Acoustic Noise Reduction for Switched Reluctance Machines

Abstract

A current profile of an excitation current supplied to an electric coil of a switched reluctance machine is controlled to reduce acoustic noise. Multiple potential current waveforms may be evaluated to select a desired waveform that reduces the acoustic noise level of the switched reluctance machine. A cumulative sound pressure level of the switched reluctance machine may be determined for each potential current waveform. The cumulative sound pressure level may be determined based on multiple harmonic sound pressure levels expected to result from the potential current waveform. A desired current waveform may be identified as the potential current waveform associated with an optimal cumulative sound pressure level. The desired current waveform may then be applied to a corresponding phase coil of the switched reluctance machine to operate the switched reluctance machine while reducing acoustic noise without sacrificing other motor performance criteria.

Description

[Technical field]

[0001] This application claims priority to U.S. Provisional Patent Application No. 63 / 332,903, filed April 20, 2022, the entirety of which is incorporated herein by reference.

[0002] The described embodiments relate generally to switched reluctance machines, and more particularly to reducing acoustic noise in switched reluctance machines (SRM). [Background technology]

[0003] Electric machines have been applied as motors and generators in a wide range of industries for over a century. Electric motors are a large consumer of electrical energy, with approximately 46% of the world's electricity demand coming from electric motors (see, e.g., U.S. Department of Energy, Office of Advanced Manufacturing, Premium Efficiency Motor Selection and Application Guide - Handbook for Industry, February 2014 [online]. Available at: http: / / www.energy.gov / ). Motor-related electrical energy consumption is expected to increase in the coming decades due to, for example, the increasing market penetration of electric vehicles, the accelerating use of renewable energy systems, the increasing sales of household appliances in jurisdictions with a growing middle class, and the growing need for cooling equipment in regions facing significant increases in temperature due to climate change. There is a continuing demand for high-efficiency, high-performance, and low-cost electric motors for a variety of applications in the transportation (e.g., traction motors), industrial (e.g., pumps), commercial (e.g., compressors), residential (e.g., electric fans), and energy generation (e.g., wind turbines) sectors.

[0004] Rare earth magnets are widely used in a variety of motor applications. The manufacturing of permanent magnets represents the largest application of rare earth metals. The rotor magnet in a permanent magnet (PM) machine provides an independent source of magnetic flux. Interior permanent magnet machines, where the magnets are embedded in the rotor, are currently used in the majority of electric powertrain applications. Surface permanent magnet machines, where the magnets are located on the surface of the rotor, are increasingly being used in residential applications. However, price fluctuations and supply chain risks associated with rare earth metals can limit the availability of permanent magnet motors. Additionally, the extraction and processing of rare earth metals has had adverse environmental impacts. Challenges with the supply of rare earth metals can prevent the electric motor industry from meeting the growing demand for high efficiency, high performance, and low cost electric motors.

[0005] A reluctance machine is an electric machine in which torque is produced by the tendency of the moving parts of the machine to move to a position where the inductance of the energized windings is maximum. A switched reluctance machine (SRM) is a type of reluctance machine whose windings are energized depending on the position of the moving parts of the machine. SRMs have salient poles on both the rotor and the stator. Permanent magnets can be omitted in SRMs.

[0006] The simple, low-cost construction and robustness of SRMs make them suitable for many applications, including traction and automotive applications, however, SRMs can have high acoustic noise and vibration, which can prevent their use in noise-sensitive applications, such as propulsion units, electrical equipment, and heating and ventilation systems. Summary of the Invention [Means for solving the problem]

[0007] The following summary is intended to introduce the reader to various aspects of the detailed description, but it is not intended to define or limit the scope of the invention.

[0008] Systems, methods, and computer program products are provided for controlling a switched reluctance machine. In some instances, the switched reluctance machine may be controlled to reduce acoustic noise generated by the switched reluctance machine. The current waveform used to drive the switched reluctance machine can be controlled to reduce acoustic noise. This allows for more widespread use of switched reluctance machines in a wide variety of applications.

[0009] A switched reluctance machine operates using electric coils that are selectively energized to induce a magnetic flux and force a rotor of the switched reluctance machine to rotate. The operation of the switched reluctance machine can be controlled by controlling how current is supplied to the electric coils. Potential current waveforms can be evaluated for a given phase coil of the switched reluctance machine to select a waveform that reduces the acoustic noise level of the switched reluctance machine. A cumulative sound pressure level of the switched reluctance machine can be determined for each potential current waveform. The cumulative sound pressure level can be determined based on multiple harmonic sound pressure levels expected to result from the potential current waveform. A desired current waveform can be identified as the potential current waveform associated with an optimal cumulative sound pressure level. The desired current waveform can then be applied to a corresponding phase coil of the switched reluctance machine to operate the switched reluctance machine while reducing acoustic noise.

[0010] According to some aspects, a method for controlling a switched reluctance machine is provided that can include determining a plurality of potential phase current waveforms for corresponding phase coils of the switched reluctance machine, determining a cumulative sound pressure level of the switched reluctance machine for each potential phase current waveform, the cumulative sound pressure level being determined based on a plurality of harmonic sound pressure levels expected to result from the potential phase current waveform, identifying a desired phase current waveform as the potential phase current waveform associated with an optimal cumulative sound pressure level, and applying the desired phase current waveform to the corresponding phase coil of the switched reluctance machine.

[0011] The desired phase current waveform may be determined using an iterative optimization process, where each iteration of the iterative optimization process may include identifying an iteration-specific potential phase current waveform, determining a number of iteration-specific circumferential orders associated with the iteration-specific potential phase current waveform, determining a number of radial force harmonics magnitudes associated with the iteration-specific potential phase current waveform, determining a number of iteration-specific forcing frequencies associated with the iteration-specific potential phase current waveform, and determining a cumulative sound pressure level in the iteration-specific potential phase current waveform using the number of iteration-specific circumferential orders, the number of iteration-specific forcing frequencies, and the magnitudes of the number of radial force harmonics.

[0012] The cumulative sound pressure level may be determined by identifying a plurality of repeat-specific harmonic sound pressure levels, where each repeat-specific harmonic sound pressure level is determined based on a combination of one of the repeat-specific circumferential orders, one of the repeat-specific forcing frequencies, and one of the radial force harmonic magnitudes, and determining the cumulative sound pressure level as a sum of the plurality of repeat-specific harmonic sound pressure levels.

[0013] The specific harmonic sound pressure level for each iteration may be determined using a look-up table that is defined based on the simulated sound pressure levels associated with a particular circumferential order and a particular forcing frequency.

[0014] The look-up table may be defined during a pre-processing stage, which may include determining a number of dominant circumferential modes for the switched reluctance machine based on a geometry and a magnetic pole configuration of the switched reluctance machine; for each dominant circumferential mode, determining a natural frequency associated with that dominant circumferential mode and determining a simulated sound pressure level associated with that dominant circumferential mode by simulating a composite sound pressure level corresponding to at least one frequency value close to the natural frequency using a vibro-acoustic model of the switched reluctance machine; and storing the simulated sound pressure levels in association with the corresponding dominant circumferential mode and frequency value in the look-up table.

[0015] The multiple dominant circumferential modes may be determined by identifying a number of magnetic poles of the switched reluctance machine as a ratio between a number of stator poles of the switched reluctance machine and a number of phases of the switched reluctance machine, and identifying the dominant circumferential modes as a function of the number of magnetic poles.

[0016] The method may include, for each dominant circumferential mode, identifying higher circumferential order radial force harmonics that excite the dominant circumferential mode due to a sampling effect of radial force density harmonics, and determining a simulated sound pressure level associated with the dominant circumferential mode to include sound pressures resulting from the identified higher circumferential order radial force harmonics.

[0017] Determining the multiple iteration specific forcing frequencies may include determining a motor speed of the switched reluctance machine; determining multiple iteration specific time orders associated with the iteration specific potential phase current waveforms; and determining the multiple iteration specific forcing frequencies based on the motor speed and the multiple iteration specific time orders.

[0018] Determining the multiple iteration-specific time orders may include determining a radial force waveform associated with the iteration-specific potential phase current waveform, and calculating the multiple iteration-specific time orders by applying a Fast Fourier Transform to the radial force waveform.

[0019] The method may include determining a plurality of dominant time orders, determining a subset of dominant iteration-specific time orders corresponding to the plurality of dominant time orders, and determining a plurality of iteration-specific forcing frequencies based on the motor speed and the subset of dominant iteration-specific time orders.

[0020] Multiple dominant time orders can be determined based on the geometry and pole configuration of the switched reluctance machine.

[0021] Determining the plurality of dominant time orders may include identifying a number of strokes in one machine revolution as a product of a number of rotor poles of the switched reluctance machine and a number of phases of the switched reluctance machine, and identifying the dominant time order as a function of the number of strokes.

[0022] Determining the multiple iteration-specific circumferential orders may include determining a radial force waveform associated with the iteration-specific potential phase current waveform, and calculating the multiple iteration-specific circumferential orders by applying a Fast Fourier Transform to the radial force waveform.

[0023] Determining the magnitudes of the multiple radial force harmonics may include determining a radial force waveform associated with a repetition-specific potential phase current waveform and calculating the magnitudes of the multiple radial force harmonics by applying a Fast Fourier Transform to the radial force waveform.

[0024] The magnitudes of multiple repeat-specific circumferential orders and multiple radial force harmonics can be determined simultaneously by applying a Fast Fourier Transform to the radial force waveform.

[0025] The radial force waveform can be determined using a dynamic motor drive model based on the motor speed and DC link voltage of the switched reluctance machine.

[0026] The radial force waveform may be determined based on the motor speed, phase voltages, and phase currents of the switched reluctance machine.

[0027] According to some aspects, a system may include a switched reluctance machine comprising a shaft, a rotor attached to the shaft and having a plurality of rotor protruding poles, a stator having a plurality of stator protruding poles protruding from the stator towards the rotor poles, and a plurality of electric coils wound around the stator poles, the plurality of electric coils including a plurality of separate phase coils defining a plurality of phases of the switched reluctance machine; a power source; a power converter coupled to the power source and the switched reluctance machine; and a controller, the controller configured to: determine a plurality of potential phase current waveforms for corresponding phase coils of the switched reluctance machine; determine a cumulative sound pressure level of the switched reluctance machine for each potential phase current waveform, the cumulative sound pressure level being determined based on a plurality of harmonic sound pressure levels expected to result from the potential phase current waveforms; identify a desired phase current waveform as the potential phase current waveform associated with an optimal cumulative sound pressure level; and apply the desired phase current waveform to the corresponding phase coil of the switched reluctance machine using the power converter.

[0028] The controller may be configured to determine the desired phase current waveform using an iterative optimization process, where each iteration of the iterative optimization process may include identifying an iteration-specific potential phase current waveform, determining a number of iteration-specific circumferential orders associated with the iteration-specific potential phase current waveform, determining a number of radial force harmonics magnitudes associated with the iteration-specific potential phase current waveform, determining a number of iteration-specific forcing frequencies associated with the iteration-specific potential phase current waveform, and determining a cumulative sound pressure level in the iteration-specific potential phase current waveform using the number of iteration-specific circumferential orders, the number of iteration-specific forcing frequencies, and the magnitudes of the number of radial force harmonics.

[0029] The controller may be configured to determine the cumulative sound pressure level by identifying a plurality of iteration-specific harmonic sound pressure levels, each iteration-specific harmonic sound pressure level being determined based on a combination of one of the iteration-specific circumferential orders, one of the iteration-specific forcing frequencies, and one of the radial force harmonic magnitudes, and determining the cumulative sound pressure level as a sum of the plurality of iteration-specific harmonic sound pressure levels.

[0030] The system can include a non-transitory storage memory having stored therein a look-up table defined based on simulated sound pressure levels associated with particular circumferential orders and particular forcing frequencies, and the controller can be configured to determine each iteration-specific harmonic sound pressure level using the look-up table.

