FIELD OF THE INVENTION

The invention relates to a control method and control mechanism for electrical machines supported by magnetic bearings. In connection with the present invention, the electrical machine may be machine configured to convert electrical energy to mechanic power, such as an electrical motor, or it may be a machine configured to convert mechanical energy to electrical power, such as a generator.

BACKGROUND OF THE INVENTION

The theory and operating principles of magnetic levitation are well understood by those skilled in the art. The following description of the theory and operating principles relate to an electrical motor but they are easily adaptable to a generator.

An electrical machine that comprises a stator and a rotor generates a magnetic torque between the stator and rotor. The same magnetic flux that generates the torque can be adapted to generate an additional magnetic levitation force that acts perpendicular to the axle of the rotor.

In an exemplary case, wherein a cylindrical rotor is configured to rotate within stator having the general shape of a cylindrical ring, the magnetic flux traverses a cylinder-shaped air gap between the stator and rotor. The flux density in the air gap may be divided into Fourier components, wherein the base wave is the most significant component. The wavelength of the base wave is twice the pole pitch of the machine, and its wave number (pole pair number) is designated by p. By utilizing complex notation, the radial compo- nent of the air gap flux density caused by the base wave flux can be expressed as follows:

Herein B _p is the magnitude of the base wave, φ is the angular coordinate of a cylindrical coordinate system, ω is the frequency of the power feed, t is time, and a is the phase angle of the base wave. In the following, the phase angle a will be assumed zero, to simplify the description.

The magnetic force acting on the rotor of an electrical machine can be calculated by utilizing Maxwell tension tensor. When a radial flux machine is examined by means of a two-dimensional model, the magnetic force between the stator and rotor is yielded by:

^{F =} Τ _μ - s - ^{+ 2Β} τΒ _φ ^β _φ ]ά5 [2]

Herein, μ ₀ is the permeability of the vacuum, B _r and Β _ψ are the radi- al and tangential component, respectively, of the flux density in the air gap of the machine, and S is cylindrical surface coaxial with the rotor in the air gap.

Since the radial component of the flux density is normally significantl larger than the tangential component, equation 2 can be simplified as:

In studies of rotor dynamics, forces acting on the rotor are commonly presented as complex numbers, wherein the magnitude and angle of the complex number correspond to the strength and direction of the force, respectively. By resorting to complex presentation and by substituting the expression for the cylindrical surface element, we obtain the complex expression F (or F ) of the force as follows:

Herein / denotes the axial length of the air gap of the machine, while R is the radius of the air gap.

The base wave by itself cannot produce the combined force acting on the rotor. This can be verified by substituting equation 1 of the base wave of the flux density into equation 4 of the force. Generation of a magnetic force is only possible by providing the air gap with a second harmonic component whose wave number v differs from the wave number of the base wave, denoted by p. In this case the flux density of the air gap is given by:

B = B _V + B _V = Β _ν ε^- ^ω ^ + B _v e^- ^Wt+C ^ [5]

Substitution of equation 5 into equation 4 and a small calculation show that generation of the force is only possible when the two wave numbers differ from each other b one. For the case wherein v = p+"\ :

For the case wherein v = ρ+λ : nlR - Ιί-α

F

2μ ₀

The harmonic wave, wherein v = p±\ , can be produced by providing the stator with an additional three-phase winding, for example, and the additional winding may be supplied by a frequency converter. The feed voltage and the rotational velocity determine the magnitude and frequency of the base wave. The magnitude and rotational velocity may be adjusted by varying the magnitude and frequency of the harmonic wave having a wave number of v = p±\ .

For a better overall view of the equipment and power needed for generating the force, an exemplary calculation of the generation of a static force will be presented next. The magnitude of the static force shall be set equal to the weight of the rotor. In other words, the static force is sufficient to support the rotor against the force of gravity. The magnitude of the harmonic wave (v = p±\ ) to be generated in the air gap is given by: F ^p±l [8]

The frequency at which the harmonic wave generates the static force is given by:

Ω _ρρ±1 = 0 => ± (w _p±l - co) = 0 => w _p±l = ω => f _p±l = f _p [9]

The slip of the harmonic wave in relation to the rotor is an important quantity, as it indicates the power losses caused by the generation of the static force in the rotor cage of an asynchronous machine or in the damping winding of an synchronous machine. The slip can be calculated from: p p + l p + l p

Magnetic levitation of electrical motors is a field of active study. For instance, US patent application 2003/0057784 by Hideki Kanebako discloses a magnetically levitated motor and magnetic bearing apparatus. But a few residual problems remain.

