WO2001098676A2

WO2001098676A2 - Magnetic bearing having features for low power consumption, reduced weight, and fault-tolerant operation

Info

Publication number: WO2001098676A2
Application number: PCT/US2001/020007
Authority: WO
Inventors: Paul E. Allaire
Original assignee: Trinity Flywheel Power
Priority date: 2000-06-21
Filing date: 2001-06-21
Publication date: 2001-12-27
Also published as: WO2001098676A3; AU2001268687A1

Abstract

A magnetic bearing for suspending a rotatable shaft comprises a backiron (1), and a plurality of poles (2) fixedly coupled to the backiron and extending radially inward toward the rotatable shaft (6). The magnetic bearing also comprises a plurality of electromagnetic coils (4) each being circumferentially disposed around the backiron or the poles, and a magnetic target ring (5) fixedly coupled to the rotor shaft. Each of the poles is separated from the magnetic target ring by an airgap (3). The magnetic bearing further comprises one or more electrical circuits electrically coupled to the plurality of electromagnetic coils and being adapted to energize the plurality of electromagnetic coils so that magnetic flux is induced in the airgaps. The magnetic bearing includes features providing relating to fault detection and fault-tolerant operation. The magnetic bearing also includes mechanical and control-system features that reduce the power consumption of the magnetic bearing.

Description

MAGNETIC BEARING HAVING FEATURES FOR LOW POWER CONSUMPTION, REDUCED WEIGHT, AND FAULT-TOLERANT OPERATION

Field of the Invention

The present invention relates generally to magnetic bearings. More particularly, the invention relates to improvements in the physical design and control methodology for magnetic bearings.

Background of the Invention

Magnetic bearings are commonly used to support rotating components by magnetically levitating (suspending) the shafts on which such components are suspended. Figures 1 and 2 depict a conventional magnetic bearing 200. The magnetic bearing 200 includes a backiron 201, eight straight-sided, radially-oriented poles 202, a plurality of electromagnetic coils 203, a rotor 205, and a magnetic target ring 204.

Magnetic bearings provide non-contact operation, and thereby possess long service lives in comparison to other types of bearings, e.g., rolling-element bearings. Furthermore, magnetic bearings consume relatively low amounts of power. Power consumption in a magnetic bearing results from ohmic losses in the coil wires of the bearing, and rotor losses caused by eddy currents and magnetic hysteresis. It is commonly known by those skilled in the art of magnetic-bearing design that rotor losses can be substantially reduced by constructing the magnetic target ring 204 using thin magnetic laminations similar to those utilized in transformers and electric motors.

Power losses can be further reduced by minimizing eddy currents and magnetic hysteresis caused by the passage of the rotor 205 through the varying magnetic fields produced by the stationary components of the bearing 200. Power losses can also be reduced by lowering the current levels in the coils 203. Thus, a need exists for a magnetic- bearing design that minimizes power losses by minimizing eddy currents and magnetic hysteresis, and by reducing current levels in the coils 203.

Another aspect of the present invention relates to the magnitude and distribution of the bias flux within a magnetic bearing. (Bias flux refers to the steady-state component of the total magnetic flux needed to maintain the rotor 205 in a suspended condition). Bias flux levels are governed by the magnetic flux levels in the airgap between the magnetic target ring 204 and the tips of the poles 202. The noted magnetic flux is produced by the electrical currents witliin the coils 203, in conjunction with the magnitude and distribution of magnetic flux density in the magnetic target ring 204. Higher flux levels produce higher load capacity and higher force slew rates, but increase the power consumption of the magnetic bearing 200. Thus, a need currently exists for a magnetic bearing design that yields high levels of bias flux with low power consumption.

Magnetic bearings are typically designed with multiple radially-oriented poles 202 having substantially straight sides, as shown for example in U.S. Patent No. 5,111,102 ("Meeks"). This design is currently favored by magnetic-bearing designers because the minimum cross-sectional area of the pole 202 occurs at the tip of the pole 202, i.e., at the location nearest the magnetic target ring 204. Thus, magnetic-flux saturation will occur at the pole tip. Alternatively, circumferential extensions, or flux spreaders, may be used on the tips of the poles 202 to broaden the flux path at the location nearest the air gap between the tips of the poles and the magnetic target ring 204. The use of flux spreaders, however, is generally considered poor design practice because the additional magnetic material added by the flux spreaders increases the weight and cost of the bearing 200 without improving magnetic-saturation performance. Foregoing the use of flux spreaders near the pole tips causes the gaps between adjacent pole tips to be relatively large. Such large gaps cause the rotor flux to undergo substantial variations along the angular path of the magnetic target ring 204. These variations, in turn, induce substantial power losses.

Eddy currents and magnetic hysteresis in the rotor 205 are generated by the passage of each point on the magnetic target ring 204 through time varying magnetic fluxes generated by the stationary portion of the bearing 200. In conventional magnetic bearings having straight-sided poles 202, the magnetic flux density is maximum directly adjacent the tip of each pole 202, and drops to nearly zero between the poles.

So-called heteropolar bearings have a single set of radially-oriented poles 202 (see Figures 1 and 2). The individual poles 202 are either north (N) poles or south (S) poles. The coils 203 of these types of bearings are wound and controlled so that the north and south poles alternate, e.g., N-S-N-S-N-S-N-S. In the alternative, the north and south poles may be magnetically paired, e.g., N-N-S-S-N-N-S-S in an eight-pole, single-radial- plane bearing. The magnetic flux density (B) generated in the rotor target is +B at the north poles, zero (B=0) between the poles, and -B at the south poles. As the rotor 205 passes the pole tips during it normal rotation, the rate of change of the flux density with ^" respect to time varies substantially, e.g., +B, 0, -B, 0, +B, 0, -B, 0, +B, 0, -B in an eight- pole, single-radial-plane bearing wound in alternating (N-S-N-S-N-S-N-S) fashion. Hence, the magnitude of the change in the flux density is approximately 200% of B, and the flux density changes four times during one complete rotation of the magnetic target ring 204. Hence, the magnetic target ring 204 sees a substantial rate-of-change in magnetic flux, thereby leading to relatively large eddy-current and magnetic-hysteresis losses. A need therefore exists for a magnetic bearing that minimizes the change in magnetic flux experienced by a rotor such as the rotor 205 as it rotates within a bearing such as the bearing 200.

Magnetic bearings may, in the alternative, be of the so-called homopolar design. A homopolar bearing includes two or more sets of radially-oriented poles 202. The poles 202 in each set are disposed in a common radial plane. The common radial planes are axially spaced by a predetermined distance. The poles 202 in a particular radial plane are all of one type, i.e., either north or south. For example, an eight pole, two-plane bearing may have one plane of poles disposed in an N-N-N-N arrangement, and another plane disposed in an S-S-S-S arrangement. The magnetic flux density (B) generated in the magnetic target ring 204 is +B at the north poles, zero (B=0) between the poles, and -B at the south poles. Hence, the flux density in an eight pole, two-plane bearing varies between +B, 0, +B, and 0 in the plane of the north poles, and between -B, 0, -B, and 0 in the plane of the south poles. Thus, the change in flux magnitude in any one of the two planes is 100% of the bias flux B, and this change occurs four times over one rotor rotation. This pattern causes the eddy-current and magnetic-hysteresis power loss in the rotor 205 to be less in a heteropolar bearing than in a comparable homopolar bearing. One version of this type of design is described in Meeks, which discloses a permanent magnet or electromagnet placed in the axial flux path to generate constant axial-bias-flux levels.