[0031] The look-up table may be defined during a pre-processing stage, which may include determining a plurality of dominant circumferential modes for the switched reluctance machine based on a geometry and a magnetic pole configuration of the switched reluctance machine; for each dominant circumferential mode, determining a natural frequency associated with the dominant circumferential mode; determining a simulated sound pressure level associated with the dominant circumferential mode by simulating a composite sound pressure level corresponding to at least one frequency value close to the natural frequency using a vibro-acoustic model of the switched reluctance machine; and storing the simulated sound pressure levels in association with the corresponding dominant circumferential mode and frequency value in the look-up table.

[0032] The multiple dominant circumferential modes may be determined by identifying a number of magnetic poles of the switched reluctance machine as a ratio between a number of stator poles of the switched reluctance machine and a number of phases of the switched reluctance machine, and identifying the dominant circumferential modes as a function of the number of magnetic poles.

[0033] For each dominant circumferential mode, a higher circumferential order radial force harmonic may be identified that excites the dominant circumferential mode due to a sampling effect of the radial force density harmonics, and a simulated sound pressure level associated with the dominant circumferential mode may be determined to include sound pressure resulting from the identified higher circumferential order radial force harmonic;

[0034] The controller may be configured to determine the multiple iteration-specific forcing frequencies by determining a motor speed of the switched reluctance machine, determining multiple iteration-specific time orders associated with the iteration-specific potential phase current waveforms, and determining the multiple iteration-specific forcing frequencies based on the motor speed and the multiple iteration-specific time orders.

[0035] The controller may be configured to determine the multiple iteration-specific time orders by determining a radial force waveform associated with the iteration-specific potential phase current waveform and calculating the multiple iteration-specific time orders by applying a fast Fourier transform to the radial force waveform.

[0036] The controller may be configured to determine a plurality of dominant time orders, determine a subset of dominant iteration-specific time orders corresponding to the plurality of dominant time orders, and determine a plurality of iteration-specific forcing frequencies based on the motor speed and the subset of dominant iteration-specific time orders.

[0037] The multiple dominant time orders may be predetermined based on the geometry and pole configuration of the switched reluctance machine.

[0038] The multiple dominant time orders may be predetermined by identifying a number of strokes in one machine revolution as a product of a number of rotor poles of the switched reluctance machine and a number of phases of the switched reluctance machine, and identifying the dominant time orders as a function of the number of strokes.

[0039] The controller may be configured to determine the multiple iteration-specific circumferential orders by determining a radial force waveform associated with the iteration-specific potential phase current waveform and calculating the multiple iteration-specific circumferential orders by applying a Fast Fourier Transform to the radial force waveform.

[0040] The controller may be configured to determine the magnitudes of the multiple radial force harmonics by determining a radial force waveform associated with a repetition-specific potential phase current waveform and calculating the magnitudes of the multiple radial force harmonics by applying a Fast Fourier Transform to the radial force waveform.

[0041] The controller may be configured to simultaneously determine the magnitudes of multiple repeat-specific circumferential orders and multiple radial force harmonics by applying a Fast Fourier Transform to the radial force waveform.

[0042] The controller may be configured to determine the radial force waveform using a dynamic motor drive model based on the motor speed and DC link voltage of the switched reluctance machine.

[0043] The controller may be configured to determine a radial force waveform based on a motor speed, a phase voltage, and a phase current of the switched reluctance machine.

[0044] According to some aspects, a non-transitory computer readable medium is provided that stores computer executable instructions that, when executed by a computer processor, cause the computer processor to perform a method of controlling a switched reluctance machine, the method including determining a plurality of potential phase current waveforms for corresponding phase coils of the switched reluctance machine, determining a cumulative sound pressure level of the switched reluctance machine for each potential phase current waveform, the cumulative sound pressure level being determined based on a plurality of harmonic sound pressure levels expected to result from the potential phase current waveform, identifying a desired phase current waveform as the potential phase current waveform associated with an optimal cumulative sound pressure level, and applying the desired phase current waveform to the corresponding phase coil of the switched reluctance machine.

[0045] The non-transitory computer readable medium can store computer executable instructions that, when executed by a computer processor, cause the computer processor to perform a method of controlling a switched reluctance machine, the method being described herein.

[0046] Other features and advantages of the present application will become apparent from the following detailed description taken in conjunction with the accompanying drawings. It should be understood, however, that the detailed description and specific examples, while indicating preferred embodiments of the present application, are given by way of illustration only, since various changes and modifications within the spirit and scope of the present application will become apparent to those skilled in the art from this detailed description.

[0047] For a better understanding of the various embodiments described herein, and to show more clearly how these may be put into practice, reference is made to the accompanying drawings, in which at least one illustrative embodiment is shown, by way of illustration, and which are briefly described herein. [Brief description of the drawings]

[0048] [Figure 1A] 1 is a block diagram of an exemplary system for controlling a switched reluctance machine. [Figure 1B] FIG. 2 is a cross-sectional view of an exemplary switched reluctance machine that can be used with the system shown in FIG. [Diagram 2] 1 is a flowchart of an exemplary method for controlling a switched reluctance machine. [Diagram 3] 3 is a flow chart of an exemplary method for determining a cumulative sound pressure level that can be used in the method shown in FIG. 2. [Figure 4] 4 is an exemplary flowchart of data flow between sub-processes that can be used in the methods shown in FIGS. 2 and 3. [Diagram 5] FIG. 13 illustrates a plot of natural frequencies identified for vibration modes 0 and 6 for an exemplary switched reluctance machine with 18 stator poles and 12 rotor poles. [Figure 6] FIG. 1 illustrates an example plot of surface displacement from vibration mode 0 for an example switched reluctance machine with 18 stator poles and 12 rotor poles. [Figure 7] FIG. 13 illustrates an example plot of surface displacement from vibration mode 6 plotted for an example switched reluctance machine with 18 stator poles and 12 rotor poles. [Figure 8] FIG. 13 illustrates an example plot of excitation of lower circumferential order vibration modes caused by higher circumferential order radial force harmonics due to sampling effects for an example switched reluctance machine with 18 stator poles. [Figure 9] FIG. 13 shows a plot of sound pressure levels from vibration mode 0 at different frequencies for an exemplary switched reluctance machine with 18 stator poles and 12 rotor poles including excitations induced by higher circumferential modes due to sampling effects. [Figure 10] FIG. 13 illustrates an exemplary plot of sound pressure levels for a simulated switched reluctance machine when different force magnitudes are applied. [Figure 11] FIG. 13 illustrates an example plot of sound pressure level of vibration mode 0 over a frequency range for an example switched reluctance machine with 18 stator poles and 12 rotor poles when unit force is applied. [Figure 12] FIG. 13 illustrates an example plot of acoustic emissivity of vibration mode 0 over a frequency range for an example switched reluctance machine with 18 stator poles and 12 rotor poles. [Figure 13] FIG. 13 illustrates an example plot of sound pressure levels of vibration mode 6 over a frequency range for an example switched reluctance machine with 18 stator poles and 12 rotor poles when unit force is applied. [Figure 14] FIG. 13 illustrates an example plot of acoustic emissivity of vibration mode 6 over a frequency range for an example switched reluctance machine with 18 stator poles and 12 rotor poles. [Figure 15] FIG. 13 illustrates an example plot of a radial force density waveform determined for one electrical cycle of a switched reluctance machine using an example dynamic motor drive model. [Figure 16] FIG. 16 illustrates an example plot of a Fast Fourier Transform decomposition of the radial force density waveform shown in FIG. 15 for an example switched reluctance machine with 18 stator poles and 12 rotor poles. [Figure 17]FIG. 17 illustrates an example plot of radial force density harmonics from FIG. 16 in the first quadrant of the uv-plane for an example switched reluctance machine with 18 stator poles and 12 rotor poles. [Figure 18] FIG. 10 illustrates an example plot of dynamic current values ​​of a switched reluctance machine and simulated sound pressure levels determined according to the methods described herein compared to sound pressure levels determined based on finite element analysis simulations. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0049] As will be appreciated by those skilled in the art, the drawings described below are for illustrative purposes only. The drawings are not intended to limit the scope of the applicant's teachings in any way. It will also be appreciated that for simplicity and clarity of illustration, elements shown in the figures have not necessarily been drawn to scale. Further, where considered appropriate, reference numerals may be repeated among the figures to indicate corresponding or analogous elements.

[0050] Unless otherwise indicated, the definitions and examples set forth in this section and other sections are intended to be applicable to all applicable embodiments and aspects of the application described herein, as would be understood by one of ordinary skill in the art.

[0051] In understanding the scope of the present application, the term "comprising" and its derivatives as used herein are intended to be open-ended terms specifying the presence of stated features, elements, components, groups, integers, and / or steps, but not excluding the presence of other unstated features, elements, components, groups, integers, and / or steps. The same also applies to words of similar meaning, such as the terms "including," "having," and their derivatives. The term "consisting of" as used herein and its derivatives are intended to be closed terms specifying the presence of stated features, elements, components, groups, integers, and / or steps, but excluding the presence of other unstated features, elements, components, groups, integers, and / or steps. The term "consisting essentially of" as used herein is intended to specify the presence of stated features, elements, components, groups, integers, and / or steps, as well as the presence of things that do not substantially affect the basic and novel properties of the features, elements, components, groups, integers, and / or steps.

[0052] As used herein, terms of degree such as "substantially," "about," and "approximately" refer to a reasonable amount of deviation from the modified term so as not to materially alter the end result. These terms of degree should be interpreted as including a deviation of at least 5% from the modified term if this deviation does not negate the meaning of the word it modifies.

[0053] As used in this application, the singular forms "a," "an," and "the" include plural references unless the content clearly dictates otherwise.

[0054] In embodiments with "additional" or "second" components, a second component, as used herein, is physically distinct from the other components or from the first component. A "third" component is distinct from the other first and second components, and further listed or "additional" components are similarly distinct.

[0055] As used herein, the term "and / or" means that the listed items may be present or used either individually or in any combination. In effect, the term means that "at least one of" or "one or more" of the listed items are used or present.

[0056] The systems, methods, and devices described herein may be implemented as a combination of hardware or software. In some cases, the systems, methods, and devices described herein may be implemented, at least in part, by using one or more computer programs executing on one or more programmable devices including at least one processing element and data storage element (including volatile and non-volatile memory and / or storage elements). Depending on the nature of the device, these devices may have at least one input device (e.g., push button keyboard, mouse, touch screen, etc.) and at least one output device (e.g., display screen, printer, wireless radio, etc.).

[0057] Some elements used to implement at least some of the systems, methods, and devices described herein may be implemented via software written in a high-level procedural language, such as object-oriented programming. Thus, the program code may be written in any suitable programming language, such as Python or C. Alternatively or additionally, some of these elements implemented via software may be written in assembly language, machine language, or firmware, as appropriate. In either case, the language may be a compiled or interpreted language.

[0058] At least some of these software programs may be stored in a storage medium (e.g., but not limited to, a computer readable medium such as a ROM, a magnetic disk, an optical disk, etc.) or in a device readable by a general-purpose or special-purpose programmable device, the software program code, when read by the programmable device, configures the programmable device to operate in a new specific, predefined manner to perform at least one of the methods described herein.

[0059] Additionally, at least some of the programs associated with the systems and methods described herein may be capable of being distributed in a computer program product that includes a computer-readable medium that holds computer usable instructions for one or more processors. The medium may be provided in a variety of forms, including, but not limited to, one or more diskettes, compact discs, tapes, chips, and non-transitory forms such as magnetic and electronic storage devices. Alternatively, the medium may be transitory in nature, such as, but not limited to, wired transmissions, satellite transmissions, Internet transmissions (e.g., downloads), media, digital signals, and analog signals. The computer usable instructions may be in a variety of forms, including compiled and non-compiled code.

[0060] Switched reluctance machines have a simple, low-cost, and robust construction. SRMs can operate without permanent magnets, which have significant supply chain issues. Expanding the utility of switched reluctance machines can contribute to low-cost, reliable, high-performance machines for a variety of applications without the supply chain risks inherent to machines that require permanent magnets.