One of the residual problems relates to the fact that the magnetic levitation force depends on the magnetic flux density and the cross-sectional area traversed by the magnetic fluxes of the magnetic bearing. Because flux density is restricted by physical constraints, a strong magnetic levitation force requires physically large magnetic bearings. This is one of the problems remaining in the field.

Another problem is that many publications of the prior art only address the problem of levitating the rotor of an electrical motor. What this means is that the prior publications only address the question of how to fully or partially compensate for the weight of the rotor. A residual problem is that questions relating to the rotor's orientation or vibrations remain unanswered.

SUMMARY OF THE INVENTION

It is an object of the invention to eliminate or alleviate one or more of the problems identified in the background section. This object is achieved by means of a method and apparatus as claimed in the attached independent claims. The dependent claims as well as the following detailed description and drawings describe specific embodiments which solve additional problems and/or provide additional benefits.

A first aspect of the invention is a method for controlling operation of an electrical machine; wherein the electrical machine comprises a stator and a rotor; wherein the rotor is at least partially supported by at least two magnetic bearings, each magnetic bearing having a set of windings for providing a support force on a respective portion of the rotor, and a set of position sensors for indicating a position of the respective portion of the rotor in said at least two degrees of freedom.

The fact that the position sensors, such as air gap sensors, indicate the position of the respective portion of the rotor in at least two degrees of freedom. Should be understood as follows. Firstly, each of at least two mag- netic bearings provide a support force on a respective portion of the rotor. For example, each of two magnetic bearings placed near the ends of the rotor axle supports a respective end of the rotor axle. A set, such as a pair, of position sensors, installed for each magnetic bearing, indicates the position of the respective portion (such as end) of the rotor in two dimensions that are prefera- bly orthogonal, such as x-y or angle-radius.

The stator comprises a first set of windings, with a first pole pair number, for conversion between electrical energy and a rotating magnetic flux, wherein the rotating magnetic flux either drives the rotor (as in a motor) or is driven by it (as in a generator). The stator further comprises a second set of windings, with a second pole pair number, wherein the second pole pair num- ber differs from the first pole pair number by +1 or -1 . It is the purpose of the second set of windings to generate the magnetic levitation force that fully or partially bears the weight of the rotor and thereby reduces the load of other load-bearing systems, such as the magnetic bearings. Because the pole pair number of the second set of windings differs from the pole pair number of the first set of windings by +1 or -1 , the second set of windings can be used to generate a static magnetic levitation force.

The inventive method further comprises performing the following steps by an automated controller: a) controlling operation of a first variable power source, to adjust said conversion between the electrical energy and the rotating magnetic flux via the first set of windings; b) maintaining a range of nominal positions for each of the portions of the rotor in said at least two degrees of freedom; c) receiving from the sets of position sensors a position indication for each of the one or more portions of the rotor; d) monitoring a devia- tion between the nominal positions and the respective position indications, and e) based on the monitored deviation, controlling operation of a second variable power source, to supply compensative current to the second set of windings.

Step a, namely controlling operation of a first variable power source, to adjust said conversion between the electrical energy and the rotating mag- netic flux via the first set of windings, relates to the primary function of an electrical motor or generator, which is conversion from electrical energy to mechanical energy or vice versa. Step b, maintaining a range of nominal positions for each of the portions of the rotor in said at least two degrees of freedom, means that for each portion of the rotor equipped with a magnetic bearing and a set of position sensors, there is a range of nominal positions in the coordinate systems that comprises at least two, preferably orthogonal, coordinates. The range of nominal positions can be infinitesimally thin, such as a single ideal value for each of the (preferably) orthogonal coordinates. Or the range of nominal positions can extend over a range of permissible values such that no action needs to be taken if the position indications remain within the ranges of permissible values. Step c, receiving from the sets of position sensors a position indication for each of the one or more portions of the rotor, may mean that the automated controller receives data numbers that directly indicate the position of the portions of the rotor. Or it may mean that the automated controller receives data numbers from which the position of the portions of the rotor can be calculated. Step d, namely monitoring a deviation between the nominal po- sitions and the respective position indications, means that the automated controller monitors if the indicated positions of the portions of the rotor are at the nominal positions or within the position ranges, or if the indicates positions deviate from the nominal values.