Using thin magnetic laminations reduces the generation of eddy currents in the magnetic target ring 204 to a certain extent, but does not completely eliminate the eddy currents. Thus, a need currently exists for a magnetic bearing having provisions to further reduce eddy currents.

It is desirable to employ two or more sets of radial poles 202 in a magnetic bearing such as the bearing 200 to obtain low power operation, as indicated by the above discussion. This method of bearing construction normally requires the use of some physical means for generating a substantially constant axial bias flux between the two or more sets of radial poles 202, as taught in Meeks. Conventional magnetic bearings utilize physical means to provide the substantially constant axial bias flux, e.g., permanent magnets located between the sets of radially-oriented poles 202, or electromagnets formed by coils of wire wrapped around the axial magnetic structures that join the sets of radial poles 202.

The force slew rate of a magnetic bearing such as the bearing 200, i.e., the rate at which the bearing can generate the magnetic forces needed to maintain the rotor 205 in a desired position when external forces are applied to the rotor 205, is proportional to the bias flux. Hence, it is desirable is maintain the bias flux at a relatively high level. However, the axial-bias-flux feature taught by Meeks produces high rotor eddy-current and magnetic-hysteresis power losses because the axial-bias-flux feature causes the flux to vary between a maximum value proximate the tips of the poles 202 and a minimum value that approaches zero in the areas between the poles 202.

Rotating machines that are supported by magnetic bearings often operate in various modes that require different load capacities or force slew rates. For example, an energy-storage flywheel used for power storage or momentum control in a communications satellite may require a very small load capacity to operate at a fixed orientation in space. A relatively large load capacity may be required, however, when the satellite undergoes changes in orientation. In general, the bias flux of the bearing is chosen based on the desired load capacity and force slew rate for the bearing. Also, a point at which the lower or higher levels of bias flux will be required is usually known well in advance of that point. Conventional magnetic bearings, in general, are not capable of exerting a variable bias flux. Hence, a need exists for a magnetic bearing in which the bias flux can be varied based on expected changes in required load capacity and force slew rate. The stationary portion of a conventional magnetic bearing such as the bearing 200 typically comprises a plurality of radial poles 202, a back iron 201, and coils 203 wound around the poles 202 or the back iron 201, as noted above. The transition area between the poles 202 and the back iron 201 is typically formed as a sharp corner. Magnetic saturation usually occurs at or near these sharp corners. The magnetic saturation substantially limits the magnetic flux that can be generated in the airgaps between the tips of the poles 202 and the magnetic target ring 204, and thereby limits the amount of magnetomotive force that can be produced by the coils 203 (the magnetomotive force is equal to the number of turns in the coils 203 multiplied by the electrical current through the coils 203). Hence, a need exists for a magnetic bearing that is capable of generating a maximal amount of magnetic flux in the airgaps between the poles 203 and the magnetic target ring 204.

Typical magnetic bearings such as the bearing 200 are sometimes subject to partial mechanical or electronic failures. For example, failures can occur due to breakage of the wire in one or more of the coils 203, partial failure of a digital signal processor, channel failures in an analog controller, failures of one or more power amplifiers, digital- to-analog or analog-to-digital board failures, or a failure of other components of the bearing 200. The various signal channels of the bearing's control system can be used as input channels where information about the state of the system is desired, e.g., rotor position or coil current. The channels can also be used as output channels to control the individual bearing components that maintain the rotor 205 in magnetic suspension, e.g., the coils 203, analog-to-digital converters, or digital-to-analog converters.

Typical magnetic bearings such as the bearing 200 include a plurality of poles 202 and one or more means for controlling the magnetic bias flux produced by the poles 202. The poles 202 and the control form independent or coupled magnetic and electronic circuits that produce the magnetic forces that suspend the rotor 205. For example, an eight-pole homopolar radial bearing has eight electromagnetic coils 203. Hence, eight magnetic/electronic channels exist in such a bearing where independent control of the coils 203 is utilized. The bearing is only required, however, to produce two perpendicular forces. In the event of a multiple-channel failure, the remaining channels can potentially support continued operation of the bearing if the bearing's controller is equipped with appropriate logic. Hence, a need exists for a magnetic bearing having a controller that can support operation of the bearing following a multiple-channel failure.

Summary of the Invention

A presently-preferred magnetic bearing for suspending a rotatable shaft comprises a backiron, and a plurality of poles fixedly coupled to the backiron and extending radially inward toward the rotatable shaft. Each of the plurality of poles has a curved surface portion that adjoins the backiron. The magnetic bearing further comprises a plurality of electromagnetic coils each being circumferentially disposed around one of the backiron and a respective one of the poles, and a magnetic target ring fixedly coupled to the rotor shaft. Each of the plurality of poles is separated from the magnetic target ring by an airgap. The magnetic bearing also comprises one or more electrical circuits electrically coupled to the plurality of electromagnetic coils and being adapted to energize the plurality of electromagnetic coils so that magnetic flux is induced in the airgaps.

A presently-preferred magnetic bearing system for suspending a rotatable shaft comprises a backiron positioned around the rotatable shaft, and a plurality of poles extending radially inward from the backiron and each terminating at a respective tip portion. Each of the poles comprises a transition section having a curved surface portion that adjoins the backiron. The magnetic bearing system also comprises a plurality of electromagnetic coils each being wound around one of the backiron and a respective one of the poles, and a magnetic target ring fixedly coupled to the rotor shaft and facing the tip portion to that an airgap is formed between each of the pole tips and the magnetic target ring. The magnetic bearing system further comprises one or more sources of electrical power for selectively energizing the electromagnetic coils so that magnetic flux is induced in the airgaps.

A presently-preferred embodiment of a magnetic bearing system comprises a first plurality of substantially co-planar, radially-oriented poles, a first plurality of flux spreaders each being fixedly coupled to an end of a respective one of the first plurality of poles, and a first plurality of electromagnetic coils each being wound around one of the backiron and a respective one of the first plurality of poles. The magnetic bearing system further comprises a second plurality of substantially co-planar, radially-oriented poles axially spaced from the first plurality of substantially co-planar poles, a second plurality of flux spreaders each being fixedly coupled to an end of a respective one of the second plurality of poles, and a second plurality of coils each being wound around one of the backiron and a respective one of the second plurality of substantially co-planar poles.

The presently-preferred embodiment of the magnetic bearing system also comprises a control system electrically coupled to the first and the second plurality of electromagnetic coils and being adapted to energize the first and the second plurality of electromagnetic coils so that the first plurality of electromagnetic coils and the first plurality of poles generate a radially-polarized magnetic flux having a first polarity, and the second plurality of electromagnetic coils and the second plurality of poles generate a radially-polarized magnetic flux having a second polarity, where the second polarity is substantially opposite the first polarity.

