This application is related to U.S. patent application Ser. No. 15/040,795 filed on Feb. 10, 2016, the contents of which are expressly incorporated herein by reference.
The present disclosure relates generally to switched reluctance machines, and more particularly, to flux estimation systems and methods implemented for controlling switched reluctance machines.
An electric machine such as an electrical motor, power generation system, genset, or the like, is generally used to convert one form of energy into another and may operate in a motoring mode to convert electrical input into rotational or otherwise mechanical output, or operate in a generating mode to convert rotational or otherwise mechanical input into electrical output. Among the various types of machines available for use with an electric drive, switched reluctance (SR) machines have received great interest for being robust and cost-effective. While currently existing systems and methods for controlling such electric machines may provide adequate control, there is still room for improvement.
Among other factors, proper determination of the position and speed of the rotor of the SR machine during relatively low speed operations may have significant impacts on overall performance and efficiency. Some conventional control schemes rely on mechanically aligned speed wheels and sensors to detect and determine the position of the rotor relative to the stator at machine standstill or low speed operations. However, such sensor-based control schemes typically require costly implementations and are susceptible to error. For instance, an error of 2 degrees in the detected mechanical rotor position of an SR machine, caused by a skewed sensor, a mechanical misalignment of the speed wheel, or the like, may correspond to a 0.5% decrease in efficiency of the electric drive assembly at full load.
Although sensorless solutions also exist, conventional sensorless control schemes must implement two or more distinct processes for different ranges of operating speeds or operating modes. For instance, a conventional control scheme for low speed operations, such as that of U.S. Pat. No. 5,525,886 to Lyons, et al., may inject current signals and refer to lookup maps to estimate the rotor position, while a conventional control scheme for high speed operations may apply observers to phase currents to emulate and determine the rotor position. Such a need to simultaneously operate between distinct processes depending on the speed or mode of operation can be inefficient, cumbersome and unnecessarily waste computational resources.
In addition, although the lookup tables or maps used during low speed processes can quickly output the rotor position based on injected current signals, the accuracy of the rotor position at the output is only as good as the quality of the current signal that is read at the input. More specifically, because lookup tables or maps are not capable of sufficiently filtering out noise or distinguishing errors induced by noise from the targeted signal, the rotor position ultimately output can be based on noise-induced errors and thus susceptible to inaccuracies. Such noise may, for example, manifest as current noise, which may lead to flux error and/or inaccurate flux estimation. Furthermore, while conventional systems typically derive rotor speed based on the rotor position, derivations or calculations based on noisy rotor position information can further compound noise-induced errors, output even noisier rotor speed information, and adversely impact the overall performance of the associated SR machine.
Accordingly, there is a need to provide control schemes for controlling SR machines that are not only less costly and easier to implement, but also more efficiently performed without compromising overall reliability. Moreover, there is a need to provide a control system that accurately predicts and/or corrects estimated flux values so that control systems can operate across wider ranges of operating speeds or operating modes of an SR machine and consume less of the computational resources allocated for use with the SR machine. There is also a need to provide a solution that is more reliable and robust to error, specifically errors caused by inefficient or ineffective flux estimation at multiple phases of an SR machine. The systems and methods disclosed herein are directed at addressing one or more of the aforementioned needs.
In one aspect of the present disclosure, a control system for a multi-phase switched reluctance (SR) machine having a rotor and a stator is provided. The stator may include, at least, a first phase winding and a second phase winding, the first phase winding corresponding with a first phase of the multi-phase SR machine and the second phase winding corresponding with a second phase of the multi-phase SR machine. The control system may include a converter circuit in electrical communication between the stator and a common bus, and a controller configured to monitor a bus voltage of the converter circuit and a phase current of the multi-phase SR machine. The controller may include, at least, a phase voltage estimator module configured to determine, at least, a first phase voltage associated with the first phase and a second phase voltage associated with the second phase, each of the first and second phase voltages based on one or more pulses, a flux estimator module configured to determine a first estimated flux, the first estimated flux associated with the first phase and based on the first phase voltage and an associated first mutual voltage, and a second estimated flux, the second estimated flux associated with the second phase and based on the second phase voltage and an associated second mutual voltage, a position observer module configured to determine a rotor position based at least partially on the first estimated flux, the second estimated flux, and a main pulse control module configured to control the SR machine based on the rotor position and a desired torque.
