quality ^ ^^^ in the HX. The heat is exchanged only through
walls of the HX lateral to the direction of flow of refrigerant. There is no heat
loss in the direction of the flow of the refrigerant. The walls are thin enough to not have
any loss of heat at the walls and the wall can be lumped with the refrigerant. [00315] Heat transfer coefficient across the heat exchanger i
s constant. [00316] Refrigerant R134a does not deviate from its propertie
s as shown in the ph graph in FIG. 20. [00317] The pressure remains constant and there is no fricti
onal loss in the HX. This allows the pressure of the refrigerant at the HX to
be represented as a single state P. ൫ ^^ ^ ൌ ^^ ^, in ൌ ^^ ^, out , ^^ ^ ൌ ^^ ^, in ൌ ^^ ^, out ൯ with no change in the enthalpy of the refrigerant.
^ℎ ଶ ൌ ℎ ଷ ^ [00319] The Superheat ^SH^ and subcool ^SC^ can be calibrate
d to achieve a constant superheat temperature Δ ^^ ௌு ൌ ^^ ସ െ ^^ ^ ^ ^^ ^ ^ ൌ 10 ∘ C and a constant subcool temperature Δ ^^ ௌ^ ൌ ^^ ଶ െ ^^ ^ ^ ^^ ^ ^ ൌ 5 ∘ C. [00320] Implementations of the present disclosure include a m
oving interface mathematical model. MCC Ref. No.: 103361‐369WO1 [00321] The fundamental PDEs for the conservation of mass an
d energy are reduced, inspired from He (1996), and integrated over the len
gth (L) of the ^^ ^^. The final form is proposed based on average vapor quality ^ ^^ ^ ^ and the two‐phase pressure (P) in Khuntia e
t al. (2022c). The flow rate of the refrigerant entering a
nd exiting the refrigerant is assumed to be constant hence reducing the two ODEs into one in Pr
essure as compared to that proposed in Singh (2021). The equation proposed in Singh (2021)
is shown in eq(11) and the final reduced form is shown in eq(13). ^ 0322] ௩మ ^^ ௗ^ ௗ ^ ^ మ ^^ ௗ^ ^ ൌ ^˙^^^ℎin െ ^˙^out ℎout ^ ^^ ^ ˙ 11^ [0 ^ ௧ ௩^ ௗ௧ [00323] [00324] ^˙^ in ൌ ^˙^ out [00325] This reduces the above equations into one equation r
epresenting HX pressure dynamics: [00326] ^ ௩ ^మ ^ ^^ െ ^^ ா ^ ௗ^ ௗ ௧ ൌ ^˙^^ℎ in െ ℎ out ^ ^ ^˙^ (13) is represented as ^˙^. Going by the convention of the direction of f
low of heat, the ^˙^ for the MCC Ref. No.: 103361‐369WO1 evaporator and the condenser takes opposite signs. It
can be calculated using the e‐NTU method as it allows us to calculate the exit proper
ties of the refrigerant using only the inlet properties and is highlighted in Eq. (16) ^^ ൌ ೌ^˙^ ൌ 1 െ ^ ିே்^ ^ ˙^౮ ^ ^16^ ൯ ^17^ [00330] ൌ ^min [00331] The calibration activity for the HX is to estimate
the overall heat transfer coefficient, UA. The UA is calculated based on the
Eq. (20). It is highly dependent on the HX geometry which is constant for a particular HX and
is provided in the supplier datasheet. The authors calibrate this as a lumped value. Although t
he heat transfer coefficient ^^ is a function of the crossflowing air speed, hence is not constant
. Due to limited data points and a limited range of operation, it can be assumed as a constant
. [00332] ^^ ^^ ൌ ^^^1 െ ^^ ^^^ ൫1 െ ^^ ^^^ ൯൧ ^^ ^ (20) transfer (shown in Eq. (22) for the evaporator and
Eq. (21) for condenser) the study determined the exit air temperatures. [00334] ^˙^^ ൌ ^˙^^^^ ^^ ^^^^^൫ ^^^^^,^^ െ ^^^^^,^௨௧൯ ^21^ [00335] [00336] ^^^^^, out ൌ ^^^ ^ ൫ ^^^^^,^^ െ ^^^൯ ^^ିே்^^ ^23^ ି ே்^ ^ ^^^^, out ൌ ^^^ ^ ൫ ^^^^^,^^ െ ^^^൯ ^^ ^ ^24^ MCC Ref. No.: 103361‐369WO1 [00337] The equation Eq. (25) represents the final model equ
ation by using Eq. (19) to Eq. (24) along with the mention of the calibration
parameters in red. [00338] ^ ௩ ^మ ^ ^^ െ ^^ ா ^ ௗ^ ௗ ௧ ൌ ^^ ^ ^˙^ ^ ℎ^^ െ ℎ^௨௧ ^ ^ ^^ ^˙^^^^ ^^ ^^^^^ ^ ^^^^^,^^ െ ^^ு^ ^ parameter 'M' helps in better capturing the transienc
e in the data. The UA is the overall heat transfer coefficient as discussed earlier in this wor
k. ^^ is a correction term for the inaccuracies that may have come about due to the estimation of
flow rates. The calibration parameter ' ^^ ' acts as the larger knob for tuning the heat exchang
e pressure while the UA acts as the fine‐ tuning knob. [00340] At the condenser, the refrigerant exits as a sub‐c
ooled liquid. While going through the EXV, its pressure drops at a constant e
nthalpy and enters the evaporator as a two‐ phase fluid. The pressure and the sub‐cool at the
condenser are used to identify the enthalpy, and the pressure at the evaporator is used to pinpo
int the properties of the refrigerant from the ph curve (FIG. 20). [00341] The refrigerant properties are calculated using a fre
e property open source software COOLPROP using the python wrapper Bell et a
l. (2014) in MATLAB. [00342] The refrigerant exits the evaporator as a superheated
gas. This helps in avoiding any liquid entering the compressor. However,
the degree of superheat is kept small as very hot and dry gas entering the compressor can al
so sabotage the machine. MCC Ref. No.: 103361‐369WO1 ^^ ^ ^^ଶ ൬ ^^^ െ ^^^ ^^^ ^ ^ ^ ^^ ^^^ ^ ^ ൌ ^^^^ ^˙^^^^ ^ ℎଷ െ ℎସ ^ ^ ^^^ ^˙^^൧ ^26^ ^,^ ^ ^^ [00343] relationships and using one steady state data point
for the inlet and exit temperatures of the refrigerant. Then ^^ ^ and ^^ ^ are manually tuned. The equations are executed
in the MATLAB/Simulink environment and the results are shown
in FIGS. 23A‐23D. The results of the simulation and the data did not match as expected.