Another presently-preferred embodiment of a magnetic bearing system for suspending a rotatable shaft comprises a magnetic target ring fixedly coupled to the rotatable shaft, a first backiron, and a first plurality of substantially co-planar poles extending radially inward from the backiron. The magnetic bearing system also comprises a first plurality of flux spreaders each being fixedly coupled to an end of a respective one of the first plurality of poles so that the first plurality of flux spreaders are spaced apart from the magnetic target ring by a first plurality of airgaps. The magnetic bearing system further comprises a first plurality of electromagnetic coils each being wound around one of the backiron and a respective one of the first plurality of poles.

The presently-referred magnetic bearing system also comprises a second backiron axially spaced from the first backiron, and a second plurality of substantially co- planar poles extending radially inward from the second backiron and being axially spaced from the first plurality of substantially co-planar poles. The magnetic bearing system also comprises a second plurality of flux spreaders each being fixedly coupled to an end of a respective one of the second plurality of poles so that the second plurality of flux spreaders are spaced apart from the magnetic target ring by a second plurality of airgaps. The magnetic bearing system further comprises a second plurality of electromagnetic coils each being wound around one of the backiron and a respective one of the second plurality of substantially co-planar poles. The presently-referred magnetic bearing system also comprises a control system electrically coupled to the first and the second plurality of electromagnetic coils and being adapted to energize the first and the second plurality of electromagnetic coils so that a magnetic flux having a first polarity is produced in the fist plurality of airgaps and a magnetic flux having a second polarity is produced in the second plurality of airgaps, where second polarity is substantially opposite the first polarity.

Another presently-preferred embodiment of a magnetic bearing system for suspending a rotating shaft comprises a backiron, a plurality of poles fixedly coupled to the backiron and extending radially inward from the backiron, and a plurality of electromagnetic coils wound around a circumference of one of te backiron and a respective one of the poles. The magnetic bearing system also comprises a magnetic target ring fixedly coupled to the shaft and separated from the plurality of poles by a plurality of airgaps, and a control system electrically coupled to the plurality of electromagnetic coils and being adapted to energize the plurality of electromagnetic coils so that the plurality of electromagnetic coils and the plurality of poles generate a variable, axially-polarized magnetic flux.

A presently-preferred method for detecting faults in a magnetic bearing system comprises altering a current in an electromagnetic coil of the magnetic bearing system to cause a perturbation in the current, measuring a response of a component of the magnetic bearing system to the perturbation, and comparing the response of the component to a predetemiined response of the component to a substantially identical perturbation.

A presently-preferred magnetic bearing system having fault-detection capabilities comprises a backiron, a plurality of poles fixedly coupled to the backiron and extending radially inward from the backiron, and a plurality of electromagnetic coils each being wound around one of the backiron and a respective one of the poles. The magnetic bearing system also comprises a magnetic target ring fixedly coupled to the shaft and separated from the poles by a plurality of airgaps.

The magnetic bearing system further comprises a control system comprising a source of electrical power electrically coupled to the electromagnetic coil, a microprocessor electrically coupled to the source of electrical power, a memory-storage device electrically coupled to the microprocessor, and a set of computer-executable instructions stored on the memory-storage device. The computer-executable instructions alter a current in an electromagnetic coil of the magnetic bearing system to cause a perturbation in the current, measure a response of a component of the magnetic bearing system to the perturbation, and compare the response of the component to a predetermined response of the component to a substantially identical perturbation.

A presently-preferred method for operating a magnetic bearing system in a fault-tolerant manner comprises controlling a first set of outputs from a control system of the magnetic bearing system based on a first set of inputs to the control system and using a first set of mathematical equations, and detecting the presence of a fault in one or more of the control inputs and the control outputs. The presently-preferred method also comprises modifying the first set of mathematical equations to form a second set of mathematical equations, and controlling a second set of outputs from the control system based on second set of inputs to the control system and using the second set of mathematical equations, where the second set of outputs and the second set of inputs exclude the one or more of the control inputs and the control outputs having the fault therein.

A presently-preferred a fault-tolerant magnetic bearing system comprises a backiron, a plurality of poles fixedly coupled to the backiron and extending radially inward from the backiron, and a plurality of electromagnetic coils each being wound around one of the backiron and a respective one of the poles. The fault-tolerant magnetic bearing system also comprises a magnetic target ring fixedly coupled to the shaft and separated from the poles by a plurality of airgaps, and a control system. The control system comprises a source of electrical power electrically coupled to the electromagnetic coil, a microprocessor electrically coupled to the source of electrical power, a memory-storage device electrically coupled to the microprocessor, and a set of computer-executable instructions stored on the memory-storage device.

The computer-executable instructions of the fault-tolerant magnetic bearing system control a first set of outputs from a control system of the magnetic bearing system based on a first set of inputs to the control system and using a first set of mathematical equations, and detect the presence of a fault in one or more of the control inputs and the control outputs. The computer-executable instructions also modify the first set of mathematical equations to form a second set of mathematical equations, and control a second set of outputs from the control system based on second set of inputs to the control system and using the second set of mathematical equations, where the second set of outputs and the second set of inputs exclude the one or more of the control inputs and the control outputs having the fault therein.

Another presently-preferred magnetic bearing for suspending a rotatable shaft comprises a backiron, a plurality of poles fixedly coupled to the backiron and extending radially inward toward the rotatable shaft, and a plurality of electromagnetic coils each being circumferentially disposed around one of the backiron and a respective one of the poles. The presently-preferred magnetic bearing also comprises a magnetic target ring fixedly coupled to the rotor shaft, where each of the plurality of poles are separated from the magnetic target ring by an airgap. The presently-preferred magnetic bearing further comprises one or more electrical circuits electrically coupled to the plurality of electromagnetic coils and being adapted to energize the plurality of electromagnetic coils so that magnetic flux is induced in the airgaps, where the one or more electrical circuits comprise an external coil.

Another presently-preferred a magnetic bearing system for suspending a rotatable shaft comprises a first backiron, a first plurality of substantially co-planar poles extending radially inward from the backiron, and a first plurality of electromagnetic coils each being wound around one of the backiron and a respective one of the first plurality of poles. The presently-preferred magnetic bearing system also comprises a second backiron axially spaced from the first backiron, and a second plurality of substantially co-planar poles extending radially inward from the second backiron and being axially spaced from the first plurality of substantially co-planar poles.

The presently-preferred magnetic bearing system also comprises a second plurality of coils each being wound around one of the backiron and a respective one of the second plurality of substantially co-planar poles, where each of the second plurality of electromagnetic coils has an angular position substantially equal to an angular position of a respective one of the first plurality of electromagnetic coils. The magnetic bearing system also comprises a control system electrically coupled to the first and the second pluralities of electromagnetic coils and being adapted to energize the first and the second pluralities of electromagnetic coils so that substantially identical electrical currents are sent to the electromagnetic coils having substantially identical angular positions, whereby the magnetic bearing generates an axially-polarized magnetic flux.