In step e, the automated controller controls operation of a second variable power source on the basis of the monitored deviation, so as to supply compensative current to the second set of windings. The compensative current causes the second set of windings to generate a magnetic levitation force that supports the rotor and thereby causes a reduction in said deviation between the nominal positions and the respective position indications.

Other aspects of the invention include an automated controller for carrying out the method of the first aspect, as well as an electrical machine, such as a motor or a generator, that comprises the automated controller. There is a wide range of implementation techniques for constructing the automated controller according to the invention. For instance, an embedded programmed data processor can be used for this purpose. Alternatively, Field- Programmable Gate Arrays (FPGA) or discrete analogue, digital or hybrid (analogue + digital) implementations are possible. Finally, the controller can be implemented by utilizing operational amplifiers.

In the context of the present invention, "magnetic bearing" refers to components or subsections of the electrical machine that do not contribute to the conversion between electrical energy and mechanical rotational energy, except indirectly, by supporting the weight of the rotor and thereby reducing frictional losses. The magnetic bearings need not be firmly integrated with the electrical machine, and it is common practice to obtain the electrical machine and the magnetic bearings from separate suppliers. The principal function of magnetic bearings, as the term is used in the present context, is to support the weight of the rotor by utilizing a magnetic flux. Thus, for the purposes of the present invention, the magnetic levitation force generated by the second set of windings of the stator does not qualify as a magnetic bearing. Another way to distinguish magnetic bearings from a magnetically levitated rotor is to say that the electrical machine comprises at least two sets of magnetic bearings which are separated from each other, as measured along the rotor's longitudinal axis, and that the rotor is placed between these two sets of magnetic bearings.

The very fact that the electrical machine and the magnetic bearings are frequently obtained from separate suppliers has caused a residual problem in the prior art, namely the fact that control of the magnetic bearings on one hand and control of the magnetic levitation of the rotor on the other hand have not been integrated into a common controller. This residual problem has the consequence that it has not been possible to monitor the performance of the magnetic bearings and adjust the rotor's magnetic levitation, via the second set of windings, depending on the monitored performance of the magnetic bearings. Monitoring the performance of the magnetic bearings may include monitoring the position of the ends of the rotor, ie, monitoring the deviation of the actual positions from their nominal values; monitoring vibration of the rotor by means of additional sets of position sensors, or a combination of these.

As regards the operating principle of the position sensors, they can be optical, ultrasonic, Hall effect, capacitive, inductive or eddy current sensors or any combination of these techniques.

Brief Description of the Drawings

Specific embodiments of the present invention will be described in the following, with reference to the attached drawings in which:

Figure 1 is a block diagram depicting the major functional blocks of an electrical machine equipped with a controller according to an embodiment of the present invention;

Figures 2A to 2C show how vibration effects of the rotor may be detected by utilizing additional position sensors; and

Figure 3 is a block diagram of an automated controller implemented as a programmed data processor.

Detailed Description of Specific Embodiments

In the attached drawings like reference numerals refer to like items and a repeated description is usually omitted.

Figure 1 is a block diagram depicting the major functional blocks of an electrical machine equipped with a controller according to an embodiment of the present invention. The invention will be described in connection with a horizontally installed electrical machine 100 that comprises a stator 1 10 and a rotor 120. In the drawing, the front half of the stator 1 10 has been removed such that the rotor 120 is visible. The rotor 120 is at least partially supported by at least two magnetic bearings 130A, 130B. Each magnetic bearing has a set of windings for providing a support force on a respective portion 120A, 120B of the rotor 120. Magnetic bearings are well known in the art, and a detailed de- scription is omitted. The magnetic bearings 130A, 130B are provided with a set of position sensors 132A, 133A; 132B, 133B for indicating a position of the respective portion 120A, 120B of the rotor in at least two degrees of freedom. The degrees of freedom refer to a generally orthogonal coordinate system, such as an x-y system or radius/angle system.