There are inconsistencies that can be seen. One arises mainly during the AC clutch ON‐OFF acti
on. It is also reasonable since the model being lumped parameter working with the given assumpt
ions can capture the bulk average behavior but not the behavior with the high transien
ce. The second inconsistency can be seen after t ൌ 60 min. The pressure trajectory predicted by the mod
el is the same as that of the data but shifted by a few units up. To match the
shift in pressure, the model is re‐tuned and another set of parameters is used and the results a
re shown in FIGS. 24A‐24D. While one set of values for ^^ ^^ ^ , ^^ ^ , ^^ ^ satisfied the transient behavior while another set
of parameters satisfied the steady‐state values. [00344] This discrepancy is further investigated. Matching the
pressure dynamics from the data to the refrigerant flow rate (FIG. 22B) it
was seen that the flow rate is increased occasionally when the compressor engages and disengage
s. This should indicate the pressure to decrease at the suction. This is predicted by t
he model but not indicated by the data. At time t ൌ 60 min, the oscillations are stopped, the data displ
ays the oscillations to have stopped at a MCC Ref. No.: 103361‐369WO1 higher pressure, but the model indicates the oscillat
ions to be stopped at the lower pressure value. This explains the shift in the pressure indic
ated by the model and the data. [00345] The model does not seem to replicate the data exact
ly, but the pattern and trends predicted by the model and the data are simi
lar. Possible sources of error can include: missing information in the data including which in t
he modeling should reduce the error and/or that the assumptions in the model make it incapable
to capture the pressure dynamics of the HX pressure accurately. [00346] To check the validity of the example model and iden
tify the source of the error, another model (Zhang and Canova (2013)) was c
hosen and compared to the existing results later. This model relies on the time‐invari
ant mean void fraction in the two‐phase region of an HX to calculate its pressure. This model is
verified against experimental data and so serves as a good baseline for comparison. The model however
demands an additional calibration parameter for the weight of the refrigerant at the
HX and the definition of the void fraction for this model is estimated (Dandekar and Brooks (2016))
and is shown in Eq. (29): ^‾^ ^ ൌ ^ ^ ି௩ೌ ^ ௩ ೌ ^^ି௫య^^^ି௩ೌ^మ ln ^ ^^^ ^ ^ 1 െ ^^^ ^ ^^ଷ ^ (29) segments of the data. FIG. 23A illustrates initial n
o flow, FIG. 23B illustrates transience FIG. 23C illustrates flow rate oscillations; FIG. 24D illustrat
es steady‐state with calibration parameter set 1. [00349] It can be seen from FIG. 23A‐24D that in the fir
st segment where the flow rate is 0 , both the proposed model and Zhang et.al. mo
del do not capture the increase in the MCC Ref. No.: 103361‐369WO1 pressure. Following that, both models closely capture
the transience and both reach the same steady state value at nearly the same time as indic
ated by the data. Both models do not capture the oscillations but show the same trend. The steady
‐state condition at the end is not tracked by either of the models but both show a similar tr
end and are nearly identical to each other. The steady state is however reached a little before
but overall, the models are consistent with each other and the last portion of the data. [00350] Breaking down the response segment‐wise can also be
used to show clearer results, as shown in FIGS. 24A‐24D. The two models
match each other but they can be disconnected from the data at some points. While the
oscillations are still captured, they are not captured with the same magnitude however the ste
ady state is captured by both models accurately. They can hence be of use to capture the
overall averaged pressure which can be used for energy estimation. [00351] FIGS. 24A‐24D illustrate calibrated evaporator models
comparison on 4 segments of the data. FIG. 24A illustrates initial n
o flow. FIG. 24B illustrates transience. FIG. 24C illustrates flow rate oscillations. FIG. 24D illustrat
es steady‐state with calibration parameter set 2. [00352] FIG. 25 illustrates a summary of the RMSE for each
segment of a evaporator pressure using both models (proposed model and Zhang
model). It is clear that the proposed model shows a better response in representing the re
al data. [00353] This solidifies the accuracy of the evaporator model
and this is then used and calibrate the condenser. The refrigerant enters the c
ondenser as a superheated gas and starts turning into two‐phase at the saturated vapor line.