[0061] Switched reluctance machines can suffer from problems related to acoustic noise and vibration. The large radial forces resulting from the protruding pole structure of switched reluctance machines can deform the stator core and frame, causing vibration and acoustic noise. This can be particularly problematic in noise-sensitive applications such as propulsion equipment, home appliances, and heating and ventilation systems.

[0062] The acoustic noise performance of a switched reluctance machine is related to both the electrical and mechanical aspects of the SRM. This presents an additional challenge in reducing the acoustic noise during operation of the SRM compared to controlling other performance parameters of the SRM such as torque and torque ripple, which are more easily modeled.

[0063] Acoustic noise in an SRM is primarily generated when radial forces excite the principal vibration modes of the motor structure. The radial forces in an SRM vary at different positions on the rotor as it rotates. The radial forces also vary at different angular positions of the air gap. Vibrations and acoustic noise are generated when the temporal and spatial harmonics of the radial forces interact with the vibration mode shapes of the motor structure.

[0064] Radial force density, which is the radial force per unit surface area, is a function of radial and tangential magnetic flux densities. The magnetic flux density in a switched reluctance machine is related to various parameters such as the pole configuration, motor geometry, and current control. The magnitude and forcing frequency of the radial force density harmonics, modal characteristics, motor emissivity, and properties of the acoustic medium contribute to the vibration and acoustic noise levels.

[0065] The present application provides systems and methods that can reduce acoustic noise and / or vibration in a switched reluctance machine by controlling the current supplied to the switched reluctance machine. The current can be controlled to directly reduce the sound pressure level of the switched reluctance machine. In particular, the current waveforms applied to the phase coils of the switched reluctance machine can be determined to reduce the vibration and / or acoustic noise of the switched reluctance machine.

[0066] The geometry and pole configuration of the switched reluctance machine can be used to determine the sound pressure levels resulting from radial force harmonics of a given frequency and circumferential order when the switched reluctance machine is operating. Predicted sound pressure levels for various combinations of frequency and circumferential order can be determined in a pre-processing stage by simulating the sound pressure levels. These predicted sound pressure levels can then be used to control the current waveform during real-time operation of the switched reluctance machine. For example, for a given SRM geometry and pole configuration, predicted sound pressure levels can be defined as a function of frequency and circumferential order. These predicted sound pressure levels can be stored (e.g., in a look-up table) for quick and easy access during operation of the SRM.

[0067] Predicted sound pressure levels can be determined for the dominant temporal and circumferential modes of the radial force harmonics of the SRM, which allows for the determination of a preferred current waveform based on the radial force harmonics that contribute most to the acoustic noise of the SRM.

[0068] The predicted sound pressure level can also be determined to include acoustic noise resulting from higher order modes that can excite the dominant modes due to sampling effects, thereby further reducing the acoustic noise generated by the SRM.

[0069] The current waveforms can be prescribed for the complete electrical cycle of the SRM. For example, a dynamic model of acoustic noise generation in a switched reluctance motor drive can be prescribed. This model can be used to determine the currents for the complete electrical cycle.

[0070] The current waveforms can be determined in real time using models of the vibration and acoustic noise generated by the switched reluctance machine. The current waveforms can be determined in the absence of vibration sensors in the SRM. This simplifies the construction of the SRM and ensures that all the most relevant time and azimuth modes are taken into account when determining the current waveforms.

[0071] The current waveform can be determined to directly reduce the sound pressure level, which can include considering the radiation efficiency at various frequencies and azimuthal modes, ensuring that the waveform is selected to reduce the acoustic noise that would be radiated from the SRM.

[0072] The sound pressure levels of the SRM can be predicted based on various characteristics of the SRM. For example, the acoustic noise generation of the SRM can be modeled based on characteristics such as the magnetic pole configuration, motor geometry, operating speed, load requirements, dominant temporal and circumferential modes, sampling effects, and emissivity. The modeled acoustic noise generation can be used to determine predicted sound pressure levels of the SRM under various operating conditions. The predicted sound pressure levels can then be stored for later access and retrieval. For example, the predicted sound pressure levels can be stored in one or more look-up tables that are categorized or indexed based on the specified operating conditions of the SRM.

[0073] The predicted sound pressure level can be used when determining a current waveform to be applied to the electrical coil of the SRM during operation. For example, the predicted sound pressure levels of different potential current waveforms can be compared. A preferred current waveform can then be selected based on the comparison of the predicted sound pressure levels. For example, the preferred current waveform can be selected as the potential current waveform that provides an optimal cumulative sound pressure level for the SRM. The optimal cumulative sound pressure level can be determined using a variety of different optimization parameters (e.g., performance metrics). For example, the optimal cumulative sound pressure level can be determined as the minimum sound pressure level achievable along with other specified performance objectives and / or constraints such as average torque, torque ripple, and efficiency.

[0074] 1A and 1B simultaneously, FIGURE 1A illustrates a block diagram of an exemplary system 100 for controlling a switched reluctance machine (SRM) 102. FIGURE 1B illustrates an example of a switched reluctance machine 102 that may be used in system 100.

[0075] 1A , the system 100 includes a switched reluctance machine 102, a power source 104, a power converter 106, and a controller 108. The system 100 may also include a memory 114 coupled to the controller 108. Optionally, the system 100 may also include a remote processing device 112.

[0076] 1A, the controller 108 may operate to drive the switched reluctance machine 102, for example, by controlling current supplied to phase coils of the switched reluctance machine 102 by the power supply 104 and the converter 106. The controller 108 may be any suitable processor, controller, or digital signal processor capable of providing sufficient processing power depending on the configuration, purpose, and requirements of the reluctance machine as known to those skilled in the art. Similarly, the processor may be coupled to volatile and non-volatile memory (e.g., memory 114) necessary for the processes performed by the reluctance machine 102.

[0077] As shown in FIG. 1B, the SRM 102 is an example of a switched reluctance machine with eight (8) stator poles and six (6) rotor poles. The SRM 102 includes salient poles on the rotor and stator. The salient pole configuration on both the rotor and stator results in a relatively simple manufacturing process and robust operation of the SRM. The example SRM 102 is shown with concentrated coil windings.

[0078] As shown, the switched reluctance machine 102 includes a stator 105 and a rotor 120. The rotor 120 is attached to a rotatable shaft 115. In the illustrated example, the shaft 115 extends axially, and the rotor 120 and the stator 105 also extend axially. As will be apparent, the present disclosure is applicable to other configurations of switched reluctance machines including, for example, radial flux machines, axial flux machines, internal rotor machines, external rotor machines, etc.

[0079] As shown, the stator 105 and rotor 120 may be concentrically disposed with respect to each other and to the shaft 115. In the SRM 102, the rotor 120 is positioned radially inward of the stator 105. Alternatively, the rotor may be positioned radially outward from the stator.

[0080] Both the stator 105 and the rotor 120 include a plurality of protrusions that define protruding teeth or protruding magnetic poles. The stator 105 has a stator core that includes a plurality of stator teeth 110, in this case eight stator teeth 110. The rotor 120 includes a plurality of rotor teeth 125, in this case six rotor teeth 125. The stator teeth 110 protrude radially from the stator core towards the rotor 120. Similarly, the rotor teeth 125 protrude radially from the rotor 120 towards the stator 105.

[0081] An air gap 135 is also provided between the rotor poles 125 and the stator poles 110. The shaft 115 may be positioned within a central bore of the machine 102. As previously mentioned, the SRM 102 is an example of an eight stator pole and six rotor pole switched reluctance machine with concentrated coil windings. That is, the stator 105 has a coil winding 130 around each stator pole 110.

[0082] The coils 130 are wound around each stator pole 110 and connected together to create a phase winding for each phase. The coils 130 may include multiple separate phase coils that define multiple phases of the switched reluctance machine 102.

[0083] SRMs can be designed with various numbers of stator and rotor poles and various numbers of phases. In general, SRMs typically do not include an excitation source on the rotor 120. As shown, the SRM 102 omits any excitation source (e.g., permanent magnets) except for the phase coils 130.

[0084] In an SRM, such as SRM 102, different phase windings 130 are energized depending on the position of the rotor 120. Thus, the rotor position also affects the air gap 135 and phase inductances of the SRM 102.

[0085] In an SRM, such as the SRM 102, reluctance torque is the only torque generating mechanism. When a phase winding on the stator pole 110 of the SRM 102 is energized with current, the energized pole attracts and aligns with the nearest rotor pole 125, minimizing the reluctance of the phase. To create motoring torque, the phase is energized with an ascending slope in the inductance profile.

[0086] As shown in SRM 102, a concentrated winding scheme is used and strategically manipulated to maintain motoring torque. In particular, current is sequentially applied to different phases of SRM 102 to maintain motoring torque. The current applied to the different phases of SRM 102 can include a variety of different profiles including, for example, a pulsating square current profile.

[0087] The power converter 106 may control the current in each phase of the SRM 102. For example, the power converter 106 may include an asymmetric bridge converter operable to control the current in each phase. The asymmetric bridge converter may couple the power source 104 to multiple phase coils 130 using multiple switching sections. The power converter 106 may operate to selectively (and independently) energize the phase coils of each phase of the SRM 102. Optionally, individual phase coils may be energized or de-energized such that only one phase is active (i.e., receives current) at a given time. Alternatively, multiple phases may be active simultaneously.

[0088] In the SRM 102, the phase windings 130 of different phases can be electrically isolated from each other. An asymmetric bridge converter can be used to electrically isolate the phase windings 130 of the SRM 102.

[0089] As discussed above, the controller 108 may drive the switched reluctance machine 102 by controlling the current provided to the phase coils of the switched reluctance machine 102 by the power source 104 and the converter 106. The controller 108 may be configured to drive the SRM 102 to minimize or reduce acoustic noise generated by the SRM 102 during operation. The controller 108 may identify a desired phase current waveform for each phase coil of the SRM 102. The desired phase current waveform may be identified as a phase current waveform associated with an optimal cumulative sound pressure level. For example, the controller 108 may be configured to implement a method for controlling a switched reluctance machine, such as an example method 200 described herein below with reference to FIG. 2.

[0090] The controller 108 can be communicatively coupled to one or more remote processing devices 112, for example, using a wired connection and / or a wireless communication module (e.g., Bluetooth, Bluetooth Low-Energy, WiFi, ANT+IEEE 802.11, etc.). The controller 108 can also be communicatively coupled to one or more remote processing devices 112 over a wide area network, for example, the Internet. The remote processing devices 112 can be any type of processing device, such as, but not limited to, a personal computer, a tablet, a cloud server, and a mobile device, such as a smartphone.

[0091] The remote processing device 112 typically includes a processing unit, output devices (such as a display, a speaker, and / or a haptic feedback device), a user interface, an interface unit for communicating with other devices, input / output (I / O) hardware, a wireless unit (e.g., a radio that communicates using CDMA, GSM, GPRS, or Bluetooth protocols according to standards such as IEEE 802.11a, 802.11b, 802.11g, or 802.11n), a power unit, and a memory unit, which may include RAM, ROM, one or more hard drives, one or more flash drives, or some other suitable data storage element such as a disk drive, etc.

[0092] The processing unit may be any suitable processor, controller, or digital signal processor capable of controlling the operation of the remote processing device 108 and providing sufficient processing power depending on the desired configuration, purpose and requirements of the system 100.

[0093] Aspects of the monitoring, storage, and analysis of the current waveforms, sound pressure levels, and predicted sound pressure levels may be performed by one or more of the controller 108 and / or remote processing devices 112. For example, a non-transitory storage memory (e.g., memory 114) of one or more of the controller 108 and / or remote processing devices 112 may store various sound pressure level related data and pre-processed data, such as, for example, predicted sound pressure levels, dynamic drive models, and SRM geometry and magnetic pole configuration data.

[0094] For example, the pre-processing step 420 may be performed by one or more of the controller 108 and / or the remote processing device 112. The pre-processed data (e.g., predicted sound pressure levels) may then be stored in a non-transitory storage memory (e.g., memory 114) for use in real time when the operating step 415 is performed in real time to control the switched reluctance machine.