A presently-preferred method of operating a magnetic bearing to produce axially-polarized magnetic flux using a first plurality of substantially co-planar, radially- oriented electromagnetic coils axially spaced from a second plurality of substantially co- planar, radially-oriented electromagnetic coils comprises aligning each of the first plurality of radially-oriented electromagnetic coils with a respective one of the second plurality of radially-oriented electromagnetic coils so that each of the first plurality of radially-oriented electromagnetic coils has an angular position substantially equal to an angular position of the respective one of the second plurality of radially-oriented electromagnetic coils. The presently-preferred method also comprises directing substantially identical electrical currents to the electromagnetic coils having substantially identical angular positions.

Another presently-preferred magnetic bearing for suspending a rotatable shaft comprises a backiron positioned around the rotatable shaft, a plurality of poles fixedly coupled to the backiron, and a plurality of electromagnetic coils each being wound around one of the backiron and a respective one of the poles. The plurality of electromagnetic coils are adapted to generate magnetic flux in response to electrical currents sent thereto. The presently-preferred magnetic bearing further comprises a magnetic target ring fixedly coupled to the shaft, and a control system electrically coupled to the plurality of electromagnetic coils. The control system is adapted to energize the plurality of electromagnetic coils so that the magnetic flux varies in response to predicted changes in operating conditions of the magnetic bearing.

Another presently-preferred magnetic bearing for suspending a rotatable shaft comprise a backiron, a plurality of poles fixedly coupled to the backiron and extending radially inward from the backiron, and a plurality of electromagnetic coils each being wound around one of the backiron and a respective one of the poles.

The presently-preferred magnetic bearing further comprises a magnetic target ring fixedly coupled to the shaft, and a control system. The control system comprises a source of electrical power electrically coupled to the plurality of electromagnetic coils, a microprocessor electrically coupled to the source of electrical power, a memory-storage device electrically coupled to the microprocessor, and a set of computer-executable instructions stored on the memory-storage device. The computer- executable instructions cause the control system to energize the electromagnetic coils so that the magnetic bearing exerts a first predetermined level force on the rotatable shaft under a first set of operating conditions, and the magnetic bearing exerts a second predetermined level of force on the rotatable shaft under a second set of operating conditions.

A presently-preferred method of minimizing power consumption in a magnetic bearing that supports a rotatable shaft comprises energizing electromagnetic coils of the magnetic bearings with a first set of electrical currents so that the magnetic bearing generates a first level of force required to suspend the rotatable shaft under a first set of operating conditions. The presently-preferred method also comprises predicting changes in the operating conditions of the magnetic bearing using predetermined information, and determining a second level of force required to suspend the rotatable shaft after the predicted changes in the operating conditions of the magnetic bearing. The presently- preferred method further comprises energizing the electromagnetic coils with a second set of electrical currents so that the magnetic bearing exerts the second level of force on the rotatable shaft.

Brief Description of the Drawings

The foregoing summary, as well as the following detailed description of a presently-preferred embodiment, is better understood when read in conjunction with the appended drawings. For the purpose of illustrating the invention, the drawings show an embodiment that is presently preferred. The invention is not limited, however, to the specific instrumentalities disclosed in the drawings. In the drawings:

Fig. 1 is a cross-sectional axial view of a conventional heteropolar magnetic bearing;

Fig. 2 is a cross-sectional side view of the conventional heteropolar magnetic bearing shown in Fig. 1;

Fig. 3 A is a cross-sectional axial view of a homopolar magnetic bearing in accordance with the present invention; Fig. 3B is a magnified view of the area designated "3B" in Figure 3A;

Fig. 4 is a cross-sectional side view of the homopolar magnetic bearing shown in Figs. 3A and 3B;

Fig. 5 is a cross-sectional axial view a first plane of the homopolar magnetic bearing shown in Figs. 3A-4 including electrical circuits of the bearing;

Fig. 6 is a cross-sectional axial view of a second plane of the homopolar magnetic shown in Figs. 3A-5 including electrical circuits of the bearing;

Fig. 7 A is a diagrammatical illustration of a control system of the magnetic bearing shown in Figs. 3 A-6;

Fig. 7B is a block diagram depicting the operation of the control system shown in Fig. 7A;

Fig. 8A is a graphical illustration of a variation in bias flux experienced by a rotating target of the magnetic bearing shown in Figs. 3A-6;

Fig. 8B is a graphical illustration of a variation in bias flux experienced by a rotating target of a conventional magnetic bearing of the type shown in shown in Figs. 1 and 2;

Fig. 9 is a side view of the magnetic bearing shown in Figs. 3 A-6 including axial electrical circuits of the bearing; and

Fig. 10 is a diagrammatical illustration of an alternative embodiment of an electrical circuit of the magnetic bearing shown in Figs. 3A-6 and 9.

Description of Preferred Embodiments

Figures 3-7A and 9 depict a preferred embodiment of an eight-pole homopolar magnetic bearing 90 in accordance with the present invention. This particular bearing configuration is described in detail for exemplary purposes only. The various features of the invention can be used in conjunction with other types of bearing configurations, including heteropolar bearings and bearings having more or less than eight poles. The magnetic bearing 90 may be used in, for example, an energy storage flywheel utilized for power storage or momentum control in a communications satellite; an energy- storage flywheel for non-interruptible power supply in a ground-based application; or a flywheel in a power system for an electric vehicle. Other applications for the magnetic bearing 90 are also possible.

The magnetic bearing 90 comprises a backiron 1, eight radially-oriented poles 2, a plurality of flux spreaders 3, a plurality of electromagnetic coils 4, a rotor shaft 6, and a magnetic target ring 5 fixedly coupled to an outer circumference of the shaft 6. Four of the poles 2 are positioned in a first radial plane, and form a first set 2a of the poles 2 (see Figure 4). The remainder of the poles 2 are positioned in a second radial plane, and form a second set 2a of the poles 2. Hence, the first and the second sets 2a and 2b are axially spaced, as shown in Figure 4. In one particular embodiment of the invention, each pole 2 in the first set 2a is substantially aligned with a respective one of the poles 2 in the second set 2b so that the poles 2 in the first and second sets 2a, 2b have substantially identical angular (clock) positions. The significance of this feature is discussed below.

The back iron 1 provides a path for magnetic flux between adjacent poles 2. The poles 2 conduct magnetic flux from the backiron 1 to the flux spreaders 3. The flux spreaders 3 are disposed on radially-inward tips 2c of the poles 2, and minimize changes in magnetic flux experienced by the magnetic target ring 5 as the target ring 5 rotates (see Fig. 3A). The electromagnetic coils 4 and the poles 2 interact to produce magnetic flux in airgaps 8 between the magnetic target ring 5 and the pole tips 2c. The magnetic flux generates a magnetomotive force that magnetically suspends the target ring 5 and the rotor shaft 6.