It liquefies as it moves towards the MCC Ref. No.: 103361‐369WO1 saturated liquid line where it changes completely int
o liquid. Thereafter it changes to a subcooled liquid and then the refrigerant enters the
EXV to be depressurized for the evaporator. [00354] FIGS. 26A‐26F illustrate experimental data compared
to a proposed model error study on calibration parameter set 1, according
to the studied implementation of the present disclosure. FIG. 26A illustrates a comparison
of experimental data to the proposed model for segment 1. FIG. 26B illustrates an error
histogram for segment 1. FIG. 26C illustrates a comparison of experimental data to the proposed mo
del for segment 1. FIG. 26D illustrates an error histogram for segment 1. FIG. 26E illustrat
es a comparison of experimental data to the proposed model for segment 1. FIG. 26F illustrates a
n error histogram for segment 1. [00355] FIG. 27 illustrates a comparison of RMSE error for
models of example condenser pressures. [00356] FIGS. 28A‐28F illustrate experimental data compared
to a proposed model error study on calibration parameter set 2, according
to A studied implementation of the present disclosure. FIG. 28A illustrates a comparison
of experimental data to the proposed model for segment 1. FIG. 28B illustrates an error
histogram for segment 1. FIG. 28C illustrates a comparison of experimental data to the proposed mo
del for segment 1. FIG. 28D illustrates an error histogram for segment 1. FIG. 28E illustrat
es a comparison of experimental data to the proposed model for segment 1. FIG. 28F illustrates a
n error histogram for segment 1. [00357] Similar to the evaporator model, the condenser can a
lso optionally have two sets of parameters since the pressure dynamics are c
oupled. One captures the first portion well while the other captures the end steady state. As t
he flow rate oscillations change the pressure MCC Ref. No.: 103361‐369WO1 dynamics change direction, however, the data doesn't
suggest any of these switches. On analyzing section‐wise it can be seen that segments
1 and 3 are in the acceptable range, while segment 2 , due to the oscillation shows a high de
viation from the data more than the 10% reference. [00358] FIG. 29A illustrates condenser pressure with parameter
set 1, and FIG. 29B illustrates condenser pressure with parameter set 2.
FIG. 29B shows that segment 2 shows similar behavior, even in terms of the magnitude of
the error. While segment 1 is captured well, segment 3 shows a high deviation from data and the
േ10% deviation. This is due to the spike that wasn't captured in the model at time t ൌ 70 min. [00359] The compressor regulates the flow rate and hence the
pressure difference to achieve the right cooling effect. In this application
, an electric compressor is used which converts the battery energy to mechanical work with
some electrical efficiency. The other losses associated with the compressor are the (1) is
entropic losses when the compressor displays some deviation from the isotropic line when
increasing the pressure from ^^ ^ to ^^ ^ , (2) volumetric efficiency where the amount of refrigerant
entering is not the same as the amount of refrigerant exiting the compressor, i.e., the idea
listic reduction in the compressor capacity, (3) electrical and mechanical losses, i.e when conver
ting electric work to mechanical work and losses due to friction. [00360] These losses are captured as quasi‐static elements
as a function of the pressure ratio and the RPM of the compressor using
steady‐state data. FIG. 31A, FIG. 31B, and FIG. 31C show the maps used for the mechanical and
electrical efficiency combined, volumetric MCC Ref. No.: 103361‐369WO1 efficiency and the isentropic efficiency. FIG. 31A il
lustrates mechanical and electrical efficiency, FIG. 31B illustrates volumetric efficiency, and FIG.
31C illustrates isotropic efficiency. [00361] The volumetric efficiency is used to calculate the m
ass flow rate given the volumetric displacement as in Eq. (31). [00362] ^^ ^˙௩ ௩ ൌ ^ ^ ˙^^ೞ^ (31) implementation studied. The isentropic efficiency is u
sed to find the enthalpy at the exit of the compressor or the entry of the condenser using the
relation in Eq. (32). [00364] ℎ ^భ,ೞି^ర ^ ൌ ℎ ସ ^ ఎisen ^^ratio ,ோ^ெ^ (32) shown in Eq. (33) and is observed to have a mean
of 1.08 and a standard deviation of 0.016 for the entire data, and hence is used as a constant f
or the simulation. [00366] ^^ ൌ ୪୭^ ^^ratio ^ ୪ ୭^ ^ఘout /ఘin ^ (33) ^షభ ^^ ൌ ^ ^^ ^ ^ ^ ^ ^^ ^ ^ െ 1^ cycle simulator capturing the coupling between the va
por compression cycle, i.e., HXs, compressor, and cabin, and is shown in Fig. 32. An
'Ambient' subsystem that describes the external conditions of the system like the solar GHI
, ambient temperature, and relative wind speed. A setpoint temperature is set at the driver
subsystem and the control module is a MCC Ref. No.: 103361‐369WO1 placeholder for any control strategy desired to be i
mplemented in the system. The current simulation is run with a cabin vent blower and comp
ressor and radiator fan RPM control. [00370] The objective of this simulator is to calculate the
compressor power to maintain a certain temperature at the cabin and also
to study any control development. As an example, a cabin modeling section of Khuntia et al.