In another aspect of the present disclosure, an electric drive is provided. The electric drive may include an SR machine having a stator and a rotor rotatably disposed relative to the stator, the stator including, at least, a first phase winding and a second phase winding, the first phase winding corresponding with a first phase of the SR machine and the second phase winding corresponding with a second phase of the SR machine. The electric drive may further include a converter circuit configured to electrically communicate with the stator and a common bus, and a controller in electrical communication with at least the converter circuit. The controller may be configured to monitor a bus voltage of the common bus and a phase current of the SR machine, generate main pulses and any diagnostic pulses, determine a phase voltage based on one of the main pulses and the diagnostic pulses, determine a mutual voltage for the first phase, the mutual voltage representative of coupling effects of, at least, the second phase, determine a decoupled estimated flux based on the phase voltage and the mutual voltage, engage a position observer to determine a rotor position based at least partially on the decoupled estimated flux, and control the SR machine based on the rotor position and a desired torque.
In yet another aspect of the present disclosure, a method for determining rotor position of an SR machine being operated through a converter circuit is provided. The SR machine may include a stator and a rotor rotatably disposed relative to the stator, the stator including, at least, a first phase winding and a second phase winding, the first phase winding corresponding with a first phase of the SR machine and the second phase winding corresponding with a second phase of the SR machine. The method may include monitoring a bus voltage of the converter circuit and a phase current of the first phase of the SR machine, generating main pulses and any diagnostic pulses, determining a phase voltage for the first phase based on one of the main pulses and the diagnostic pulses, determining a mutual voltage for the first phase, the mutual voltage representative of coupling effects of, at least, the second phase, determining an estimated flux based on the phase voltage and the mutual voltage, engaging a position observer to determine a rotor position of the SR machine based at least partially on the decoupled estimated flux, and controlling an output torque of the SR machine based on the rotor position and a desired torque.
Referring to
Mechanical energy that is supplied by the primary power source 102 may be converted into electrical power by the electric drive 100 for use by the connected electrical loads 104. Conversely, any electrical power that may be supplied by the electrical loads 104 and/or the electric drive 100 may be supplied to drive mechanical power to the primary power source 102. As shown in the particular embodiment of
As shown in
During a generating mode of operation, as the rotor 110 of the SR machine 106 is rotated within the stator 112 by the primary power source 102, electrical current may be induced within the stator 112 and supplied to the converter circuit 116. The converter circuit 116 may in turn convert the electrical signals into the appropriate direct current (DC) voltage for distribution to the electrical load 104 and/or any other device via the common bus 114. The common bus 114 may provide terminals 118, such as positive and negative or ground lines, across which the common bus 114 may communicate a bus voltage or DC link voltage between one or more electrically parallel devices of the electric drive 100. The electrical loads 104 may include circuitry for converting the DC voltage supplied by the converter circuit 116 into the appropriate electrical signals for operating any one or more devices associated with the electric drive 100. Additionally, during a motoring mode of operation, or when the electrical loads 104 become the source of electrical power, the SR machine 106 may be enabled to cause rotation of the rotor 110 in response to electrical signals that are provided to the stator 112, and its associated phase windings 111, from the common bus 114.
As shown in
As illustrated in
The controller 128 of
As shown in
While the main pulse control module 136 may be suited for use with high speed operating modes or relatively high operating speeds of the SR machine 106, low speed operating modes or relatively low operating speeds of the SR machine 106 may be managed by a diagnostic pulse control module 138 as shown in
Still referring to
As demonstrated by the architecture of the controller 128 in
To determine rotor position or rotor speed, the controller 128 of
For example,
In some examples, the controller 128 may provide a stator resistance module 141. The stator resistance module 141 may be provided to estimate a voltage drop due to stator resistance, within the SR machine 106 based on phase current and an estimation of stator resistance. An example of the stator resistance module 141 is depicted schematically in
The controller 128 may further include a mutual voltage estimator module 142 configured to determine the associated mutual voltage, for instance, with reference to one or more preprogrammed lookup tables, maps, or the like, which predefine relationships between mutual flux values, phase current values, estimated rotor position values, and the like. Mutual voltage (VmutualA) for a given phase of the SR machine 106 is a voltage at a given phase that is caused by inductance and/or coupling effects of other phases.