In accordance with the present invention, each of the poles 2 includes a transition section 7 that adjoins the backiron 1. Each of the transition sections 7 includes rounded surfaces 7a that extend between the respective pole 2 and the backiron 1. The rounded surfaces 7a reduce magnetic saturation at the interface between the poles 2 and the backiron 1 in comparison to a conventional pole-bacl iron interface in which the pole and backiron form a substantially square corner. Reducing magnetic saturation at the interface between the poles 2 and the baclάron 1 results in higher levels of magnetic-flux density in the airgaps 8, and thus lowers the power requirements of the magnetic bearing system 90. Hence, the magnetic bearing 10 can be constructed with a lighter and more compact configuration than a conventional magnetic bearing of comparable capabilities. The magnetic bearing 90 further comprises eight electrical circuits 101 (see Figures 5 and 6; the flux spreaders 3 are not shown in Figures 5 and 6 for clarity). Each electrical circuit 101 includes a control digital signal processor 105 electrically coupled to a low-power-consumption switching power amplifier 106.

Each electromagnetic coil 4 is electrically coupled to one of the electrical circuits 101, and is wound around one of the poles 2. The electromagnetic coils 4 are powered and controlled by a control system 300 (see Figure 7A). The control system 300 comprises a controller 301, a digital-to-analog converter 302, an analog-to-digital converter 304, a power amplifier 305, one or more displacement probes 306, and an anti- alias filter 307. The controller 301 comprises a microprocessor 308, a memory-storage device 310, and a set of computer-executable instructions 312 stored on the memory- storage device 310. The operation of the control system 300 is illustrated in the form of a block diagram in Figure 7 A. The controller 301 functions as both a state-space controller 301a and an anti-imbalance controller 301b, as denoted in the Figures 7A and 7B.

The electromagnetic coils 4 and the poles 2 produce a radially-polarized magnetic flux in the airgaps 8 when the electromagnetic coils 4 are energized by the control system 300. The electrical currents to the electromagnetic coils 4 are controlled in a manner that causes the poles 2 in the first set 2a to act as "north" poles, and the poles 2 in the second set 2b to act as "south" poles. In other words, the magnetic flux generated by the poles 2 in the first set 2a has a polarity substantially opposite that of the magnetic flux generated by the poles 2 in the second set 2b.

The use of the flux spreaders 3, in conjunction with the axially-spaced sets 2a, 2b of poles 2, represents a further aspect of the present invention. More particularly, the flux spreaders 3 minimize the distance between adjacent pole tips 2c, as shown in Figure 3. In addition, the poles 2 are positioned so that poles 2 of the same polarity are grouped in a common plane. This arrangement minimizes the variation in magnetic flux experienced by an individual point on the magnetic target ring 5 as that point rotates into and out of proximity with the tips 2c of the poles 2. Minimizing the variation in magnetic flux produces a corresponding reduction in the eddy currents and magnetic hysteresis within the target ring 5 and the shaft 6, and thereby lowers the power requirements of the magnetic bearing system 90. Hence, this particular aspect of the invention permits a lighter and more compact construction for the magnetic bearing 90 in comparison to a conventional magnetic bearing of comparable capabilities.

Figure 8 A conceptually illustrates the variation in the magnetic flux experienced by an individual point on the magnetic target ring 5 due to the effects of the flux spreaders 3 and the axially-spaced sets 2a, 2b of poles 2. The magnetic flux is presented as a function of the angular position of the individual point. The magnetic flux generated by each combination of pole 2 and electromagnetic coil 4 is designated "B" in Figure 8A (for clarity, only bias flux is depicted in this example). The minimum magnetic flux experienced at a point on the magnetic target ring 5 is designated as a fractional value " " of the magnetic flux B ( may have a value within a range of, for example, approximately 0.8 to 0.95 in the exemplary magnetic bearing 90). Hence, the variation in the magnetic flux experienced by that point on the magnetic target ring 5 varies between B, a B, B, B, B, α B, B, B during one rotation of the magnetic target ring 5. The magnetic flux therefore decreases by an amount equal to (l-α)B, where the term 1-α would range from approximately 0.20 to 0.05, or approximately 20% to 5% of the bias flux value B, in this particular example. The magnetic-flux variation in a conventional magnetic bearing operating under the same conditions, by contrast, would decrease by an amount approximately equal to B (see Figure 8B).

Alternative embodiments of the invention may comprise relatively thin magnetic strips that connect adjacent flux spreaders 3 so that no gap exists between the flux spreaders 3. This connecting strip, or saturation link, is designed to be magnetically saturated by the bias flux. This feature causes the magnetic fluxes of the poles to be sufficiently separated from one another to allow for the differences in radial flux that generate the radial load capacity in the bearing 90. Furthermore, each of the poles 2 preferably comprises a second transition section 9 having rounded surfaces 9a that extend between the respective pole 2 and the corresponding flux spreader 3. The rounded surfaces 9a reduce magnetic saturation at the interface between the poles 2 and the flux spreaders 3, and thereby produces the previously-noted benefits associated with such reductions.

Another aspect of the present invention relates to the generation of axial flux without the use of stationary permanent magnets, and without the use of electromagnetic coils used exclusively for that purpose. Rather, the magnetic bearing 90 generates an axially-polarized magnetic field by way of the electromagnetic coils 4, a plurality of axial electrical circuits 401 each electrically coupled to a respective one of the electromagnetic coils 4, and the controller 301 (see Figure 9). This feature is achieved using digital signal processing control methods that cause the electromagnetic coils 4 to produce electromagnetic fields that result in both radial and axial magnetic-flux paths.

The axial electrical circuits 401 may be controlled independently, as shown in Figure 9. Alternatively, the axial electrical circuits 401 may be electrically coupled, and controlled as a single circuit. Each axial electrical circuit 401 includes a digital signal process controller 405 and a switching power amplifier 406 electrically coupled to the controller 301, as shown in Figure 9. The controller 301 controls the flow of current to the electromagnetic coils 4 so that the electromagnetic coils 4 simultaneously generate the axial and radial flux paths noted above. More particularly, the controller 301 directs substantially identical electrical currents to poles 2 having substantially identical angular positions. (As previously noted, each pole 2 in the first set 2a of poles 2 is substantially aligned with a respective one of the poles 2 in the second set 2b so that the poles 2 in the first and second sets 2a, 2b have substantially identical angular positions. Hence, each of the electromagnetic coils 4 positioned on the first set 2a of poles 2 has a substantially identical angular position in relation to one of the electromagnetic coils 4 positioned on the second set 2b of poles 2).

The use of substantially identical electrical currents in electromagnetic coils 4 having substantially identical angular positions generates an axially-polarized magnetic flux. (Note: The currents used in radially-opposed electromagnetic coils 4 should, in general, be unequal in order to achieve the net radial force needed to suspend the rotor 6.) The generation of axially-polarized magnetic flux through the exclusive use of the electromagnetic coils 4 obviates the need for permanent magnets or separate electromagnetic coils to accomplish this function. This feature can thereby lead to substantial reductions in the overall weight, volume, and power consumption of the bearing 90. In addition, this feature permits the axial flux to be varied in response to changes in the operating conditions of the magnetic bearing 90. More particularly, the axial flux produced by the electromagnetic coils 4 can be increased or decreased via the controller 301 in response to anticipated changes in the for axial flux. This feature permits the power consumption of the magnetic bearing 90 to be more closely tailored to the operating conditions of the magnetic bearing 90 than would otherwise be possible. Details concerning this additional aspect of the invention follow.