(2022a) is emulated and the corresponding load is calculated. That is maintaining the cabin at
a setpoint temperature of 24.5 ∘ C. [00371] It was found that due to the size of the cabin, t
he temperature of the recirculating air going into the evaporator was not
the same as the average cabin temperature. However, it had similar dynamics to it. Hence a pse
udo node is introduced and added to the cabin model that represents the air entering the eva
porator. [00372] ^^ ௗ h vac ^ ௗ ௧ ൌ ^^ ௪^^,^ ^ ^^ ௪ െ ^^ ^ ^ ^ ^^ roof,c ^ ^^ roof െ ^^ ^ ^ ^ ^^ ^˙^ ^^ ^^ ^^^ ^ ^^ hvac െ ^^ cab ^ a rules‐based controller, the results and the contr
ol actions are shown in FIGS. 33A‐33C. The model is calibrated for AC operations. It can be se
en in FIG. 33B that the data for cabin internal air temperature is provided. In the experiment, until
time ^^ ൌ 4 h, the HVAC is operated as a heat pump. As the daytime temperature rises the cabi
n is maintained at a setpoint temperature and as the temperature falls, the cabin temperature
also falls. The load corresponding to the hoteling period is calculated and is shown in FIG.
33C and the corresponding evaporator and condenser pressure is shown in FIG. 33A. It is wort
h noting that as the temperature of the cabin needs to be maintained, the pressure of the refriger
ant has to fall to be able to lower its own temperature for effective heat transfer, while at the
condenser the temperature has to increase higher for effective heat transfer with the MCC Ref. No.: 103361‐369WO1 [00374] If the model is not calibrated as a heat pump, it
is not activated when it should be used for heating, hence it takes longer t
o rise to the setpoint temperature. [00375] No cross‐flowing air at the condenser end can help
in the rise of pressure at the condenser. Condenser fans are turned off at this
time. When they turned on, the pressure at the condenser decreased. The compressor RPM is co
mmanded using the difference in the ambient and the cabin inlet temperature, and it in
turn commands a flow rate for the refrigerant using the relationship in Eq. (31). FIG.
33C shows the corresponding electrical power. This is only corresponding to the compressor
work and not the vent blowers in the cabin. The total energy required at the compressor w
as found to be 15.21kWhr. [00376] Implementations of the present disclosure can include
a data driven Model. HVAC may not be the only source of load on the on
board battery pack during the hotel period. The driver engages in activities like having food/dri
nks, relaxing, and watching TV. The cabin in a long‐haul sleeper truck comes equipped with devic
es to support these activities like a microwave, refrigerators, etc. There is a load associ
ated with using these devices. The time and duration of using these activities (device usage) det
ermines the total energy consumption during the hotel period as shown in FIG. 4. While
HVAC load can be predicted using Physics‐ based HVAC modeling, the example implementation can c
apture the other activities of the driver using Machine Learning techniques. [00377] Any machine learning model requires a lot of data t
o be trained on. In this case, the data is the time and duration of a corre
sponding device being used during the 10‐hour hotel period. Currently, this data is not being acti
vely recorded on the truck, and in cases where it is recorded, it is not easily available to be s
hared. It is also not possible to have enough surve
y MCC Ref. No.: 103361‐369WO1 data to supplement the training of a Neural Network
(NN) which can require 100’s of data points to be accurate enough. Hence, the synthetic d
ata is generated in this work based on the existing survey samples. [00378] FIG. 4 illustrates different activities and the x‐a
xis shows the different time stamps. FIG. 4 represents the survey data recorded f
or the SuperTruck project. While some of these activities require the use of an electrical ap
pliance like a TV, the ones that don't are represented by 0W. FIG. 5 represents the normalized
electric power for the 10 hours of the hotel period for these activities. The majority of t
he time the load is low and there are some large spikes in the data. These spikes correspond to
the microwave and the coffee maker is turned on. The long during between the two sets of
spikes is the time in which the driver sleeps. The other loads can considered be small and
insignificant. [00379] According to implementations of the present disclosure
, predicting the duration of each activity independently is a multivar
iate problem and increases the complexity of the machine learning problem. By adding all the
corresponding loads at a particular time stamp, such as shown in FIG. 5 and predicting this
profile can reduce this multi‐variate problem to a uni‐variate problem eliminating the need for
creating categories/labels for each activity/group of activities. The long and Short Term
Memory (LSTM) algorithm can be used in implementations of the present disclosure for predicti
ng this time series data and is set up as a regression problem. The input and output are set up
as a predictor response format where a series is fed as an input and the output is a val