[0095] Optionally, the pre-processing stage 420 and the operating stage 415 may be performed by one or more of the controller 108 and / or the remote processing device 112 in an offline operation. This offline operation may be defined to identify desired phase current waveforms (such as, for example, 230 herein below) at various different load conditions of the switched reluctance machine. These desired phase current waveforms (also referred to as predetermined desired phase current waveforms) may then be stored in a non-transitory storage memory (e.g., memory 114) for real-time use during operation of the switched reluctance machine. These predetermined desired phase current waveforms may then be used to control the operation of the switched reluctance machine in real-time (e.g., by providing current references to phases of the switched reluctance machine).

[0096] The remote processing device 112 may provide additional processing resources not available to the controller 108. Some aspects of determining the predicted sound pressure level for a given switched reluctance machine may be performed by the remote processing device 112. For example, the remote processing device 112 may perform aspects of the pre-processing stage 420. The controller 108 and the remote processing device 112 may communicate in real time to determine the cumulative sound pressure levels of potential current waveforms and / or to determine the desired current waveform.

[0097] Alternatively or additionally, pre-processed data such as predicted sound pressure levels for a given switched reluctance machine may be stored using memory 114. This enables controller 108 to determine cumulative sound pressure levels in potential current waveforms and / or to determine desired current waveforms locally and in real-time.

[0098] 2, an example method 200 for controlling a switched reluctance machine is shown. Method 200 is illustrative of a method for controlling a switched reluctance machine that may be used to operate a switched reluctance machine with reduced noise. Method 200 may be performed by various systems for controlling and driving a switched reluctance machine, such as example system 100 previously described herein.

[0099] Throughout the description of method 200 (and method 300 below), reference is made simultaneously to Figure 4, which illustrates example data flow between sub-processes that may be used to determine the cumulative sound pressure level and desired phase current waveforms for a switched reluctance machine. Sub-processes 401-412 shown in Figure 4 are examples of various sub-processes that may be used to determine the cumulative sound pressure level and desired phase current waveforms for a switched reluctance machine.

[0100] A pre-processing stage 420 may be used to determine features and characteristics of the switched reluctance machine that remain static during operation of the switched reluctance machine. The data generated during the pre-processing stage may then be used during the operation stage 415 of the switched reluctance machine (e.g., in sub-processes 408-412) to determine in real time desired current waveforms that may be applied to phase coils of the switched reluctance machine.

[0101] At 401, a pole configuration for the switched reluctance machine can be defined. The pole configuration can be defined based on the number of stator and rotor poles for the switched reluctance machine for which the current waveform is optimized. The pole configuration can include various arrangements of the stator and rotor, including internal or external rotor SRM. The pole configuration for the switched reluctance machine can be used to determine circumferential and time harmonics of the radial force waveform for the switched reluctance machine that impact the noise characteristics of the SRM.

[0102] At 402, a switched reluctance machine geometry can be defined. The switched reluctance machine geometry can include all the geometric and material details of the motor, such as stator geometry and steel properties, rotor geometry and steel properties, stator coils and material properties, stator housing and its material properties and boundary conditions (e.g., how the motor is mounted), etc.

[0103] The geometry of the switched reluctance machine can be used to determine various characteristics of the SRM such as the motor's flux density, torque, and voltage levels, as well as the radial force levels. Considering the acoustic noise generated by the SRM, the geometry of the switched reluctance machine can affect both the electrical and mechanical properties of the SRM.

[0104] The geometry of the switched reluctance machine may be used for further pre-processing and modeling of the switched reluctance machine. For example, the geometry of the SRM may be used to determine static electromagnetic properties of the SRM through electromagnetic finite element analysis (FEA), as described in more detail below in connection with 407. The geometry of the SRM may also be used to determine dominant mode properties of the SRM through vibro-acoustic FEA, as described in more detail below in connection with 404. Once the geometry of the switched reluctance machine is determined at 402, the static electromagnetic properties (at 407) and dominant mode properties (at 404) may be determined. The static electromagnetic properties (at 407) and dominant mode properties (at 404) may be determined simultaneously through respective electromagnetic FEA and vibro-acoustic FEA processes.

[0105] 2, a number of potential phase current waveforms may be determined for phase coils of the switched reluctance machine, at 210. Each potential phase current waveform may include a current profile that may be applied to a corresponding phase coil to drive rotation of a rotor of the switched reluctance machine.

[0106] Each potential phase current waveform can be defined for a complete electrical cycle of the switched reluctance machine. Each potential phase current waveform can include a current profile with a unique shape. Each current profile can have a unique combination of profile characteristics such as shape, magnitude, excitation period, etc.

[0107] The potential phase current waveforms may be determined based on operational constraints associated with the switched reluctance machine and / or associated drive system. In some cases, the potential phase current waveforms may be identified using an iterative process in which a first potential phase current waveform is defined and then iteratively modified to generate other potential phase current waveforms.

[0108] For example, the potential phase current waveform (or at least a first potential phase current waveform) may be determined using one or more predetermined current waveforms for a switched reluctance machine.

[0109] Alternatively, the potential phase current waveforms may be determined (at least in part) using conventional control methods. For example, a first potential phase current waveform may be identified using conventional control methods. This first potential phase current waveform may then be iteratively varied to generate a plurality of potential phase current waveforms.

[0110] At 220, a cumulative sound pressure level may be determined for each potential phase current waveform from 210. An example process 300 for determining the cumulative sound pressure level in a given phase current waveform is described in further detail herein below in connection with FIG.

[0111] The cumulative sound pressure level may be determined based on multiple harmonic sound pressure levels expected to result from the potential phase current waveform, i.e., the cumulative sound pressure level may be determined based on a combination of sound pressure levels from different harmonics expected to result from application of the potential phase current waveform to corresponding phase coils of the SRM.

[0112] Optionally, the cumulative sound pressure level may be determined based on a plurality of filtered harmonic sound pressure levels. For example, harmonic sound pressure levels corresponding to sounds (e.g., frequencies) that are not perceptible to the human ear may be filtered out in determining the cumulative sound pressure level. This allows evaluation of potential phase current waveforms based on sound pressure levels that are perceptible to humans and that are more likely to result in an undesirable experience for humans in the vicinity of the switched reluctance machine when in use.

[0113] At 230, a desired phase current waveform may be identified from the plurality of potential phase current waveforms from 210. The desired phase current waveform may be identified as the potential phase current waveform associated with an optimal cumulative sound pressure level. The optimal cumulative sound pressure level may be determined directly based on a dynamic calculation of the cumulative sound pressure level predicted to result from each potential phase current waveform.

[0114] For switched reluctance machines, the phase current waveforms are typically specified to achieve one or more operating objectives. Examples of operating objectives include average torque, torque ripple, and radial force waveforms. Evaluating how the phase current waveforms perform with respect to one or more operating objectives often involves modeling the operation of the switched reluctance machine.

[0115] One or more objective functions may be used to identify operating parameters that may be used to determine the potential phase current waveform. The objective functions may be used to determine operating parameters such as average torque, torque ripple, dynamic currents, and sound pressure levels associated with the potential phase current waveform. These operating parameters may then be evaluated based on one or more operating goals for the switched reluctance machine. Illustrative examples of sub-processes that may be used as objective functions are described in further detail below with respect to sub-processes 408-411 of FIG. 4.

[0116] Optionally, an objective function may be calculated for the switched reluctance machine during an offline process to determine desired phase current waveforms. Operation of the switched reluctance machine may then be controlled in real time with reference to these predetermined desired phase current waveforms.

[0117] Alternatively, the preprocessed data corresponding to the switched reluctance machine may be used to calculate the objective function in real time for the switched reluctance machine. To determine the sound pressure level at a given potential current waveform, the radial force produced by that potential current waveform needs to be determined. Similarly, other operating parameters such as electromagnetic torque and torque ripple may be considered when evaluating a given potential current waveform. Radial force, torque, and torque ripple are electromagnetic quantities resulting from electromechanical energy conversion. Thus, the preprocessed data may include the static electromagnetic characteristics of the switched reluctance machine determined during the preprocessing stage.

[0118] Static electromagnetic properties, such as radial and tangential magnetic flux densities, may be determined as a function of operational variables, such as rotor position, spatial position within the air gap, and magnitude of current. For example, the static electromagnetic properties may be modeled using Maxwell's stress tensor, as described in more detail below in connection with sub-process 407. The static electromagnetic properties may then be stored for subsequent access and use during computation of the objective function.

[0119] The static electromagnetic characteristics may be stored in a non-transitory storage memory, such as memory 114. The stored static electromagnetic characteristics may then be accessed during operation process 415 and used to determine a desired current waveform and associated cumulative sound pressure level. For example, the static electromagnetic characteristics may be stored as a look-up table that is indexed based on the corresponding operating variables, allowing for dynamic determination of operating parameters for potential current waveforms, as described in further detail below, for example, in connection with sub-process 408.

[0120] The preprocessed data may include static vibro-acoustic characteristics of the switched reluctance machine determined during the preprocessing stage. The static vibro-acoustic characteristics, such as sound pressure level, may be determined as a function of operating variables, such as forcing frequency and circumferential order. For example, the static vibro-acoustic characteristics may be determined as described in further detail below in connection with sub-process 406 of FIG. 4. The static vibro-acoustic characteristics may then be stored for subsequent access and use during computation of the objective function.

[0121] The static vibration acoustic signature may be stored in a non-transitory storage memory, such as memory 114. The stored static vibration acoustic signature may then be accessed during operation process 415 and used to determine a desired current waveform and associated cumulative sound pressure level. For example, the static vibration acoustic signature may be stored as a look-up table that is indexed based on the corresponding operating variables, thereby allowing for dynamic determination of acoustic noise levels in potential current waveforms, as described in further detail below, for example, in connection with sub-process 411.

[0122] Current profiling may be used to determine desired phase current waveforms using outputs from objective functions (e.g., dynamic calculation of acoustic noise level at 411 and dynamic calculation of operating parameters at 408). An example of the current profiling sub-process is shown in FIG. 4 at 412. The current profiling method may be applied to achieve one or more operating objectives such as, for example, maximizing average torque, minimizing torque ripple, minimizing phase current, and minimizing sound pressure level.

[0123] The desired phase current waveform may be determined using an iterative optimization process. At each iteration of the optimization, a phase current profile of one of the potential phase current waveforms from 210 may be evaluated. The evaluation of the iteration-specific potential phase current waveform at a given iteration may include determining (at 220) a cumulative sound pressure level for that iteration-specific potential phase current waveform (e.g., using sub-process 411).

[0124] Optionally, the evaluation of the iteration-specific potential phase current waveform at a given iteration may include determining other operational targets for that iteration-specific potential phase current waveform. For example, the phase current profile of the iteration-specific potential phase current waveform may be provided as an input to a dynamic motor drive model in sub-process 408. The dynamic motor drive model may be defined to calculate operational targets for that iteration-specific potential phase current waveform using stored electromagnetic characteristics of the switched reluctance machine and other operational parameters such as motor speed and DC link voltage.

[0125] The desired phase current waveform may then be identified as one of the iteration-specific potential phase current waveforms evaluated through the iterative optimization process. The desired phase current waveform may then be identified as the iteration-specific potential phase current waveform that corresponds to the optimal cumulative sound pressure level.

[0126] The optimum cumulative sound pressure level may be determined through an optimization process defined to minimize the acoustic noise level of the SRM. Optionally, the optimum cumulative sound pressure level may be determined as the minimum achievable sound pressure level, thereby ensuring that the switched reluctance machine operates with an absolute minimum level of acoustic noise generation.

[0127] Alternatively or additionally, the optimal cumulative sound pressure level may be determined through an optimization process that considers acoustic noise levels in combination with other operational objectives, and the optimization process may be defined to determine the optimal cumulative sound pressure level as the minimum sound pressure level achievable by a potential current waveform that also achieves the other operational objectives.

[0128] Alternatively, the optimization process can be defined to minimize a multivariate cost function that includes the cumulative sound pressure level as one of the variables. The cost function can be a weighted cost function that considers the acoustic noise level in combination with other operational objectives. The optimal cumulative sound pressure level can then be determined as the sound pressure level that corresponds to the potential current waveform that results in the lowest overall cost of the cost function.