The stator 1 10 comprises a first set of windings 1 12, with a first pole pair number p, for conversion between electrical energy and a rotating magnetic flux, wherein the rotating magnetic flux either drives the rotor 120 or is driven by it. What this means is that the electrical machine 100 may act as a motor, in which case the rotor 120 is driven by the rotating magnetic flux generated by the stator. Or, the electrical machine 100 may act as a generator, in which case the rotor, which is driven by an external mechanical rotational force, generates a rotating magnetic flux, which is converted to electrical energy in the first set of windings 1 12 of the stator. Either way, it is the first set of windings 1 12, in combination with the rotor, that performs the primary function of the electrical machine, which is conversion between electrical energy and mechanical energy.

In order to support the rotor 120, the stator 1 10 further comprises a second set of windings 1 14. The second set of windings 1 14 has a second pole pair number v, which differs from the first pole pair number by +1 or -1 . In other words, p = v ± 1 . As regards mechanical construction, the second set of windings 1 14 can be installed in the grooves of the stator 1 10 or outside the stator.

The electrical machine 100 also comprises an automated controller 160, which may be implemented as an embedded data processing apparatus, which comprises one or more processors or processing cores, interfaces for input/output, management, debugging, data logging (optional). The data processing apparatus 160 may further comprise clock and interrupt handling circuitry, as is known to those skilled in the art. Still further, the apparatus 160 comprises memory for storing program code and variables. The program code contains instructions, the execution of which causes the apparatus 160 to carry out the inventive control method. The variables include actual and nominal values for process parameters, such as the positions of the portions of the rotor 120, as indicated by the position sensors.

The inventive method comprises performing the following steps or functions by the automated controller 160. For enabling the electrical machine 100 to perform its principal task, namely conversion between electrical and mechanical energy, the automated controller 160 controls operation of a first variable power source 152, such as a frequency converter, such that the variable power source 152 supplies an adequate level of power to the first set of windings 1 12. In addition, the automated controller 160 controls operation of a second variable power source 154, to supply compensative current to the second set of windings 1 14. It is the compensative current that causes the second set of windings 1 14 to generate a magnetic levitation force that supports the rotor 120 and thereby reduces the load that the weight of the rotor 120 impos- es on the magnetic bearings 130A, 130B. In order to supply the second set of windings 1 14 with an adequate level of energy, the automated controller maintains a range of nominal positions for two or more portions or the rotor 120, such as its ends, denoted by reference numerals 120A, 120B. In Figure 1 , the nominal positions are generally denoted by reference numeral 161 , and as an illustrative example, they are depicted as acceptable coordinate ranges {x1 ₀ , y1 ₀ }, {x2 ₀ , v2 ₀ } in a Cartesian coordinate system. The automated controller monitors actual positions of the portions 120A, 120B of the rotor 120, by receiving from the position sensors 132A, 133A; 132B, 133B a position indication for the portions 120A, 120B of the rotor 120. In Figure 1 , the actual position indications are generally denoted by reference numeral 162, and they are depicted as coordinates {x1 , y1 }, {x2, y2}. Alternatively, the permissible and actual coordinates can be processed in a polar coordinate system. The automated controller 160 monitors the deviation between the nominal positions 161 and the respective actual position indications 162. Based on the monitored devia- tion, the automated controller 160 instructs the second adjustable power source 154 to adjust the power supplied to the second set of windings 1 14. It is the adjustment of the power supplied to the second set of windings 1 14 that causes a reduction in the deviation between the nominal positions and the respective actual position indications.