ue representing the power load at the next time instant. For the first time step, the problem is se
t as one‐to‐one and then changes to a many‐to
one prediction from time to step two as the predict
ion is stacked to the input. MCC Ref. No.: 103361‐369WO1 [00380] The first survey is used to generate enough data se
ts to train a machine learning algorithm. A Time Allocation Matrix (TAM) an
d a Transition Matrix ^TM^ are extracted from it. A combination of the two along with a pow
er load vector is fed as the input to the NN. The algorithm gives some predictions which are then
evaluated and the performance is reported. A flowchart of the algorithm configuration
for the example implementation is illustrated as FIG. 7. [00381] The temporal information is captured using the probab
ility distribution for each activity and can be seen in FIG. 8 for the d
river sleeping, using the microwave, and using the coffee maker activities. The TM is developed usi
ng Markov property Gagniuc (2017). For ^^‐ activities a ^^‐by‐n matrix is generated, in which the element
^^‐by‐m ^ ^^, ^^ ∈ ^^^ would store the probability of going from activity ^^ to activity ^^ at any instant. [00382] The non‐limiting example algorithm in the study was
set up as: [00383] (1) Split the data into training and test sets in
a 9: 1 ratio [00384] (2) Create predictor ^^ ழ^:^ షభவ and response ^^ ழଶ:^ வ for the training sequences where ^^ ௫ is the length of the series. [00385] (3) Row wise append the predictor with TAM and TM,
i.e., create the input matrix. and train the network. [00386] (4) Use the trained network to predict the test day
. An initial guess of ^^ ழ^வ is 0 and a prediction ^^ ழ^வ is made. This prediction is columnwise appende
d to ^^ ழ^வ as ^^ ழଶவ and a prediction for 3rd time step, i.e., ^^ ழଶவ is made until ^^ ழ ^வ [00387] the power load. MATLAB NN toolbox was used to defin
e the architecture of the LSTM algorithm MCC Ref. No.: 103361‐369WO1 and is summarized below. The algorithm is trained us
ing an 'adam' solver with an initial learning rate of 0.005 and a gradient threshold of 1 for 25
0 epochs. [00388] (1) Input Layer: With 25 features [00389] (2) LSTM Layer: With 50 hidden units [00390] (3) LSTM Layer: With 50 hidden units [00391] (4) Fully Connected Layer: Multiplies the input weigh
t matrix and adds the bias vector. [00392] (5) Regression Layer: Compares the mean squared error
. [00393] The right compromise between the computation time and
accuracy is suggested using 2 layers. In order to capture the m
inor trends in the data a two‐layer deep LSTM is used. Adding another LSTM layer increases th
e computation time exponentially high. [00394] The study included model validation. Characterization
of the accuracy of the prediction can be used to define the performance of
the algorithm and also the hyperparameter tuning. Euclidean distance with Dynamic
Time Warping (DTW) is hence used. While having the total energy consumption as a metri
c can be useful, it would give an error averaged over the complete prediction horizon. Instant
aneous predicted power needs to be checked and this is where DTW becomes useful. Using
the RMSE error after warping the time axis [00395] The two hyper‐parameters define the size of the ne
twork as well as determine the computation time. The number of hidden units cho
sen is [20 50 100 150] while the training period is [10 204070100140 200]. A visualization of
the errors for the different combinations is MCC Ref. No.: 103361‐369WO1 given in FIG. 9 as the surface plot of the error
matrix, this is the Euclidean distance measured point‐wise on the warped time axis. [00396] Since the optimization did not follow a pattern, a
grid exhaustive search determines the minimum error at 50 hidden units and
a training period of 20 days. The results with this combination are reported in FIGS. 10A and
10B. [00397] FIG. 10A shows the forecast and the prediction as i
s while FIG. 10B shows its warped version. Different overlapping regions show the
accuracy of the model. However it must be noted that warping may not be used for the
final prediction, but only to gauge the accuracy of the model. It is irrelevant to capture
the exact times of the activities and more important to capture the overall trend. For instanc
e, it is more important to have the larger spikes corresponding to the usage of microwave and c
offee maker to be at the start and the end of the hoteling leaving the time in the middle
for the sleeping activity. [00398] The two‐layer LSTM algorithm trained on 20 days of
data with 50 hidden units achieved 90% accuracy in total energy prediction calc
ulated as the integrated sum of power over time. A 32 GB RAM, 2.2GHz clock rate, and 64
bit processor computer are expected to take 4 minutes for the computation. [00399] The study included an evaluation of an example drive
cycle corresponding to a full day of a class 8 vehicle on a long‐haul jou
rney is developed adhering to federal laws. The example vehicle is a mild hybrid vehicle where the
electric motor only works as a generator to charge the onboard battery pack during the drive pha
se. FIG. 34 illustrates the drive cycle and includes drive and hotel phases. The drive cycle of