[0129] Regardless of the operational objective being optimized, the optimization process can be constrained by both practical and operational constraints. For example, the optimization process can be constrained to ensure that operational parameters such as average torque, torque ripple, dynamic currents, sound pressure levels, etc., fall within prescribed operating ranges for a given application.

[0130] At 240, the desired phase current waveforms from 230 may be applied to corresponding phase coils of a switched reluctance machine. Application of the current waveforms to the phase coils of the switched reluctance machine may be used to drive operation of the switched reluctance machine.

[0131] The example method 200 described herein above relates to a process for determining a desired phase current waveform for one phase coil of a switched reluctance machine. However, the method 200 may be implemented to simultaneously determine a respective desired phase current waveform for each phase coil of a switched reluctance machine.

[0132] That is, desired phase current waveforms may be determined simultaneously. The desired phase current waveforms may include corresponding desired current waveforms in each of the phases of the switched reluctance machine. The desired phase current waveforms may be determined by optimizing the current waveforms for all phases simultaneously. This may include analyzing the acoustic noise level (and other operating parameters) based on the potential waveforms in each phase of the switched reluctance machine. For a three-phase switched reluctance machine, the desired phase current waveforms may include a desired incoming phase current waveform in an incoming phase (phase current increasing), a desired outgoing phase current waveform in an outgoing phase (phase current decreasing), and a desired conducting phase current waveform in a conducting phase.

[0133] Method 200 may also be performed continuously by a controller of a switched reluctance machine, i.e., the controller may continually / persistently determine a desired phase current waveform and / or a desired number of phase current waveforms for the switched reluctance machine, to ensure that the switched reluctance machine continues to operate with reduced / minimized acoustic noise.

[0134] Optionally, steps 210-230 of method 200 may be performed in real time to determine desired phase current waveforms for the switched reluctance machine, which may include measuring operating parameters of the switched reluctance machine and using these values ​​for various sub-processes (e.g., using the measured phase currents as inputs to sub-process 412 rather than the phase currents generated by the drive model).

[0135] Alternatively, steps 210-230 of method 200 may be performed offline to predetermine the desired phase current waveforms for the switched reluctance machine, including determining the desired phase current waveforms at various load conditions and operating parameters of the switched reluctance machine. The predetermined desired phase current waveforms may then be stored in a non-transitory storage memory, and at 240, the predetermined desired phase current waveforms may be accessed during operation of the switched reluctance machine to determine the desired phase current waveforms to be applied.

[0136] 3, an example method 300 for determining cumulative sound pressure levels associated with operation of a switched reluctance machine is shown. Method 300 is an example process that may be used to determine the cumulative sound pressure levels expected to result from application of particular phase current waveforms to corresponding phase coils of a switched reluctance machine.

[0137] Method 300 may be used to evaluate a plurality of potential phase current waveforms, for example, in step 220 of method 200. As previously described, a desired phase current waveform may be determined through an iterative optimization process that is used to evaluate a plurality of potential phase current waveforms. Method 300 may be repeated for each potential phase current waveform in the plurality of potential phase current waveforms. For each iteration of method 300, a corresponding iteration-specific potential phase current waveform may be identified.

[0138] At 310, a number of repetition-specific circumferential orders associated with the repetition-specific potential phase current waveforms may be identified.

[0139] The multiple iteration-specific circumferential orders may be determined based on the radial force waveform associated with the iteration-specific potential phase current waveform. For example, the multiple iteration-specific circumferential orders may be calculated by applying a Fast Fourier Transform to the radial force waveform in sub-process 409.

[0140] At 320, magnitudes of a number of radial force harmonics associated with the repetition-specific potential phase current waveforms may be determined.

[0141] The magnitudes of the multiple radial force harmonics may also be determined based on the radial force waveform associated with the repetition-specific potential phase current waveform. For example, the magnitudes of the multiple radial force harmonics may be calculated by applying a Fast Fourier Transform to the radial force waveform in sub-process 409.

[0142] Although 310 and 320 are shown as separate steps, the multiple iteration specific circumferential orders and the magnitudes of the multiple radial force harmonics may be determined simultaneously. For example, the multiple iteration specific circumferential orders and the magnitudes of the multiple radial force harmonics may be determined simultaneously by applying a two-dimensional fast Fourier transform to the radial force waveform in sub-process 409. By applying a two-dimensional fast Fourier transform to the radial force waveform, the circumferential and time orders (discussed further below in 330) associated with the dynamic radial force waveform may be identified along with the magnitudes of the radial force harmonics at the respective time and circumferential orders.

[0143] Optionally, the multiple iteration-specific circumferential orders determined in 310 may include only those circumferential orders that are dominant for the switched reluctance machine, i.e., the multiple iteration-specific circumferential orders determined in 310 may be limited to those circumferential orders determined by applying a two-dimensional Fourier transform to the radial force waveform, which are also identified as the dominant circumferential orders for the switched reluctance machine.

[0144] Dominant circumferential orders generally refer to those circumferential orders (e.g., more than other circumferential orders for SRM) that contribute substantially to acoustic noise and vibration. The radial forces in switched reluctance motors can have many harmonics. However, due to the pole configuration, number of phases, and number of poles, certain circumferential orders contribute more to acoustic noise and vibration than other frequency orders. From a computational perspective, it is often not efficient to evaluate all possible circumferential orders for a switched reluctance machine. Therefore, the dominant circumferential orders can be determined during a pre-processing stage. This can allow a more efficient evaluation of the predicted acoustic noise levels associated with potential current waveforms, with minimal impact on the accuracy of the prediction.

[0145] The dominant circumferential orders may be determined in sub-process 403 of pre-processing stage 420. For a switched reluctance machine, multiple dominant circumferential vibration modes may be determined based on the magnetic pole configuration (e.g., from 401) of the switched reluctance machine.

[0146] The number of magnetic poles of the switched reluctance machine can be determined. For example, the magnetic poles can be determined as the ratio between the number of stator poles Ns of the switched reluctance machine and the number of phases m of the switched reluctance machine. The dominant circumferential vibration modes can be identified depending on the number of magnetic poles.

[0147] The difference between the dominant circumferential orders of vibration modes in an SRM can be determined by the number of magnetic poles. For example, the dominant circumferential order v of a vibration mode of an SRM can be determined using a function of the number of magnetic poles and any integer number i according to the following equation:

number

[0148] The dominant circumferential order v may include each circumferential order vibration mode determined for any integer number i according to:

number

[0149] The dominant circumferential modes may then be stored for later use during pre-processing stage 420 and operation stage 415 for the switched reluctance machine. For example, the dominant circumferential modes for the switched reluctance machine may be stored in a non-transient storage memory, such as memory 114. Harmonic sound pressure levels may then be determined (e.g., in sub-process 411) only for those harmonics that correspond to the dominant circumferential modes determined for the switched reluctance machine.

[0150] Optionally, higher circumferential orders may also be identified that may affect multiple dominant circumferential modes due to sampling effects. Higher vibration modes usually have high natural frequencies. As a result, it is typically difficult for higher circumferential radial force harmonics to excite corresponding vibration modes of the SRM. However, higher circumferential radial force harmonics may excite lower circumferential vibration modes due to sampling effects. Including the impact of sampling effects on dominant circumferential modes ensures that the sound pressure levels determined for the potential current waveforms take into account higher circumferential modes that may contribute significantly to the acoustic noise generated by the SRM.

[0151] For each dominant circumferential mode, higher circumferential order radial force harmonics that excite lower order vibrational modes can be identified due to sampling effects of the radial force density harmonics. For example, these higher circumferential order modes that affect the dominant circumferential mode due to sampling effects can be determined in sub-process 405.

[0152] Higher circumferential order radial force harmonics that can contribute to the sound pressure level produced by exciting lower order circumferential vibration modes can be identified as circumferential orders v, identified according to: For the stator of an internal rotor (IT) SRM, v>N s / 2 For external rotor (ER) SRM rotors, v>N r / 2 where Ns is the number of stator poles (e.g., determined from 401) and Nr is the number of rotor poles (e.g., determined from 401).

[0153] Number of stator poles N s , number of rotor poles N r For a switched reluctance machine having a pole configuration with m phases, the sampling effect can be calculated as follows: mod(v,N s )=0, this radial force harmonic excites mode 0. mod(v,N s ) ≠ 0 and mod(v,N s When / m)=0, this radial force harmonic is mode N s Exciting / m

[0154] Thus, higher circumferential order modes that influence a dominant circumferential mode may be identified as the higher circumferential order modes whose corresponding radial force harmonics excite the dominant circumferential mode.

[0155] For switched reluctance machines, multiple higher circumferential order modes that influence the dominant circumferential mode can be identified. The impact of higher circumferential orders of radial force density harmonics on vibration and acoustic noise decreases as the circumferential order increases due to sampling effects. Therefore, the set of circumferential orders included in the multiple higher circumferential order modes that influence the dominant circumferential mode can be selected based on a trade-off between accuracy and computational cost. This may depend on the specific implementation of the SRM drive.

[0156] Optionally, the multiple higher circumferential order modes are up to 2N for the stator of the inner rotor SRM. s , or max. 2N for external rotor SRM rotors r This cutoff may provide good accuracy at relatively little computational cost.

[0157] 8(a)-8(d) show exemplary plots of the sampling effect on high circumferential harmonics for a switched reluctance machine with 12 stator poles, in which the stator is shown cut in half and laid flat for ease of illustration.

[0158] Figure 8(a) shows the sampling effect on the 8th circumferential mode. Before sampling, this circumferential mode has 8 peaks and 8 valleys. This harmonic generates a force on the stator poles as shown in Figure 8(a). When the 8th circumferential mode is sampled by the stator poles it interacts with (i.e., the 2nd, 5th, 8th, and 11th stator poles), it creates 4 peaks and 4 valleys that can excite vibration mode 4.

[0159] Figure 8(b) shows the sampling effect of the 12th circumferential mode. Before sampling, this circumferential mode has 12 peaks and 12 valleys. For this circumferential mode, the stator poles sample the same points on the waveform. Due to the sampling effect, this harmonic can excite vibration mode 0.

[0160] Figure 8(c) shows the effect of sampling the 16th circumferential mode. Before sampling, this circumferential mode has 16 peaks and 16 valleys. This harmonic generates a force on the stator poles. When the 16th circumferential mode is sampled by the stator poles it interacts with (i.e., the second, fifth, eighth, and eleventh stator poles), it again has four peaks and four valleys that can excite vibration mode 4.

[0161] Figure 8(d) shows the effect of sampling the 20th circumferential mode. Before sampling, this circumferential mode has 20 peaks and 20 valleys. This harmonic generates forces on the stator poles. When the 20th circumferential mode is sampled by the stator poles it interacts with (i.e., the first, fourth, seventh and tenth stator poles), it again has four peaks and four valleys which can excite vibration mode 4.

[0162] FIG. 9 shows a plot of the sound pressure level (SPL) generated by vibration mode 0 of a switched reluctance machine with 12 stator poles and 16 rotor poles. The plot shown in FIG. 9 shows the sound pressure level generated for different frequency ranges when considering the sampling effect of higher circumferential orders. For the plot shown in FIG. 9, the magnitude of radial force density harmonics of different circumferential orders is the same.

[0163] The natural frequency of vibration mode 0 is 7636.5 Hz. As shown in the plot of FIG. 9, the natural frequency of vibration mode 0 corresponds to the maximum SPL at each plotted circumferential order. As shown in the plot of FIG. 9, the radial force harmonics associated with circumferential modes 12 and 24 contribute to the vibration of mode 0 due to sampling effects. These higher order circumferential modes excite vibration mode 0 in a manner similar to the radial force circumferential mode 0.

[0164] Figure 9 also shows that circumferential mode 4 also excites mode 0, causing a small sound pressure level due to forced excitation. Circumferential modes 8, 16, and 20 also contribute to the sound pressure in a similar manner to mode 4 due to sampling effects. Although these modes may contribute less to the overall sound pressure level, at certain forcing frequencies they can still have a significant impact on the overall acoustic noise level of a switched reluctance machine.