Figures 2A to 2C show how vibration effects of the rotor may be detected by utilizing more than two pairs of position sensors. Figure 2A is a schematic side view of the elements shown in Figure 1 , apart from the fact that the stator 1 10 has been omitted to show the rotor 120 in its entirety. In this side view, only one position sensor from each pair of position sensors 132A, 133A and 132B, 133B is shown. Figure 2B shows a scenario in which the rotor 120 undergoes vibration such that the first pair of position sensors 132A, 133A at the left end of the rotor 120 and the second pair of position sensors 132B, 133B at the right end of the rotor 120 indicate no abnormal positions. Yet the rotor 120 vibrates such that the center point is clearly offset from its nominal position. Such vibrations can be detected by installing additional position sensors. Figure 2B shows an implementation wherein an additional pair of position sensors 232C, 233C is installed at the rotor's longitudinal center point. Again, only one sensor of the pair is shown in this side view.

A residual problem in this implementation is that even higher vibration modes can cause vibration nodes to occur at the three pairs of position sensors 132A, 133A; 132B, 133B; and 232C, 233C. This residual problem can be solved by installing an even higher number of position sensors. Alternatively, the residual problem can be solved by installing the third pair of position sensors, denoted herein by reference numerals 232D, 233D at a position of the rotor that such that any foreseeable vibration mode does not create a vibration node at the location of the position sensors 232D, 233D. As shown in Figure 2C, the distance between the rotor's support points is denoted by L, and the fractional distances from the location of the position sensors 232D, 233D to the support points are LI and L2, such that L=L1+L2. If L is not an integral multiple of LI or L2, or not even close to an integral multiple, then the location of the position sensors 232D, 233D is unlikely to be a node of higher- order vibrations.

Figure 3 is a block diagram of an automated controller implemented as a programmed data processor. It should be understood that Figure 3 shows an exemplary but non-restrictive construction and many other implementations are possible. As shown in Figure 3, the automated controller 160 comprises a central processing unit 310; an internal bus 315, including address, data and control portions; an optional management interface 320; two (in the present example) Input-Output bus controllers 330, 335; circuitry for clock and interrupt functions and related tasks, generally denoted by reference numeral 350; and memory, generally denoted by reference numeral 350.

By means of the optional management interface 320, the automated controller 160 may communicate with an optional management terminal MT. Such communication may comprise outputting of statistics and/or inputting of configuration changes, for example. The first Input-Output bus controller 330 provides communication capabilities with the variable power sources, such as frequency controllers (items 152, 154 in Figure 1 ), while the second Input- Output bus controller 335 provides communication capabilities with the position sensors 232A through 233D.

The memory 350 comprises program code 360, 370, and data 380.

The first program code portion 360, when executed by the processor 310, performs torque control, by outputting adjustment instructions to the first frequency converter 152. As a result, the first frequency converter 152 adjusts the supplied energy feed to the first set of windings 1 12, which in turn affects the conversion between electrical energy and mechanical rotational energy. This portion of the program code is well known in the prior art, and a more detailed description is superfluous.

The second program code portion 370, when executed by the processor 310, performs levitation control, by outputting adjustment instructions to the second frequency converter 154. As a result, the second frequency converter 154 adjusts the supplied energy feed to the second set of windings 1 14, which in turn affects the magnetic levitation force that the stator exerts on the rotor. Adjustment of the second frequency converter 154 is based on a comparison between actual position indications, as reported by the position sen- sors 232A through 233D, and their nominal values or value ranges, which in the present implementation are stored in the data memory portion 380. Generation of the adjustment instructions to the second frequency converter 154 as a result of the comparison between actual and nominal positions may be adjusted from the management terminal MT via the management interface 320. For the optional management functions, the memory 350 comprises an optional management program, which is not shown separately.

The optional management interface 320 may be any interface that permits a data processing apparatus to communicate with a user terminal, including but not limited to: wired interfaces, such as Ethernet, RS-232, USB, or wireless interfaces, such as Bluetooth, WLAN, infrared, or a connection via a cellular network. As regards the Input-Output buses 1 and 2, they can be implemented by any industry-standard or proprietary technology.

It should be understood by those skilled in the art that the specific embodiments described in detail are provided for the purposes of illustrating the invention and not for restricting it, and many variations are possible without departing from the scope of the invention, as defined by the following claims.