FIG. 34 is generated by repeating a smaller MCC Ref. No.: 103361‐369WO1 drive cycle defined for the SuperTruckII. The driver
is on‐duty for 11 hours with two 30‐minute rests and hotels for 10 hours. [00400] The models presented in this paper estimate the tota
l energy required during the 10‐hour hotel period. This is a combination of
the HVAC load (FIG. 33C) calculated using the physics‐based models and the load from the activiti
es of the driver is predicted using the LSTM algorithm (FIGS. 10A and 10B). This is also illustra
ted in FIG. 34. [00401] After determining the energy consumption, a correspond
ing initial (beginning of hotel period) battery's SOC is determined. An opt
imal SOC trajectory is calculated using an optimal control‐dynamic programming algorithm that fi
nds the optimal instances to recharge the battery, This can be done by running the engine
at a more efficient operating point while transferring additional energy to the battery or rege
neration from the wheels Singh et al. (2022). Dynamic programming, being a non‐causal opti
mization method, requires all the information about the route and power loads from the
auxiliary devices ahead of time. [00402] The SOC is displayed in FIG. 34. These two lines c
orrespond to two different battery packs (battery pack‐1 and battery pack‐2),
the smaller‐dotted blue, and the larger pack ‐
solid blue. In this case, the energy required to su
pport hotel loads is higher than the battery capacity hence the battery needs to be charged fully
before the hotel period starts. The battery starts being used as the driver is resting inside t
he truck and using the auxiliaries. Correspondingly the SOC starts to drop until it reac
hes the minimum allowable SOC. At this point, the engine has to be idle to charge the onb
oard battery pack only to the amount as is required for the rest of the hotel period avoiding
the 10 hour hoteling which is prevalent in MCC Ref. No.: 103361‐369WO1 trucks with conventional powertrains. The smaller batt
ery requires 1.09 hours of additional idling to support this operation while the larger pa
ck required only 27.8 minutes of idling. [00403] FIG. 35 illustrates an example system according to i
mplementations of the present disclosure. HVAC load cycle information can b
e predicted, and user activity can be predicted. The HVAC load cycle information and user
activity predictions can be used to determine a state of charge (“SOC”). Optionally,
the models shown in FIG. 35 can be refined based on the actual user activities and/or actual HV
AC load. [00404] Implementations of the present disclosure include syst
ems and methods to estimate the energy in hybrid class 8 long haul tru
cks during hoteling and/or use that information to increase the freight efficiency of the
vehicle achieved using an optimal control algorithm called dynamic programming. A full day of
simulation is used herein to show the results. The estimation of energy during hoteling is
estimated using physics‐based modeling and machine learning. Thermal models for the components i
n a vapor compressor cycle are developed from the first principles. The HX is model
ed using the moving boundary method and the compressor is modeled using empirical relationship
s. A 2‐node and 3‐node cabin model is explored to estimate the temperature of the cabin at
all times. Loads other than the HVAC loads, concerning the use of different devices are p
redicted using time series forecasting using a Recurrent Neural Network (RNN) called LSTM. An exampl
e 2 layer LSTM model uses 20 days of data to predict one full 10 hour hoteling. A 20‐d
ay moving window can be used to ensure consistent low‐energy training and accuracy. Both mo
dels can work separately to produce a 10 hour hoteling load that is ultimately combined and u
sed together in the optimal control algorithm. MCC Ref. No.: 103361‐369WO1 [00405] References [00406] Although the subject matter has been described in la
nguage specific to structural features and/or methodological acts, it is
to be understood that the subject matter defined in the appended claims is not necessarily li
mited to the specific features or acts described above. Rather, the specific features and ac
ts described above are disclosed as example forms of implementing the claims. [00407] Administration, F.M.C.S. (2011). Interstate truck drive
r's guide to hours of service. [00408] Aztiria, A., Augusto, J.C., Basagoiti, R., Izaguirre,
A., and Cook, D.J. (2012). Discovering frequent userenvironment interactions in in
telligent environments. Personal and Ubiquitous Computing, 16(1), 91‐103. [00409] Gagniuc, P.A. (2017). Markov chains: from theory to
implementation and experimentation. John Wiley & Sons. [00410] Goutham, M., Stockar, S., Blaser, R., and Hanumalagut
ti, P. (2021). User activity sequence prediction in smart homes using mul
ti‐layer long short‐term memory networks. IFAC‐PapersOnLine, 54(20), 901‐906. [00411] Goutham, M. (2020). Machine learning based user activ
ity prediction for smart homes. Master's thesis, The Ohio State Universi
ty. [00412] Graves, A. (2013). Generating sequences with recurrent
neural networks. arXiv preprint arXiv:1308.0850. MCC Ref. No.: 103361‐369WO1 [00413] Khuntia, S., Hanif, A., Singh, S.P., and Ahmed, Q.