[0165] As Figure 9 shows, omitting the acoustic noise generated as a result of sampling effects can lead to an underestimation of the acoustic noise level. Therefore, by identifying the higher circumferential order modes that affect the dominant circumferential mode of the switched reluctance machine due to sampling effects, the sound pressure level for a given potential current waveform can be determined more accurately.

[0166] As previously discussed, the magnitudes of the multiple repetition specific circumferential orders and multiple radial force harmonics (and the repetition specific time orders) can be determined for the switched reluctance machine based on the radial force waveform.

[0167] The radial force waveform can be dynamically determined for the switched reluctance machine. For example, the radial force waveform can be determined based on the motor speed and DC link voltage of the switched reluctance machine. The radial force waveform can be determined using a dynamic motor drive model for the switched reluctance machine, which is represented by sub-process 408.

[0168] In sub-process 408, a dynamic motor drive model can determine a radial force waveform for a given potential phase current waveform based on the current profile in the potential phase current waveform. The dynamic motor drive model can be defined to solve phase voltage equations to dynamically calculate the radial force waveform while considering the voltage dynamics of the switched reluctance machine at a given motor speed (i.e., the current operating motor speed) and DC link voltage. The dynamic motor drive model can take into account the current tracking capabilities of the SRM drive in determining the radial force waveform.

[0169] The dynamic motor drive model can be configured to determine a radial force waveform based on the electromagnetic characteristics of the switched reluctance machine. The static electromagnetic characteristics of the switched reluctance machine can be determined during pre-processing stage 420, for example using sub-process 407.

[0170] Sub-process 407 can be prescribed to determine the static electromagnetic characteristics of the switched reluctance machine using the known geometry of the SRM. The SRM geometry includes salient pole structures that are an important part of the torque generating mechanism. As a result, the electromagnetic properties such as flux linkage and torque are functions of the rotor position. Radial forces arise from the radial and tangential air gap flux densities and are therefore functions of the spatial position of the SRM.

[0171] These characteristics also depend on the phase current waveforms applied to the SRM. When current is applied to the electrical coils of a given phase of the SRM, most of the magnetic flux produced will link with the electrical coils of the same phase. Typically, SRMs have negligible mutual flux linkages. Therefore, once the electromagnetic properties of one phase of an SRM are determined, they can be applied to the other phases of the SRM as well.

[0172] In sub-process 407, electromagnetic finite element analysis (FEA) of the SRM can be used to determine the electromagnetic characteristics of one phase of the SRM. Multiple constant currents can be applied to one phase of the SRM over one electrical cycle. Values ​​of flux linkage, torque, voltage, radial and tangential flux density, and radial force can be determined as a function of excitation current and rotor position through electromagnetic FEA. The radial and tangential flux density and radial force are also functions of spatial position in the air gap.

[0173] These values ​​of the electromagnetic properties may be stored in a non-transitory storage memory, such as memory 114. For example, the values ​​of the electromagnetic properties may be stored as a look-up table that is indexed based on operating parameters such as the excitation current (e.g., potential current waveform), rotor position, and spatial position within the air gap.

[0174] Sub-process 407 can automatically generate an electromagnetic FEA model for a given SRM geometry, pole configuration, winding layout, and core material. The model can be two-dimensional or three-dimensional depending on the particular application. Electromagnetic properties can be calculated with or without considering mutual coupling effects, regardless of whether the FEA model is two-dimensional or three-dimensional.

[0175] The values ​​of the electromagnetic properties determined in sub-process 407 may be nonlinear. The magnetic properties of the SRM often depend heavily on the properties of the magnetic steel sheet. For a particular magnetic steel sheet, the relationship between the magnetic field strength and the magnetic flux density is typically nonlinear, resulting in nonlinear electromagnetic properties of the SRM.

[0176] Referring again to sub-process 408, the radial force waveform may be determined by dynamic motor drive model 408 using stored values ​​of the electromagnetic properties (e.g., as determined by sub-process 407). The radial force waveform may be dynamically determined based on real-time operating parameters of the SRM and potential current waveforms.

[0177] The dynamic motor drive model 408 may also determine the dynamic torque and torque ripple using the stored electromagnetic property values. The dynamic torque and torque ripple may also be dynamically determined based on real-time operating parameters of the SRM and potential current waveforms. As previously mentioned, the dynamic torque and torque ripple calculated from the dynamic motor drive model in sub-process 408 may be used as part of an iterative optimization process used to determine desired phase current waveforms (e.g., as part of method 200 previously described herein).

[0178] In sub-process 408, the SRM phase voltage equations may be dynamically solved using the inverse flux linkage characteristic according to the following:

number

[0179] Typically, calculating the phase current involves applying the conduction angle to a current controller to regulate the phase current within a given conduction angle with respect to a given phase current reference (i.e., the current profile of the phase current waveform).

[0180] Referring again to sub-process 408, for the calculated switching conditions of the asymmetric bridge converter, the differential equations are solved using the phase voltages to determine the current waveforms in an offline process. To track the current waveforms, a corresponding current reference can be applied to the current controller.

[0181] Once the phase current waveforms are determined (offline), they can be used to identify corresponding stored values ​​of torque, tangential and radial flux densities, and radial force to determine the dynamic waveforms of these quantities. The radial force waveform can be calculated from the radial force using the Maxwell stress tensor method using the tangential and radial flux densities.

[0182] Alternatively, sub-process 408 may be used as part of a real-time determination of a desired phase current waveform. In such a case, the phase currents output by 408 may be replaced with measured phase current waveforms. The measured phase current waveforms may be used as feedback signals in determining the phase current waveforms.

[0183] 15 shows an example radial force density waveform determined using an example dynamic motor drive model for one electrical cycle. The example radial force density waveform shown in FIG. 15 was calculated with the Maxwell stress tensor method using tangential and radial magnetic flux densities.

[0184] 3, a number of iteration-specific forcing frequencies associated with the iteration-specific potential phase current waveforms may be determined at 330. Although 310, 320, and 330 are shown as separate steps, the number of iteration-specific circumferential orders, the number of radial force harmonic magnitudes, and the number of iteration-specific forcing frequencies may be determined simultaneously.

[0185] The sound pressure level resulting from the application of a given current waveform to a switched reluctance machine can be determined as a function of frequency. The frequencies of the radial force harmonics resulting from a given current waveform are related to the time order and the motor speed. Thus, multiple iteration specific forcing frequencies (with respect to corresponding iteration specific potential current waveforms) can be determined based on the motor speed and multiple iteration specific time orders of the SRM.

[0186] For example, multiple iteration-specific forcing frequencies may be determined in sub-process 410 by calculating the frequencies of radial force harmonics at iteration-specific time orders.

[0187] To determine the multiple iteration-specific forcing frequencies, multiple iteration-specific time orders associated with the iteration-specific potential phase current waveforms may be determined. A radial force waveform associated with the iteration-specific potential phase current waveforms may be determined (e.g., in sub-process 408 described above). The multiple iteration-specific time orders may then be calculated by applying a Fast Fourier Transform to the radial force waveforms, as described above in connection with sub-process 409. For example, the iteration-specific time orders may be calculated in sub-process 409 simultaneously with multiple iteration-specific circumferential orders and multiple radial force harmonics magnitudes.

[0188] Applying a fast Fourier transform to the radial force waveform may identify all of the time orders associated with the radial force waveform. Optionally, the multiple iteration-specific forcing frequencies may be determined based on a subset of the iteration-specific time orders that includes only the dominant time orders for the switched reluctance machine. That is, the subset of dominant iteration-specific time orders may be determined from the iteration-specific time orders by identifying the time orders determined by applying a two-dimensional Fourier transform to the radial force waveform that are also identified as dominant time orders for the switched reluctance machine.

[0189] Similar to the circumferential orders discussed above, certain time orders contribute more to acoustic noise and vibration than others due to the pole configuration, number of phases, and number of poles. From a computational perspective, it is often not efficient to evaluate all possible time orders of a switched reluctance machine. Therefore, the dominant time orders can be determined during a pre-processing stage. This allows for a more efficient assessment of the predicted acoustic noise levels associated with potential current waveforms with minimal impact on the accuracy of the prediction.

[0190] The dominant time order may be determined in sub-process 403 of pre-processing stage 420. Multiple dominant time orders may be determined for the switched reluctance machine based on the pole configuration (e.g., from 401) and geometry (e.g., from 402) of the switched reluctance machine.

[0191] The multiple dominant time orders can be determined based on the number of strokes in one mechanical revolution of the switched reluctance machine. The number of strokes can be identified as the product of the number of rotor poles (Nr) of the switched reluctance machine and the number of phases (m) of the switched reluctance machine. The specific manner in which the dominant time orders are identified depends on the construction of the particular switched reluctance machine.

[0192] The difference between the dominant time orders in an SRM is related to the number of strokes. Thus, the dominant time order can be determined as a function of the number of strokes and any integer j. The dominant time order is also related to other factors such as the direction of rotation, either clockwise (CW) or counterclockwise (CCW), whether the SRM is an internal rotor (IR) or external rotor (ER) motor, and the magnetic pole configuration. These factors change the location of the time order relative to the dominant circumferential order. The effect of these factors on the dominant time order is also related to the integer i, which can be used to define the dominant circumferential order as discussed above.

[0193] For the stator of an internal rotor SRM, if the stator pole number Ns is greater than the rotor pole number Nr and the rotor rotates in a counterclockwise direction, the dominant time order can be calculated as follows: u=mN r j+N r i

[0194] For the stator of an internal rotor SRM, if the stator pole number Ns is greater than the rotor pole number Nr and the rotor rotates in a clockwise direction, the dominant time orders can be calculated as follows: u=mN r jN r i

[0195] For the stator of an internal rotor SRM, if the stator pole number Ns is less than the rotor pole number Nr and the rotor rotates in a counterclockwise direction, the dominant time order can be calculated as follows: u=mN r jN r i

[0196] For the stator of an internal rotor SRM, if the stator pole number Ns is less than the rotor pole number Nr and the rotor rotates in a clockwise direction, the dominant time orders can be calculated as follows: u=mN r j+N r i

[0197] For the rotor of an external rotor SRM, if the stator pole number Ns is greater than the rotor pole number Nr and the rotor rotates in a counterclockwise direction, the dominant time order can be calculated as follows: u=mN r j+N s i

[0198] For the rotor of an external rotor SRM, if the stator pole number Ns is greater than the rotor pole number Nr and the rotor rotates in a clockwise direction, the dominant time order can be calculated as follows: u=mN r jN s i

[0199] For the rotor of an external rotor SRM, if the stator pole number Ns is less than the rotor pole number Nr and the rotor rotates in a counterclockwise direction, the dominant time order can be calculated as follows: u=mN r jN s i

[0200] For the rotor of an external rotor SRM, if the stator pole number Ns is less than the rotor pole number Nr and the rotor rotates in a clockwise direction, the dominant time order can be calculated as follows: u=mN r j+N s i

[0201] Referring to Figure 16, an example plot of radial force harmonics in the u-v plane at dominant circumferential modes and dominant time orders associated with a switched reluctance machine having 18 stator poles and 12 rotor poles is shown. The plot shown in Figure 16 is the result of applying a two-dimensional fast Fourier transform to the radial force density waveform for the switched reluctance machine. The plot shown in Figure 16 shows the magnitude of the radial force density harmonics at dominant circumferential and time orders.

[0202] In Figure 16, the circumferential and time axes include positive and negative values. The circumferential-time plane shown in Figure 16 can be divided into four quadrants centered at (0,0): a first quadrant (upper right quadrant) including positive circumferential values ​​and positive time values, a second quadrant (upper left quadrant) including negative circumferential values ​​and positive time values, a third quadrant (lower left quadrant) including negative circumferential values ​​and negative time values, and a fourth quadrant (lower right quadrant) including positive circumferential values ​​and negative time values.