Rotors that are supported by magnetic bearings may be subject to both large and small external forces during their normal course of operation. For example, high- performance energy-storage flywheels and control-moment gyroscopes are sometimes used in communications satellites positioned in low-earth orbit. A bearing rotor in these devices may operate for prolonged periods in a state of slow acceleration or deceleration as power is input to or removed from the rotor via a generator or a motor. The external forces imposed on the rotor may be very low under such operating conditions, and the bias flux may be reduced by the control system 300 to substantially reduce power losses due to coil currents and rotating eddy currents. The bias flux must be increased to develop a higher load capacity or force slew rate, however, when high rates of acceleration or deceleration are required, or when control moments must be produced. Note: The term "bias flux," as used throughout the specification and claims, refers to the steady-state component of the total magnetic flux required to maintain the rotor 6 in a suspended condition.

The need for changes in the performance, i.e., load capacity and slew rate, of a magnetic bearing is usually known in advance. In accordance with the present invention, this knowledge is used by the control system 300 to dynamically modify the axial bias flux. More particularly, the computer-executable instructions 312 alter the axial bias flux in response to anticipated changes in the required load capacity or force slew rate of the bearing 90. Hence, the control system reduces the total current flow to the electromagnetic coils when anticipated changes in the operating condition of the magnetic bearing 90 lower the net amount of axial bias flux needed to maintain the magnetic bearing 90 in a suspended condition. This feature can lower the overall power consumption of the magnetic bearing 90 by permitting the magnetic flux to be more closely tailored to the operating condition of the magnetic bearing 90 than would otherwise be possible. This feature can also be applied to the radial bias flux of the magnetic bearing 90. In other words, the radial bias flux can also be altered in response to anticipated changes in the level of radial flux needed to suspend the rotor 6. Another feature of the present invention relates to detection of component faults in the magnetic bearing 90. More particularly, the controller 301 is adapted to detect the presence of faults in various components of the magnetic bearing 90. The controller 301 performs this function by periodically causing a perturbation, e.g., a pulse, in the currents sent to the electromagnetic coils 4 so as to excite one or more components of the magnetic bearing system 90. The responses of the various components of the bearing 90 to the perturbations are measured and then compared to predetermined responses of the same components under normal operating conditions, i.e., without any component faults (the predetermined responses are stored in the memory-storage device 310). Differences between the as-measured and predetermined responses greater than a predetermined level are construed as an indication of a fault in the corresponding component. For example, the currents to one or more of the electromagnetic coils 4 can be pulsed periodically, and the responses of the other electromagnetic coils 4 and position sensors of the magnetic bearing 90 can be measured to facilitate condition monitoring of the vibration level of the shaft 6.

In accordance with a further aspect of the invention, the controller 301 is adapted to respond to a detected fault in a manner that permits continued operation of the magnetic bearing 90. More particularly, the computer-executable instructions 312, upon detecting the failure of a particular component of the magnetic bearing 90, reconfigures the control methodology for the magnetic bearing 90 so that the control methodology no longer relies on the input or output associated with that component. In other words, the controller 301 relies on a different combination (set) of control algorithms so that the input or output associated with the faulty component is no longer needed for the satisfactory operation of the magnetic bearing 90.

For example, in the event that one of the rotor-position or current-sensing devices in the magnetic bearing 90 fails, the computer-executable instructions 312 delete the input corresponding to that sensor from the control algorithms. In addition, the computer-executable instructions 312 dynamically reconfigure the control algorithms, i.e., the mathematical model of the magnetic bearing 90 stored in the controller 301, for continued operation without the noted input. As a further example, if a power amplifier powering one of the electromagnetic coils 4 fails, the output channel corresponding to that amplifier is deleted from the control algorithms, and the set of algorithms that controls the magnetic bearing 90 is dynamically reconfigured for continued operation without the amplifier.

Another aspect of the present invention is directed to improving the quality of the inductance in the electrical circuits 104. More particularly, the inductance in the electrical circuits 104 may ill-conditioned under certain circumstances, e.g., when the electrical circuits 104 are configured so as to be fully-independent of each other. Ill- conditioned inductance can results in ill-conditioning of the various currents throughout the magnetic bearing 90. The present invention addresses this problem by the use of an external coil in one or more of the electrical circuits of a magnetic bearing such as the magnetic bearing 90.

For example, Figure 10 depicts an electrical circuit 101a that may be used in the magnetic bearing 90. The electrical circuit 101a is substantially identical to the previously-described electrical circuit 101, with the exception that the electrical circuit 101a comprises an external coil 4a. (Components of the electrical circuit 101a that are substantially identical to those in the electrical circuit 101 are designated by common reference numerals.) The external coil 4a is used in addition to the electromagnetic coil 4 positioned on the pole 2, and is positioned external to the structure represented by the backiron 1 and the poles 2. In other words, the coil 4a is spaced apart from the backiron 1 and the poles 2. The use of the external coil 4a in the electrical circuit 101a assists in providing proper conditioning of the coil inductance matrix of the magnetic bearing 90.

It is to be understood that even though numerous characteristics and advantages of the present invention have been set forth in the foregoing description, together with details of the structure and function of the invention, the disclosure is illustrative only, and changes may be made in detail, especially in matters of shape, size, and arrangement of the parts, within the principles of the invention to the full extent indicated by the broad general meaning of the terms in which the appended claims are expressed.

Claims

What is claimed is:

1. A magnetic bearing for suspending a rotatable shaft, comprising: a baclciron; a plurality of poles fixedly coupled to the backiron and extending radially inward toward the rotatable shaft, each of the plurality of poles having a curved surface portion that adjoins the baclciron; a plurality of electromagnetic coils each being circumferentially disposed around one of the backiron and a respective one of the poles; a magnetic target ring fixedly coupled to the rotor shaft, each of the plurality of poles being separated from the magnetic target ring by an airgap; and one or more electrical circuits electrically coupled to the plurality of electromagnetic coils and being adapted to energize the plurality of electromagnetic coils so that magnetic flux is induced in the airgaps.

2. The magnetic bearing of claim 1, wherein each of the curved surface portions extend through an arc of approximately ninety degrees.

3. The magnetic bearing of claim 1, wherein the pole has a second curved surface portion adjoining the flux spreader.

4. The magnetic bearing of claim 1 , wherein the backiron and the pole are unitarily formed.

5. The magnetic bearing of claim 1, wherein the one or more electrical circuits each comprise a control digital signal processor and a low-power-consumption switching power amplifier electrically coupled to the control digital signal processor.

6. The magnetic bearing of claim 1, wherein four of the poles are fixedly coupled to the backiron.

7. A magnetic bearing system for suspending a rotatable shaft, comprising: a backiron positioned around the rotatable shaft; a plurality of poles extending radially inward from the backiron and each terminating at a respective tip portion, each of the poles comprising a transition section having a curved surface portion that adjoins the backiron; a plurality of electromagnetic coils each being wound around one of the baclciron and a respective one of the poles; a magnetic target ring fixedly coupled to the rotor shaft and facing the tip portion to that an airgap is formed between each of the pole tips and the magnetic target ring; and one or more sources of electrical power for selectively energizing the electromagnetic coils so that magnetic flux is induced in the airgaps.