(2022). Control oriented model of cabin‐hvac system in a longhaul class 8
trucks for energy management applications. Technical report, SAE Technical Paper. [00414] Marufuzzaman, M., Reaz, M., Ali, M.A.M., and Rahman,
L. (2015). A time series based sequence prediction algorithm to detect
activities of daily living in smart home. Methods of information in medicine, 54(03), 262270. [00415] Minor, B., Doppa, J.R., and Cook, D.J. (2015). Data
driven activity prediction: Algorithms, evaluation methodology, and applications. I
n Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Da
ta Mining, 805‐814. [00416] Mitler, M.M., Miller, J.C., Lipsitz, J.J., Walsh, J.K
., and Wylie, C.D. (1997). The sleep of long‐haul truck drivers. New England Journ
al of Medicine, 337(11), 755‐762. [00417] Müller, M. (2007). Dynamic time warping. Information
retrieval for music and motion, 69‐84. [00418] Smith, D., Ozpineci, B., Graves, R.L., Jones, P., Lu
stbader, J., Kelly, K., Walkowicz, K., Birky, A., Payne, G., Sigler, C., et
al. (2020). Medium‐and heavy‐duty vehicle electrification: An assessment of technology and knowl
edge gaps. Technical report, Oak Ridge National Lab.(ORNL), Oak Ridge, TN (United States). [00419] M. Margreta, C. Ford, R. Grube, et al., US freight
on the move: highlights from the 2012 Commodity Flow Survey Preliminary Data, Tech
nical Report, United States. Department of Transportation. Bureau of Transportation
Statistics, 2014. [00420] S. C. Davis, R. G. Boundy, Transportation energy dat
a book: Edition 39, Technical Report, Oak Ridge National Lab.(ORNL), Oak
Ridge, TN (United States), 2021. MCC Ref. No.: 103361‐369WO1 [00421] O. Delgado, N. Lutsey, The us supertruck program, Wa
shington DC (2014). [00422] URL: https://www.energy.gov/eere/vehicles/articles. [00423] Interstate truck driver's guide to hours of service
(fmcsa) (2022). URL: https://www.fmcsa.dot.gov/regulations/hours‐service/. [00424] ANL, Idling reduction for long‐haul trucks: An econ
omic comparison of on‐ board and wayside technologies (2012). URL: https://pu
blications. anl.gov/anlpubs/2016/10/130502.pdf. [00425] L. Gains, Economics of idling reduction options for
long‐haul trucks (2017). URL: https://www. anl.gov/sites/wwW/files/2018‐02. C.
J. Brodrick, N. P. Lutsey, Q. A. Keen, D. I. Rubins, J. P. Wallace, H. A. Dwyer, S. W. Gouse
, Truck idling trends: Results of a pilot survey in
northern california, in: SAE Technical Paper, SAE Int
ernational, 2001. URL: https : //doi.org/10.4271/2001‐01‐2828. doi:10.4271/2001‐01
2828. [00426] F. Stodolsky, L. Gaines, A. Vyas, Analysis of techno
logy options to reduce the fuel consumption of idling trucks, Technical Report,
Argonne National Laboratory, 2000. [00427] C. Kshirsagar, System Level Modeling and Optimization
of Fuel Cell Powered Auxiliary Power Unit (APU) to be used in Commercial
Vehicles, Technical Report, SAE Technical Paper, 2015. [00428] B. Surampudi, J. Redfield, G. Ray, A. Montemayor, M.
Walls, H. McKee, T. Edwards, M. Lasecki, Electrification and integration o
f accessories on a class‐8 tractor, in: SAE Technical Paper, SAE International, 2005. URL: https:
//doi .org/10.4271/2005‐01‐0016. doi:10.4271/2005‐01‐0016. MCC Ref. No.: 103361‐369WO1 [00429] B. Surampudi, M. Walls, J. Redfield, A. Montemayor,
C. Ingold, J. Abela, 42‐ volt electric air conditioning system commissioning an
d control for a class‐8 tractor, in: SAE Technical Paper, SAE International, 2004. URL: https
: //doi .org/10.4271/2004‐01‐1478. doi:10.4271/2004‐01‐1478. [00430] J. Redfield, B. Surampudi, R. Gustavo, A. Montemayor,
H. McKee, T. Edwards, M. Lasecki, Accessory electrification in class 8 trac
tors, in: SAE Technical Paper, SAE International, 2006. URL: https://doi .org/10.4271/2006
01‐0215. doi:10.4271/2006‐01‐0215. [00431] B. Surampudi, J. Redfield, A. Montemayor, G. Ray, G.
Ostrowski, H. McKee, T. Edwards, A. S. Carstensen, J. C. Lawrence, Electric
air conditioning for class 8 tractors, in: SAE Technical Paper, SAE International, 2006. URL: https:/
/doi.org/10.4271/2006‐01‐0165. doi:10.4271/2006‐01‐0165. [00432] J. Campbell, W. Watts, D. Kittelson, Reduction of ac
cessory overdrive and parasitic loading on a parallel electric hybrid city
bus, in: SAE Technical Paper, SAE International, 2012. URL: https : //doi .org/10.4271/2012‐01‐1005.
doi:10.4271/2012‐01‐1005. [00433] S. P. Singh, A. Hanif, Q. Ahmed, M. Meijer, J. Lah
ti, Applied Energy 326 ^2022^119982. S. Khuntia, A. Hanif, S. P. Singh, Q.
Ahmed, Control Oriented Model of Cabin‐ HVAC System in a Long‐Haul Class 8 Trucks for Ene
rgy Management Applications, Technical Report, SAE Technical Paper, 2022a. [00434] S. Khuntia, A. Hanif, Q. Ahmed, J. Lahti, M. Meijer
, Cabin load prediction using time series forecasting in long‐haul trucks for opt
imal energy management, IFAC‐PapersOnLine 55 (2022b) 342‐347. MCC Ref. No.: 103361‐369WO1 [00435] S. Khuntia, A. Hanif, S. P. Singh, Q. Ahmed, Contro
l Oriented Model of Cabin‐ HVAC System in a Long‐Haul Trucks for Energy Manag
ement Applications, Technical Report, SAE Technical Paper, 2022c. [00436] C. Okaeme, J. Lustbader, C. Sigler, I. Jorgensen, B.