[0203] As the plot in Fig. 16 shows, the radial force harmonics are symmetric in the four quadrants of the uv-plane. Fig. 17 shows an example plot of the radial force harmonics in the first quadrant of the uv-plane for the dominant circumferential mode and dominant time order associated with the switched reluctance machine with 18 stator poles and 12 rotor poles plotted in Fig. 16.

[0204] 3, a cumulative sound pressure level may be determined for the iteration-specific potential phase current waveforms at 340. The cumulative sound pressure level may be determined using the multiple iteration-specific circumferential orders (from 310), the multiple iteration-specific forcing frequencies (from 330), and the multiple radial force harmonic magnitudes (from 320).

[0205] A cumulative sound pressure level can be determined by combining multiple iteration-specific harmonic sound pressure levels. Each iteration-specific harmonic sound pressure level can be determined based on a combination of one of the iteration-specific circumferential orders, one of the iteration-specific forcing frequencies, and one of the radial force harmonic magnitudes. The iteration-specific harmonic sound pressure level can be determined based on simulated sound pressure levels associated with a particular circumferential order and a particular forcing frequency.

[0206] For example, in sub-process 411, multiple iteration-specific harmonic sound pressure levels may be determined for iteration-specific circumferential orders at different iteration-specific forcing frequencies using the magnitudes of the associated radial force harmonics and the simulated sound pressure levels associated with particular combinations of circumferential orders and forcing frequencies.

[0207] For example, simulated sound pressure levels associated with particular azimuthal orders and particular forcing frequencies may be determined during pre-processing stage 420 in sub-process 406. The simulated sound pressure levels may be stored in a non-transitory storage memory, such as memory 114. The simulated sound pressure levels may be accessed during operation process 415 and used to determine the cumulative sound pressure levels. For example, the simulated sound pressure levels may be stored as a look-up table indexed based on the corresponding azimuthal orders and forcing frequencies, facilitating dynamic determination of harmonic sound pressure levels in the potential current waveform in sub-process 411.

[0208] The simulated sound pressure levels can be determined using unit force values, and thus the harmonic sound pressure levels in the potential current waveform can be determined by applying the magnitude of each radial force harmonic to the stored values ​​of the simulated sound pressure levels.

[0209] A stored value of the sound pressure level may be determined for a dominant circumferential mode of the SRM as a function of frequency. To determine the simulated sound pressure level for a given switched reluctance machine, the dominant model characteristics of the SRM may be characterized. In sub-process 404, the geometry of the switched reluctance motor (from sub-process 402) may be used to characterize the dominant mode characteristics. For example, the dominant mode characteristics may be characterized through a vibro-acoustic finite element analysis of the switched reluctance machine.

[0210] In a switched reluctance machine, sound pressure is generated when radial forces excite mode shapes of the motor structure. If the forcing frequency of the radial force waveform matches the natural frequency of a vibration mode that has the same shape as a harmonic of that radial force, the vibration will be stronger. This causes resonance. Thus, the natural frequency of the vibration mode shape of the SRM will affect how a given current waveform will affect the sound pressure level of the SRM.

[0211] In sub-process 404, the geometry and structure of the SRM can be modeled using vibro-acoustic FEA software with boundary conditions. The vibro-acoustic FEA can model the SRM to identify vibration modes of the SRM that correspond to the dominant circumferential orders of the SRM (e.g., from sub-process 403).

[0212] Vibro-acoustic FEA can generate a set of natural frequencies for different mode shapes through a modal analysis of the SRM. The displacements on the motor surface for each mode shape can be identified. The displacements can be determined across the circumference of the stator.

[0213] The displacement distribution (from 403) in the dominant circumferential mode can be analytically calculated for a cylindrical shell (a cylinder with a hollow center). These analytically calculated displacements are also determined over the circumference of the stator (represented by a cylindrical shell).

[0214] The surface displacements determined based on the geometry and structure of the SRM (e.g., from vibro-acoustic FEA) can then be compared to the displacements calculated for the cylindrical shell using the dominant circumferential modes. A modal correlation coefficient can be calculated based on the comparison to identify the natural frequencies of the dominant circumferential modes. Through this modal analysis, for each dominant circumferential mode, the natural frequency associated with that dominant circumferential mode can be identified.

[0215] 5, an exemplary plot of circumferential orders of vibration modes and associated natural frequencies determined for a switched reluctance machine with 18 stator poles and 12 rotor poles is shown. The plot shown in FIG. 5 shows the natural frequencies for circumferential mode 0 and circumferential mode 6 in an SRM with 18 stator poles and 12 rotor poles.

[0216] Figure 6 shows an example plot of surface displacement for vibration mode 0 of the SRM with 18 stator poles and 12 rotor poles from Figure 5. The plot in Figure 6 shows an example shape of the dominant vibration mode of the SRM, in this case vibration mode 0. As shown in Figure 6, the natural frequency for circumferential mode 0 was identified at 6075 Hz.

[0217] Figure 7 shows an example plot of surface displacement at vibration mode 6 of the SRM with 18 stator poles and 12 rotor poles from Figure 5. The plot in Figure 7 shows an example shape of the dominant vibration mode of the SRM, in this case vibration mode 6. As shown in Figure 7, the natural frequency at circumferential mode 0 was identified at 5895 Hz.

[0218] The simulated sound pressure level associated with a given circumferential mode may also be determined using a vibro-acoustic model of the switched reluctance machine. In sub-process 406, the simulated sound pressure level associated with a given circumferential mode may be determined by simulating a composite sound pressure level corresponding to at least one frequency value close to the natural frequency (determined from 404). The simulated sound pressure level may then be stored for use during operation process 415, which is used to determine a cumulative sound pressure level.

[0219] The simulated sound pressure levels associated with a given circumferential order and forcing frequency may be defined using unit force. The unit force may be applied to the vibro-acoustic model (from 404) of a given motor structure to determine the simulated sound pressure levels resulting from a particular combination of circumferential order and forcing frequency. When determining the harmonic sound pressure levels of the potential waveform (i.e., at 340), the magnitudes of the radial force harmonics associated with a particular combination of circumferential order and forcing frequency may be applied to the simulated sound pressure levels using unit force to determine the predicted sound pressure levels in the potential waveform.

[0220] 10 shows an exemplary plot of simulated sound pressure levels determined over a frequency range. The simulated sound pressure levels were determined through vibro-acoustic FEA simulations of the SRM at a particular speed.

[0221] In the plot shown in Figure 10, a number of simulated sound pressure levels are shown corresponding to a number of different radial force harmonic magnitudes. As Figure 10 shows, a 100-fold increase in radial force magnitude results in an approximately 40 dB increase in sound pressure level.

[0222] The sound pressure level and the magnitude of the radial force harmonics are related through the relationship between displacement, sound power, and sound pressure level. If the force magnitude increases by a factor of n, the displacement also increases by a factor of n since the displacement can be directly correlated to the force magnitude according to:

number

[0223] Sound power can be related to the square of the displacement according to: 2 It can also be doubled.

number

[0224] The acoustic power level after increasing the force magnitude by a factor of n can be determined by:

number

number

[0225] The sound pressure level can then be determined according to: SPL2 = SPL1 + 20 log(n)

[0226] The plot shown in Figure 10 shows that the relationship holds even when the magnitude of the force is increased by a factor of 100. In the plot shown in Figure 10, when the force is reduced by a factor of 100, the simulated SPL was determined to be approximately 35 dB. When the force is increased by a factor of 1000, the simulated SPL was determined to be approximately 135 dB. According to the above equation, a 100 dB increase in SPL would be expected, which corresponds to the increase determined by the vibro-acoustic FEA simulation shown in Figure 10.

[0227] The simulated sound pressure level associated with a given circumferential mode may be determined by simulating multiple resultant sound pressure levels corresponding to multiple frequency values ​​close to the natural frequency. That is, multiple frequency values ​​can be simulated near the natural frequency for each circumferential mode. The simulated frequency values ​​for each circumferential order can be primarily centered near the natural frequency of the corresponding mode shape (from 404). This allows for a better estimation of the SPL level when radial forces are applied due to dynamic phase currents, since the motor tends to vibrate near the natural frequency.

[0228] For each frequency value analyzed in sub-process 406, the acoustic near-field and far-field meshes can be adjusted in the vibro-acoustic FEA to maintain simulation accuracy and reduce computation time. The near-field thickness and element size at each simulated frequency can be defined based on the wavelength of that frequency.

[0229] 11 shows an example plot of simulated sound pressure levels for circumferential mode 0 of a switched reluctance machine with 18 stator poles and 12 rotor poles. The plot shown in FIG. 11 shows the simulated sound pressure levels when a unit force is applied. These simulated sound pressure levels can be stored (e.g., in a look-up table) for use in determining harmonic sound pressure levels in a given phase current waveform during the operating stage 415.

[0230] 12 shows an example plot of circumferential mode 0 radiation coefficient for a switched reluctance machine with 18 stator poles and 12 rotor poles. The radiation coefficient can also be generated using vibro-acoustic simulation of the switched reluctance machine.

[0231] 13 illustrates an example plot of simulated sound pressure levels for circumferential mode 6 of a switched reluctance machine with 18 stator poles and 12 rotor poles. The plot shown in FIG. 13 illustrates the simulated sound pressure levels when a unit force is applied. These simulated sound pressure levels may be stored (e.g., in a look-up table) for use in determining harmonic sound pressure levels in a given phase current waveform during the operating stage 415.

[0232] Figure 14 shows an example plot of the radiation coefficient of circumferential mode 6 for a switched reluctance machine with 18 stator poles and 12 rotor poles. The radiation coefficient can also be generated using vibro-acoustic simulation of the switched reluctance machine.

[0233] As can be seen from the plots of Figures 11-14, both the sound pressure level and the emissivity are dependent on the azimuthal order and frequency. The simulated sound pressure levels as shown in Figures 11 and 13 can be used in sub-process 411 to dynamically determine (e.g., at 340) the harmonic sound pressure levels in the potential phase current waveform.

[0234] The simulated sound pressure levels generated through implementation of the methods described herein were compared to sound pressure levels generated through vibro-acoustic FEA simulations using dynamic nodal forces.

[0235] 18 shows a comparison of the sound pressure levels generated for azimuthal modes 0 and 6 through implementation of the method described herein (i.e., using sub-process 406) with the sound pressure levels calculated by directly applying the dynamic currents calculated through sub-process 408 to an electromagnetic FEA model of the SRM. Nodal forces were calculated using the electromagnetic FEA model of the SRM and those nodal forces were applied directly to a vibro-acoustic FEA model of the SRM to calculate the sound pressure levels for the same modes. As shown in FIG. 18, the sound pressure levels calculated using implementation of sub-process 406 closely match the sound pressure levels calculated directly from the electromagnetic and vibro-acoustic FEA models using the dynamic current values.

[0236] As previously described herein, the sub-processes of the pre-processing stage 420 can be used to control the current applied to a switched reluctance motor drive to reduce acoustic noise. In particular, the pre-processing stage 420 has been described in the context of determining simulated sound pressure levels and electromagnetic characteristics for a switched reluctance machine having a prescribed geometry and magnetic pole configuration. However, the sub-processes 401-407 of the pre-processing stage 420 can also be applied to the design of motor geometry and assembly to provide a switched reluctance machine that allows for further acoustic noise reduction.

[0237] While this application has been described with reference to examples, it should be understood that the claims should not be limited by the embodiments set forth in the examples, but should be accorded the broadest interpretation consistent with the description as a whole.

[0238] All publications, patents, and patent applications are herein incorporated by reference in their entirety to the same extent as if each individual publication, patent, or patent application was specifically and individually indicated to be incorporated by reference in its entirety. In the event that any term in this application is found to have a different definition in a document incorporated by reference herein, the definition provided herein shall serve as the definition of such term.