In reference to the example mutual voltage estimator module 142 depicted in
Similarly, to estimate the flux generated by the coupling effect of phase C on phase A (FluxConA), the mutual voltage estimator module 142 may receive the measured phase current for phase C (ImeasC) and an estimated rotor position of the rotor 110 relative to phase C (ΘC), which is provided by, for example, output of the position observer module 144 of the controller 128. Using ImeasC and ΘC as input to a FluxConA determiner 166, the mutual voltage estimator module 142 may be estimated by utilizing one or more preprogrammed lookup tables, maps, or the like, which predefine relationships between mutual flux values, phase current values, estimated rotor position values, and the like. For example, ΘC may be referenced against a one-dimensional look up table and then output of the one dimensional look up table may be utilized to determine FluxConA. By utilizing a one dimensional look up table for determining one or both of FluxBonA and FluxConA, a computational and/or cost effective implementation of the mutual voltage estimator module 142, in comparison to implementations using multi-dimensional look up tables, may be achieved.
Based on FluxBonA, FluxConA, and an estimated flux for phase A from a prior time step, at a given sample rate, wherein “n” is the current sample time (FluxEstA(n-1)), the mutual flux for phase A (FluxMutualA) may be determined. In the non-limiting embodiment of
Furthermore, the controller 128 may apply the phase voltage and the mutual voltage, and any suitable calculation, computation, derivation and/or manipulation thereof, as inputs to a position observer module 144 to determine rotor position and to a speed observer module 146 to determine rotor speed as shown in
While other manipulations or derivations based on the phase voltage and the mutual voltage will be apparent to those of skill in the relevant art, the controller 128 of
In the non-limiting embodiment of the flux estimator module 148 of
As such, in the example of
In some examples, the flux estimator module 148 may further include a flux integrator 174, which may be useful in accurately and continuously predicting FluxEstA with minimal error. In one example, FluxEstA may be a value which alters over the course of time in accordance with a discrete time sample at a sample rate of “k,” wherein the current time step is “n;” therefore the flux integrator 174 may utilize a time integration algorithm for continuously determining FluxEstA. In some such examples, the flux integrator 174 can be implemented by a time step 176 and a step delay 178. Of course, other algorithms, methods, or techniques known in the art for integrating the estimated flux are certainly possible.
Additionally, FluxEstA may be automatically reset by the flux estimator module 148 by utilizing the flux resetting module 180. The flux resetting module 180 may be configured to reset FluxEstA during certain conditions. For example, the flux resetting module 180 may continuously read, as input, FluxEstA and compare FluxEstA to one or both of an upper flux limit 182 and a lower flux limit 184. Both the upper flux limit 182 and the lower flux limit 184 may be continuously determined limits based on, at least, the phase current for phase A (ImeasA). In some examples, if FluxEstA equals or exceeds the upper flux limit 182, then FluxEstaA resets to the value of the upper flux limit 182 at which FluxEstA exceeded the upper flux limit 182. Additionally or alternatively, if FluxEstA is less than or equal to the lower flux limit 184 at a given time, then FluxEstA resets to the lower flux limit 184 at the point in time wherein the value was less than or equal to the lower flux limit 184. In either scenario, the flux may be constantly limited and/or reset, based on one or both of the upper flux limit 182 and the lower flux limit 184, at a flux limiter 186.
The flux resetting module 180 may be particularly useful in more accurately predicting flux by clearing error in flux calculations when the estimated flux is reset. For example, in standard operations of the SR machine 106, the phase current and flux of each phase return to zero. However, in continuous conduction mode, all phases have a phase current and flux that does not ever return to zero and, thus, flux estimation error may build within the flux integrator 174. Therefore, if the flux resetting module 180 resets the flux to the upper and lower limits twice per cycle during a continuous conduction mode, then the flux error may be cleared at each reset.
As shown, the controller 128 may further employ a current estimator module 150 which determines an estimated phase current based on the estimated flux, and a current error synthesis module 152 which determines the error between the estimated phase current and one or more phase currents of the SR machine 106. The current error may then be fed into each of the position observer module 144 and the speed observer module 146 to determine the rotor position and the rotor speed, respectively.