Grover, J. Kiesser, M. Moniot, Cabin thermal management analysis for supertru
ck ii nextgeneration hybrid electric truck design, SAE International Journal of Commercial
Vehicles 14^2021^. [00437] M. Altwieb, K. J. Kubiak, A. M. Aliyu, R. Mishra,
A new three‐dimensional cfd model for efficiency optimisation of fluid‐to‐air
multi‐fin heat exchanger, Thermal Science and Engineering Progress 19 (2020) 100658. [00438] Y. Fukuchi, K. Yoshitake, K. Yokota, Development of
simplified condenser model considering the effect of phase change, in: Pr
oceedings of 12thFKFS Conference‐ Progress in Vehicle Aerodynamics and Thermal managemen
t, 2019. [00439] Y. Ge, R. Cropper, Performance evaluations of air‐c
ooled condensers using pure and mixture refrigerants by four‐section lumped
modelling methods, Applied thermal engineering 25 (2005) 1549‐1564. [00440] M. Steinstraeter, J. Buberger, K. Minnerup, D. Trifon
ov, P. Horner, B. Weiss, M. Lienkamp, Controlling cabin heating to improve ran
ge and battery lifetime of electric vehicles, ETransportation 13 (2022) 100181. [00441] F. Widmer, A. Ritter, P. Duhr, C. H. Onder, Battery
lifetime extension through optimal design and control of traction and heating s
ystems in hybrid drivetrains, ETransportation 14 (2022) 100196. MCC Ref. No.: 103361‐369WO1 [00442] X.‐D. He, Dynamic modeling and multivariable control
of vapor compression cycles in air conditioning systems (2005). Q. Zhang,
S. Stockar, M. Canova, Energy‐optimal control of an automotive air conditioning system for
ancillary load reduction, IEEE Transactions on Control Systems Technology 24 (2015) 67‐80. [00443] J. A. Lustbader, J. P. Rugh, B. R. Rister, T. S.
Venson, CoolCalc: a long‐haul truck thermal load estimation tool, Technical Report, Nation
al Renewable Energy Lab.(NREL), Golden, CO (United States), 2011. [00444] C. C. Richter, Proposal of new object‐oriented equa
tion‐based model libraries for thermodynamic systems, Ph.D. thesis, Braunschweig,
Techn. Univ., Diss., 2008, 2008. [00445] H. Tummescheit, J. Eborn, K. Pr, Airconditioning ‐
a modelica library for dynamic simulation of ac systems, 2005. [00446] H. Tummescheit, Design and implementation of object‐
oriented model libraries using modelica, Ph.D. thesis, Lund Universit
y, 2002. [00447] M. Browne, P. Bansal, An insight into the applicatio
n of the ntu‐? approach for modelling vapour‐compression liquid chillers (1998).
[00448] M. O. McLinden, Nist thermodynamic and transport prop
erties of refrigerants and refrigerant mixtures‐refprop, NIST Standard Refer
ence Database 23^1998^ [00449] T. Wang, A. Jagarwal, J. R. Wagner, G. Fadel, Optim
ization of an automotive radiator fan array operation to reduce power consumpt
ion, IEEE/ASME Transactions on Mechatronics 20 (2014) 2359‐2369. MCC Ref. No.: 103361‐369WO1 [00450] Kenworth, Peterbilt, Kenworth: Peterbilt radiator: 2008
‐2011 t680, t880, t2000: 2008‐2011nbsp;365, 367, 387: 2011‐2012nbsp;58
7, American Radiator Pros (2012). URL: https://radiatorpros.com. [00451] T. Wang, A. Jagarwal, J. R. Wagner, G. Fadel, Optim
ization of an automotive radiator fan array operation to reduce power consumpt
ion, IEEE/ASME Transactions on Mechatronics 20 (2014) 2359‐2369. [00452] X.‐D. He, Dynamic modeling and multivariable control
of vapor compression cycles in air conditioning systems, Ph.D. thesis, Mas
sachusetts Institute of Technology, 1996. S. P. Singh, Optimal Hotel load management in 48V Class
‐8 Mild Hybrid Trucks, Master's thesis, The Ohio State University, 2021. [00453] I. H. Bell, J. Wronski, S. Quoilin, V. Lemort, Pure
and pseudo‐pure fluid thermophysical property evaluation and the open‐sourc
e thermophysical property library coolprop, Industrial & engineering chemistry resear
ch 53 (2014) 2498‐2508. [00454] Q. Zhang, M. Canova, Lumped‐parameter modeling of a
n automotive air conditioning system for energy optimization and manage
ment, in: Dynamic Systems and Control Conference, volume 56123, American Society of
Mechanical Engineers, 2013, p. V001T04A003. [00455] A. V. Dandekar, C. S. Brooks, Modeling of void frac
tion covariance in twophase flows with phase change, International Journal of Hea
t and Mass Transfer 100 (2016) 231‐242. [00456] P. A. Gagniuc, Markov chains: from theory to impleme
ntation and experimentation, John Wiley & Sons, 2017.
Previous Patent: MEDICAL DEVICE COMPONENTS, ASSEMBLIES, AND ASSOCIATED METHODS
Next Patent: CHEMICAL MECHANICAL POLISHING PADS WITH A DISULFIDE BRIDGE
Next Patent: CHEMICAL MECHANICAL POLISHING PADS WITH A DISULFIDE BRIDGE