Claims

1. 1. A method of controlling a switched reluctance machine, comprising: determining a plurality of potential phase current waveforms for corresponding phase coils of the switched reluctance machine; determining a cumulative sound pressure level of the switched reluctance machine for each potential phase current waveform, the cumulative sound pressure level being determined based on a plurality of harmonic sound pressure levels expected to result from the potential phase current waveform; identifying a desired phase current waveform as the potential phase current waveform associated with an optimal cumulative sound pressure level; applying the desired phase current waveform to the corresponding phase coil of the switched reluctance machine; A method comprising:

2. The desired phase current waveforms are determined using an iterative optimization process, each iteration of the iterative optimization process comprising: identifying iteration-specific potential phase current waveforms; determining a plurality of iteration-specific circumferential orders associated with the iteration-specific potential phase current waveforms; determining the magnitudes of a plurality of radial force harmonics associated with the repetition-specific potential phase current waveform; determining a plurality of iteration-specific forcing frequencies associated with the iteration-specific potential phase current waveforms; determining the cumulative sound pressure level in the iteration-specific potential phase current waveform using the plurality of iteration-specific circumferential orders, the plurality of iteration-specific forcing frequencies, and the magnitudes of the plurality of radial force harmonics; The method of claim 1 , comprising:

3. The cumulative sound pressure level is identifying a plurality of repeat-specific harmonic sound pressure levels, each repeat-specific harmonic sound pressure level being determined based on a combination of one of the repeat-specific circumferential orders, one of the repeat-specific forcing frequencies, and one of the radial force harmonic magnitudes; determining the cumulative sound pressure level as a sum of the plurality of iteration-specific harmonic sound pressure levels; The method of claim 2 , wherein the value is determined by

4. 4. The method of claim 3, wherein the specific harmonic sound pressure level for each iteration is determined using a look-up table defined based on simulated sound pressure levels associated with a particular circumferential order and a particular forcing frequency.

5. The lookup table is defined during a pre-processing stage, the pre-processing stage comprising: determining a plurality of dominant circumferential modes for the switched reluctance machine based on a geometry and pole configuration of the switched reluctance machine; For each dominant circumferential mode, determining a natural frequency associated with the dominant circumferential mode; determining the simulated sound pressure level associated with the dominant circumferential mode by simulating a resultant sound pressure level corresponding to at least one frequency value near the natural frequency using a vibro-acoustic model of the switched reluctance machine; storing the simulated sound pressure levels associated with the corresponding dominant circumferential modes and frequency values ​​in the look-up table; The method of claim 4, comprising the steps of:

6. The plurality of dominant circumferential modes are: determining a number of poles of the switched reluctance machine as a ratio between a number of stator poles of the switched reluctance machine and a number of phases of the switched reluctance machine; identifying the dominant circumferential mode according to the number of magnetic poles; The method of claim 5 , wherein the value is determined by

7. For each dominant circumferential mode, identifying higher circumferential order radial force harmonics that excite the dominant circumferential mode due to a sampling effect of the radial force density harmonics; determining the simulated sound pressure levels associated with the dominant circumferential modes to include sound pressures resulting from the identified higher circumferential order radial force harmonics; The method of claim 6 further comprising:

8. The step of determining the plurality of iteration-specific forcing frequencies comprises: determining a motor speed of the switched reluctance machine; determining a plurality of iteration-specific time orders associated with the iteration-specific potential phase current waveforms; determining the plurality of iteration-specific forcing frequencies based on the motor speed and the plurality of iteration-specific time orders; 8. The method of any one of claims 2 to 7, comprising:

9. The step of determining a time order specific to the plurality of iterations comprises: determining a radial force waveform associated with the iteration-specific potential phase current waveform; calculating the plurality of iteration-specific time orders by applying a Fast Fourier Transform to the radial force waveform; The method of claim 8, comprising:

10. determining a plurality of dominant time orders; determining a subset of dominant iteration-specific time orders corresponding to the plurality of dominant time orders; determining the plurality of iteration-specific forcing frequencies based on the motor speed and a subset of the dominant iteration-specific time orders; The method of claim 9 further comprising:

11. The method of claim 10 , wherein the plurality of dominant time orders are determined based on the geometry and pole configuration of the switched reluctance machine.

12. The step of determining a plurality of dominant time orders comprises: determining the number of strokes in one machine revolution as the product of the number of rotor poles of the switched reluctance machine and the number of phases of the switched reluctance machine; identifying the dominant time order according to the number of strokes; The method of claim 11 , comprising:

13. determining the plurality of iteration-specific circumferential orders by determining a radial force waveform associated with the iteration-specific potential phase current waveform; calculating the plurality of iteration-specific circumferential orders by applying a Fast Fourier Transform to the radial force waveform; The method of claim 2 , comprising:

14. determining the magnitudes of the plurality of radial force harmonics includes: determining a radial force waveform associated with the iteration-specific potential phase current waveform; calculating the magnitudes of the plurality of radial force harmonics by applying a Fast Fourier Transform to the radial force waveform; The method of claim 2 , comprising:

15. 15. The method of claim 13 or 14, wherein the magnitudes of the plurality of repeating characteristic circumferential orders and the plurality of radial force harmonics are simultaneously determined by applying the Fast Fourier Transform to the radial force waveform.

16. 15. The method of claim 13 or 14, wherein the radial force waveform is determined using a dynamic motor drive model based on a motor speed and a DC link voltage of the switched reluctance machine.

17. 15. The method of claim 13 or 14, wherein the radial force waveform is determined based on a motor speed, a phase voltage, and a phase current of the switched reluctance machine.

18. shaft, a rotor attached to the shaft and having a plurality of rotor salient poles; a stator having a plurality of stator protruding poles protruding from the stator toward the rotor poles; and a plurality of electrical coils wound around the stator poles, the plurality of electrical coils including a plurality of separate phase coils defining a plurality of phases of a switched reluctance machine; a switched reluctance machine comprising: Power supply and a power converter coupled to the power source and to the switched reluctance machine; Controller and Equipped with The controller determining a plurality of potential phase current waveforms for corresponding phase coils of the switched reluctance machine; determining a cumulative sound pressure level of the switched reluctance machine for each potential phase current waveform, the cumulative sound pressure level being determined based on a plurality of harmonic sound pressure levels expected to result from the potential phase current waveform; Identifying a desired phase current waveform as the potential phase current waveform associated with an optimal cumulative sound pressure level; applying the desired phase current waveforms to the corresponding phase coils of the switched reluctance machine using the power converter; The system is configured as follows:

19. The controller is configured to determine the desired phase current waveforms using an iterative optimization process, each iteration of the iterative optimization process comprising: identifying iteration-specific potential phase current waveforms; determining a plurality of iteration-specific circumferential orders associated with the iteration-specific potential phase current waveforms; determining the magnitudes of a plurality of radial force harmonics associated with the repetition-specific potential phase current waveform; determining a plurality of iteration-specific forcing frequencies associated with the iteration-specific potential phase current waveforms; determining the cumulative sound pressure level in the iteration-specific potential phase current waveform using the plurality of iteration-specific circumferential orders, the plurality of iteration-specific forcing frequencies, and the magnitudes of the plurality of radial force harmonics; 20. The system of claim 18, comprising:

20. The controller identifying a plurality of repeat-specific harmonic sound pressure levels, each repeat-specific harmonic sound pressure level being determined based on a combination of one of the repeat-specific circumferential orders, one of the repeat-specific forcing frequencies, and one of the radial force harmonic magnitudes; determining the cumulative sound pressure level as a sum of the plurality of iteration-specific harmonic sound pressure levels; 20. The system of claim 19, configured to determine the cumulative sound pressure level by:

21. 21. The system of claim 20, further comprising a non-transitory storage memory having stored therein a look-up table defined based on simulated sound pressure levels associated with particular circumferential orders and particular forcing frequencies, wherein the controller is configured to use the look-up table to determine each iteration-specific harmonic sound pressure level.

22. The lookup table is defined during a pre-processing stage, the pre-processing stage comprising: determining a plurality of dominant circumferential modes for the switched reluctance machine based on a geometry and pole configuration of the switched reluctance machine; For each dominant circumferential mode, determining a natural frequency associated with the dominant circumferential mode; determining the simulated sound pressure level associated with the dominant circumferential mode by simulating a resultant sound pressure level corresponding to at least one frequency value near the natural frequency using a vibro-acoustic model of the switched reluctance machine; storing the simulated sound pressure levels in association with the corresponding dominant circumferential modes and frequency values ​​in the look-up table; Steps and 22. The system of claim 21, comprising:

23. The plurality of dominant circumferential modes are: specifying a number of poles of the switched reluctance machine as a ratio between a number of stator poles of the switched reluctance machine and a number of phases of the switched reluctance machine; identifying the dominant circumferential mode according to the number of magnetic poles; 23. The system of claim 22, wherein the value is determined by:

24. For each dominant circumferential mode, identifying higher circumferential order radial force harmonics that excite the dominant circumferential mode due to a sampling effect of the radial force density harmonics; the simulated sound pressure levels associated with the dominant circumferential modes are determined to include sound pressures resulting from the identified higher circumferential order radial force harmonics.

24. The system of claim 23.

25. The controller determining a motor speed of the switched reluctance machine; determining a plurality of iteration-specific time orders associated with the iteration-specific potential phase current waveforms; determining the plurality of iteration-specific forcing frequencies based on the motor speed and the plurality of iteration-specific time orders; 25. The system of claim 19, configured to determine the plurality of iteration-specific forcing frequencies by:

26. The controller determining a radial force waveform associated with the repetition-specific potential phase current waveform; calculating the plurality of iteration-specific time orders by applying a fast Fourier transform to the radial force waveform; 26. The system of claim 25, configured to determine the iteration-specific time orders by:

27. The controller determining a plurality of dominant time orders; determining a subset of dominant iteration-specific time orders corresponding to the plurality of dominant time orders; determining the plurality of iteration-specific forcing frequencies based on the motor speed and a subset of the dominant iteration-specific time orders; 27. The system of claim 26, configured to:

28. 28. The system of claim 27, wherein the plurality of dominant time orders are predetermined based on the geometry and pole configuration of the switched reluctance machine.

29. The plurality of dominant time orders are: determining the number of strokes in one machine revolution as the product of the number of rotor poles of the switched reluctance machine and the number of phases of the switched reluctance machine; identifying the dominant time order according to the number of strokes; 29. The system of claim 28, wherein the predetermined value is determined by:

30. The controller determining a radial force waveform associated with the repetition-specific potential phase current waveform; calculating the plurality of iteration-specific circumferential orders by applying a Fast Fourier Transform to the radial force waveform; 20. The system of claim 19, configured to determine the plurality of iteration-specific circumferential orders by:

31. The controller determining a radial force waveform associated with the repetition-specific potential phase current waveform; calculating the magnitudes of the plurality of radial force harmonics by applying a fast Fourier transform to the radial force waveform; 20. The system of claim 19, configured to determine the magnitudes of the plurality of radial force harmonics by:

32. 32. The system of claim 30 or 31, wherein the controller is configured to simultaneously determine the plurality of repetitive characteristic circumferential orders and the magnitudes of the plurality of radial force harmonics by applying the Fast Fourier Transform to the radial force waveform.

33. 32. The system of claim 30 or 31, wherein the controller is configured to determine the radial force waveform using a dynamic motor drive model based on a motor speed and a DC link voltage of the switched reluctance machine.

34. 32. The system of claim 30 or 31, wherein the controller is configured to determine the radial force waveform based on a motor speed, a phase voltage, and a phase current of the switched reluctance machine.

35. 1. A non-transitory computer-readable medium storing computer-executable instructions that, when executed by a computer processor, cause the computer processor to perform a method of controlling a switched reluctance machine, the method comprising: determining a plurality of potential phase current waveforms for corresponding phase coils of the switched reluctance machine; determining a cumulative sound pressure level of the switched reluctance machine for each potential phase current waveform, the cumulative sound pressure level being determined based on a plurality of harmonic sound pressure levels expected to result from the potential phase current waveform; identifying a desired phase current waveform as the potential phase current waveform associated with an optimal cumulative sound pressure level; applying the desired phase current waveform to the corresponding phase coil of the switched reluctance machine; 1. A non-transitory computer-readable medium comprising:

36. 36. The non-transitory computer-readable medium of claim 35, wherein the method is defined according to any one of claims 2 to 7, 13, and 14.

Images