8. The magnetic bearing of claim 1, wherein each of the curved surface portions extend through an arc of approximately ninety degrees.

9. The magnetic bearing of claim 7, wherein the pole has a second curved surface portion adjoining the flux spreader.

10. The magnetic bearing of claim 7, wherein the one or more sources of electrical power comprises one or more electrical circuits each comprising a control digital signal processor and a low-power-consumption switching power amplifier electrically coupled to the control digital signal processor.

11. The magnetic bearing of claim 7, wherein four of the poles extend radially inward from the baclciron.

12. A magnetic bearing system, comprising: a first plurality of substantially co-planar, radially-oriented poles; a first plurality of flux spreaders each being fixedly coupled to an end of a respective one of the first plurality of poles; a first plurality of electromagnetic coils each being wound around one of the backiron and a respective one of the first plurality of poles; a second plurality of substantially co-planar, radially-oriented poles axially spaced from the first plurality of substantially co-planar poles; a second plurality of flux spreaders each being fixedly coupled to an end of a respective one of the second plurality of poles; a second plurality of coils each being wound around one of the backiron and a respective one of the second plurality of substantially co-planar poles; and a control system electrically coupled to the first and the second plurality of electromagnetic coils and being adapted to energize the first and the second plurality of electromagnetic coils so that the first plurality of electromagnetic coils and the first plurality of poles generate a radially-polarized magnetic flux having a first polarity, and the second plurality of electromagnetic coils and the second plurality of poles generate a radially-polarized magnetic flux having a second polarity, the second polarity being substantially opposite the first polarity.

13. The magnetic bearing system of claim 12, further comprising a first plurality of saturation links each adjoining adjacent ones the first plurality of flux spreaders, and a second plurality of saturation links each adjoining adjacent ones the second plurality of flux spreaders.

14. The magnetic bearing system of claim 12, wherein the control system comprises a controller and a power amplifier electrically coupled to the controller.

15. The magnetic bearing system of claim 14, wherein the controller comprises a microprocessor, a memory-storage device electrically coupled to the microprocessor, and a set of computer executable instructions stored on the memory- storage device.

16. The magnetic bearing system of claim 13, wherein the first plurality of saturation links are adapted to be magnetically saturated by the first radially-polarized magnetic flux and the second plurality of saturation links are adapted to be magnetically saturated by the second radially-polarized magnetic flux.

17. The magnetic bearing system of claim 12, wherein the first and the second plurality of poles each have a curved surface portion that adjoins a respective one of the first and second plurality of flux spreaders.

18. The magnetic bearing system of claim 17, wherein the curved surface portion extends through an arc of approximately ninety degrees.

19. The magnetic bearing system of claim 12, wherein the first plurality of substantially co-planar, radially-oriented poles comprises four of the substantially co- planar, radially-oriented poles and the second plurality of substantially co-planar, radially- oriented poles comprises four of the substantially co-planar, radially-oriented poles.

20. A magnetic bearing system for suspending a rotatable shaft, comprising: a magnetic target ring fixedly coupled to the rotatable shaft; a first backiron; a first plurality of substantially co-planar poles extending radially inward from the baclciron; a first plurality of flux spreaders each being fixedly coupled to an end of a respective one of the first plurality of poles so that the first plurality of flux spreaders are spaced apart from the magnetic target ring by a first plurality of airgaps; a first plurality of electromagnetic coils each being wound around one of the backiron and a respective one of the first plurality of poles; a second backiron axially spaced from the first backiron; a second plurality of substantially co-planar poles extending radially inward from the second baclciron and being axially spaced from the first plurality of substantially co-planar poles; a second plurality of flux spreaders each being fixedly coupled to an end of a respective one of the second plurality of poles so that the second plurality of flux spreaders are spaced apart from the magnetic target ring by a second plurality of airgaps; a second plurality of electromagnetic coils each being wound around one of the baclciron and a respective one of the second plurality of substantially co- planar poles; and a control system electrically coupled to the first and the second plurality of electromagnetic coils and being adapted to energize the first and the second plurality of electromagnetic coils so that a magnetic flux having a first polarity is produced in the fist plurality of airgaps and a magnetic flux having a second polarity is produced in the second plurality of airgaps, the second polarity being substantially opposite the first polarity.

21. The magnetic bearing system of claim 20, wherein the first plurality of substantially co-planar poles comprises four of the substantially co-planar poles and the second plurality of substantially co-planar poles comprises four of the substantially co- planar poles.

22. The magnetic bearing system of claim 20, wherein the control system comprises a controller and a power amplifier electrically coupled to the controller.

23. The magnetic bearing system of claim 22, wherein the controller comprises a microprocessor, a memory-storage device electrically coupled to the microprocessor, and a set of computer executable instructions stored on the memory- storage device.

24. A magnetic bearing system for suspending a rotating shaft, comprising: a backiron; a plurality of poles fixedly coupled to the baclciron and extending radially inward from the baclciron; a plurality of electromagnetic coils wound around a circumference of one of the backiron and a respective one of the poles; a magnetic target ring fixedly coupled to the shaft and separated from the plurality of poles by a plurality of airgaps; and a control system electrically coupled to the plurality of electromagnetic coils and being adapted to energize the plurality of electromagnetic coils so that the plurality of electromagnetic coils and the plurality of poles generate a variable, axially-polarized magnetic flux.

25. A method for detecting faults in a magnetic bearing system, comprising: altering a current in an electromagnetic coil of the magnetic bearing system to cause a perturbation in the current; measuring a response of a component of the magnetic bearing system to the perturbation; and comparing the response of the component to a predetermined response of the component to a substantially identical perturbation.

26. The method of claim 25, wherein altering a current in an electromagnetic coil of the magnetic bearing system to cause a perturbation in the current comprises pulsing the current.

27. The method of claim 25, wherein altering a current in an electromagnetic coil of the magnetic bearing system to cause a perturbation in the current comprises pulsing the current on a periodic basis.

28. A magnetic bearing system having fault-detection capabilities, comprising: a baclciron; a plurality of poles fixedly coupled to the backiron and extending radially inward from the baclciron; a plurality of electromagnetic coils each being wound around one of the backiron and a respective one of the poles; a magnetic target ring fixedly coupled to the shaft and separated from the poles by a plurality of airgaps; and a control system comprising: a source of electrical power electrically coupled to the electromagnetic coil; a microprocessor electrically coupled to the source of electrical power; a memory-storage device electrically coupled to the microprocessor; and a set of computer-executable instructions stored on the memory-storage device, wherein the computer-executable instructions: alter a current in an electromagnetic coil of the magnetic bearing system to cause a perturbation in the current; measure a response of a component of the magnetic bearing system to the perturbation; and compare the response of the component to a predetermined response of the component to a substantially identical perturbation.