The position observer module 144 of
Similar to the position observer module 144, the speed observer module 146 may employ a state observer system to at least partially emulate the internal state of a real SR machine 106, receive current error as input, and generate rotor speed as output. Additionally, although the speed observer module 146 may be configured to determine rotor speed based on current error, the speed observer module 146 may be modified to employ other inputs, such as the phase voltage, mutual voltage, estimated flux, phase current, or any other suitable parameter adapted by the controller 128 to assess rotor speed. In other modifications, the speed observer module 146 may be omitted entirely, and derivations of the rotor position with respect to time may be used to determine rotor speed. However, it will be understood that such indirect estimations of rotor speed may magnify any noise or other errors untreated by the position observer module 144. Furthermore, the controller 128 may optionally include a speed processing module 156 configured to process the output of the speed observer module 146 as needed to further refine and/or calibrate the estimated rotor speed.
In general, the foregoing disclosure finds utility in various applications relating to switched reluctance (SR) machines or any other suitable electric machine being employed as a motor and/or generator. In particular, the disclosed systems and methods may be used to provide more efficient and accurate flux estimation for control of SR machines that are typically employed in association with the electric drives of power generation machines, industrial work vehicles, and other types of machines commonly used in the art. The present disclosure may also be implemented with other variable-speed drives commonly used in association with industrial and consumer product applications. The present disclosure may further be used with integrated starters, generators, or the like, commonly associated with automotive, aerospace, and other comparable mobile applications.
One exemplary algorithm or controller-implemented method 200 for operating an SR machine 106 is diagrammatically provided in
Additionally, the controller 128, according to block 220, of
According to block 230, the controller 128 may determine a stator resistance using, for example, the stator resistance module 141. In some examples, the stator resistance may be based on temperature information associated with the SR machine 106. The stator resistance may be used to determine a stator resistance voltage for the SR machine 106 based on the modeled stator resistance of block 230 and the phase current, according to block 245.
A mutual voltage may be determined based on one or more of previously determined estimated flux, previously determined rotor positions, and phase currents, according to block 240. Such a mutual voltage determination may be performed in accordance with the modules and/or processes of the mutual voltage estimator module 142, discussed above, or any other techniques for determining mutual voltage known in the art.
The controller 128, such as via the flux estimator module 148 of
At block 250, the method 200 may determine if the flux exceeds any flux limits determined by, for example, the flux upper limit 182 and the flux lower limit 184 of the flux resetting module 180. If block 250 determines that the decoupled estimated flux requires resetting, then the decoupled estimated flux is reset in accordance with block 255. Otherwise, the method 200 continues to blocks 260 and/or 265. In some examples, at block 255, resetting the decoupled estimated flux is performed during the idling period of each cycle in a discontinuous conduction mode or a continuous conduction mode of the SR machine 106.
Still referring to
Furthermore, the controller 128 may be configured to engage a speed observer, such as with the speed observer module 146 in
Based on the foregoing, the present disclosure provides a simplified and yet robust solution for operating an SR machine across a much wider range of operating speeds. More particularly, the present disclosure provides a control architecture which streamlines the processes used for determining the rotor position of an SR machine to conserve computational resources and excess costs associated therewith. The present disclosure also employs independent position and speed observers which naturally filter and/or correct for noise-induced errors to provide for more reliable results. The present disclosure thereby provides a sensorless solution that eliminates the need for costly position or proximity sensors without compromising performance. It will be appreciated that while only certain embodiments have been set forth for the purposes of illustration, alternatives and modifications will be apparent from the above description to those skilled in the art. These and other alternatives are considered equivalents and within the spirit and scope of this disclosure and the appended claims.
Number | Name | Date | Kind |
---|---|---|---|
5072166 | Ehsani | Dec 1991 | A |
5701064 | Horst | Dec 1997 | A |
5864217 | Lyons | Jan 1999 | A |
5929590 | Tang | Jul 1999 | A |
5982117 | Taylor | Nov 1999 | A |
6288514 | Direnzo | Sep 2001 | B1 |
6366865 | Chalupa | Apr 2002 | B1 |
6456031 | Gallegos-Lopez | Sep 2002 | B1 |
6731083 | Marcinkiewicz | May 2004 | B2 |
6801012 | Islam | Oct 2004 | B1 |
6831439 | Won | Dec 2004 | B2 |
6853163 | Slater | Feb 2005 | B2 |
6989668 | Mayes | Jan 2006 | B2 |
8610389 | Lee | Dec 2013 | B2 |
8723462 | Jang | May 2014 | B2 |
8866433 | Gerdes | Oct 2014 | B2 |
20150084575 | Magee | Mar 2015 | A1 |