29. A method for operating a magnetic bearing system in a fault-tolerant manner, comprising: controlling a first set of outputs from a control system of the magnetic bearing system based on a first set of inputs to the control system and using a first set of mathematical equations; detecting the presence of a fault in one or more of the control inputs and the control outputs; modifying the first set of mathematical equations to form a second set of mathematical equations; and controlling a second set of outputs from the control system based on second set of inputs to the control system and using the second set of mathematical equations, wherein the second set of outputs and the second set of inputs exclude the one or more of the control inputs and the control outputs having the fault therein.

30. The method of claim 29, wherein detecting the presence of a fault in one or more of the control inputs and the control outputs comprises: altering a current in an electromagnetic coil of the magnetic bearing system to cause a perturbation in the current; measuring a response of a component of the magnetic bearing system to the perturbation; and comparing the response of the component to a predetermined response of the component to a substantially identical perturbation.

31. A fault-tolerant magnetic bearing system, comprising: a backiron; a plurality of poles fixedly coupled to the baclciron and extending radially inward from the backiron; a plurality of electromagnetic coils each being wound around one of the baclciron and a respective one of the poles; a magnetic target ring fixedly coupled to the shaft and separated from the poles by a plurality of airgaps; and a control system comprising: a source of electrical power electrically coupled to the electromagnetic coil; a microprocessor electrically coupled to the source of electrical power; a memory-storage device electrically coupled to the microprocessor; and a set of computer-executable instructions stored on the memory-storage device, wherein the computer-executable instructions: control a first set of outputs from a control system of the magnetic bearing system based on a first set of inputs to the control system and using a first set of mathematical equations; detect the presence of a fault in one or more of the control inputs and the control outputs; modify the first set of mathematical equations to form a second set of mathematical equations; and control a second set of outputs from the control system based on second set of inputs to the control system and using the second set of mathematical equations, wherein the second set of outputs and the second set of inputs exclude the one or more of the control inputs and the control outputs having the fault therein.

32. A magnetic bearing for suspending a rotatable shaft, comprising: a backiron; a plurality of poles fixedly coupled to the baclciron and extending radially inward toward the rotatable shaft; a plurality of electromagnetic coils each being circumferentially disposed around one of the baclciron and a respective one of the poles; a magnetic target ring fixedly coupled to the rotor shaft, each of the plurality of poles being separated from the magnetic target ring by an airgap; and one or more electrical circuits electrically coupled to the plurality of electromagnetic coils and being adapted to energize the plurality of electromagnetic coils so that magnetic flux is induced in the airgaps, wherein the one or more electrical circuits comprise an external coil.

33. The magnetic bearing of claim 32, wherein the external coil is spaced apart from the baclciron and the plurality of poles.

34. The magnetic bearing of claim 32, wherein the one or more electrical circuits are each independently driven.

35. A magnetic bearing system for suspending a rotatable shaft, comprising: a first baclciron; a first plurality of substantially co-planar poles extending radially inward from the backiron; a first plurality of elecfromagnetic coils each being wound around one of the backiron and a respective one of the first plurality of poles; a second backiron axially spaced from the first baclciron; a second plurality of substantially co-planar poles extending radially inward from the second backiron and being axially spaced from the first plurality of substantially co-planar poles; a second plurality of coils each being wound around one of the baclciron and a respective one of the second plurality of substantially co-planar poles, each of the second plurality of electromagnetic coils having an angular position substantially equal to an angular position of a respective one of the first plurality of electromagnetic coils; and a control system electrically coupled to the first and the second pluralities of electromagnetic coils and being adapted to energize the first and the second pluralities of electromagnetic coils so that substantially identical electrical currents are sent to the electromagnetic coils having substantially identical angular positions, whereby the magnetic bearing generates an axially-polarized magnetic flux.

36. A method of operating a magnetic bearing to produce axially- polarized magnetic flux using a first plurality of substantially co-planar, radially-oriented electromagnetic coils axially spaced from a second plurality of substantially co-planar, radially-oriented elecfromagnetic coils, comprising: aligning each of the first plurality of radially-oriented electromagnetic coils with a respective one of the second plurality of radially-oriented electromagnetic coils so that each of the first plurality of radially-oriented electromagnetic coils has an angular position substantially equal to an angular position of the respective one of the second plurality of radially-oriented electromagnetic coils; and directing substantially identical electrical currents to the electromagnetic coils having substantially identical angular positions.

37. A magnetic bearing for suspending a rotatable shaft, comprising: a baclciron positioned around the rotatable shaft; a plurality of poles fixedly coupled to the backiron; a plurality of electromagnetic coils each being wound around one of the baclciron and a respective one of the poles, the plurality of electromagnetic coils being adapted to generate magnetic flux in response to electrical currents sent thereto; a magnetic target ring fixedly coupled to the shaft; and a control system electrically coupled to the plurality of electromagnetic coils, wherein the control system is adapted to energize the plurality of electromagnetic coils so that the magnetic flux varies in response to predicted changes in operating conditions of the magnetic bearing.

38. The magnetic bearing of claim 37, wherein the control system comprises: a source of electrical power electrically coupled to the plurality of elecfromagnetic coils; a microprocessor electrically coupled to the source of electrical power; a memory-storage device electrically coupled to the microprocessor; and a set of computer-executable instructions stored on the memory-storage device.

39. The magnetic bearing of claim 37, wherein the confrol system is adapted to energize the plurality of electromagnetic coils so that an axial bias component of the magnetic flux varies in response to the predicted changes in the operating conditions of the magnetic bearing.

40. The magnetic bearing of claim 37, wherein the control system is adapted to energize the plurality of electromagnetic coils so that a radial bias component of the magnetic flux varies in response to the predicted changes in the operating conditions of the magnetic bearing.

41. A magnetic bearing for suspending a rotatable shaft, comprising: a baclciron; a plurality of poles fixedly coupled to the baclciron and extending radially inward from the backiron; a plurality of electromagnetic coils each being wound around one of the baclciron and a respective one of the poles; a magnetic target ring fixedly coupled to the shaft; and a control system comprising: a source of electrical power electrically coupled to the plurality of electromagnetic coils; a microprocessor electrically coupled to the source of electrical power; a memory-storage device electrically coupled to the microprocessor; and a set of computer-executable instructions stored on the memory-storage device, wherein the computer-executable instructions cause the control system to energize the electromagnetic coils so that the magnetic bearing exerts a first predetermined level force on the rotatable shaft under a first set of operating conditions, and the magnetic bearing exerts a second predetermined level of force on the rotatable shaft under a second set of operating conditions.

42. A method of minimizing power consumption in a magnetic bearing that supports a rotatable shaft, comprising: energizing elecfromagnetic coils of the magnetic bearings with a first set of electrical currents so that the magnetic bearing generates a first level of force required to suspend the rotatable shaft under a first set of operating conditions; predicting changes in the operating conditions of the magnetic bearing using predetermined information; determining a second level of force required to suspend the rotatable shaft after the predicted changes in the operating conditions of the magnetic bearing; and energizing the electromagnetic coils with a second set of electrical currents so that the magnetic bearing exerts the second level of force on the rotatable shaft.