A Vibration and Noise Suppression Control Method for Permanent Magnet Synchronous Motors

By extracting real-time operating status data of permanent magnet synchronous motors and calculating compensation current commands using stator mechanical admittance models, and generating suppressive voltage waveforms using space vector pulse width modulation, the problem of electromagnetic vibration and noise suppression of permanent magnet synchronous motors under dynamic operating conditions is solved, achieving real-time and accurate suppression of electromagnetic vibration and noise and reduction of stator deformation.

CN123122376APending Publication Date: 2026-11-17JIANGMEN ZHENHUA MOTOR MFG CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611504606.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-09-24
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Existing technologies struggle to suppress electromagnetic vibration and noise in real time, accurately, and effectively, as the operating conditions of permanent magnet synchronous motors change dynamically, especially under wide speed and variable load conditions, where changes in electromagnetic excitation frequency and amplitude lead to a sudden increase in vibration and noise.

Method used

By acquiring real-time operating status data of permanent magnet synchronous motors, the characteristic spatial order electromagnetic force wave components that cause stator deformation are extracted. Multi-frequency compensation current commands are calculated based on the stator mechanical admittance model, and a suppressive voltage waveform is generated through space vector pulse width modulation to drive the motor, thereby achieving real-time reverse cancellation of electromagnetic vibration noise.

Benefits of technology

It achieves real-time and accurate suppression of electromagnetic vibration noise under dynamic changes in the operating conditions of permanent magnet synchronous motor, reduces stator forced deformation and its radiated noise, and takes into account the basic drive performance and control stability of the motor.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123122376A_ABST
    Figure CN123122376A_ABST
Patent Text Reader

Abstract

This invention relates to a vibration and noise suppression control method for permanent magnet synchronous motors (PMSMs), belonging to the field of PMSM technology. The method includes the following steps: acquiring real-time operating status data of the PMSM and extracting characteristic spatial order electromagnetic force wave components that cause stator deformation; calculating multi-frequency compensation current commands for generating reverse cancelling force waves based on the characteristic spatial order electromagnetic force wave components and a preset stator mechanical admittance model; injecting the multi-frequency compensation current commands into the basic current control loop, and generating a suppressive voltage waveform through space vector pulse width modulation to drive the PMSM. This method solves the technical problem of how to suppress electromagnetic vibration and noise that migrates with the operating conditions of a PMSM in real time, accurately, and effectively under dynamically changing operating conditions.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of permanent magnet synchronous motor technology, and in particular to a method for vibration and noise suppression control of permanent magnet synchronous motors. Background Technology

[0002] Permanent magnet synchronous motors (PMSMs) are characterized by high power density, high efficiency, and wide speed range, and are widely used in new energy vehicles, industrial servo systems, and high-end equipment. Existing technologies typically reduce electromagnetic vibration and radiated noise generated by the air gap magnetic field by optimizing motor slot-pole matching, using skewed rotor poles or stator slots, adjusting magnetic pole shape, and adding stator reinforcing ribs, vibration damping pads, and sound insulation structures. Some control schemes also reduce motor vibration and noise under typical operating conditions by improving current waveforms, reducing torque ripple, or avoiding specific resonant speeds. These schemes are generally configured based on structural parameters or typical operating conditions determined during the motor design phase, and drive control is implemented according to predetermined control parameters during motor operation.

[0003] However, when the speed and load of a permanent magnet synchronous motor change, the frequency and amplitude of the electromagnetic excitation also change, and different electromagnetic excitation components may couple with the stator's natural vibration modes under different operating conditions. Existing structural noise reduction measures and fixed control parameters set according to typical operating conditions are difficult to adjust specifically to changes in operating conditions, leading to the problem of sudden increases in local frequency vibration and noise during wide-speed, variable-load operation of the motor. Therefore, the urgent technical problem to be solved by existing technologies is: how to suppress electromagnetic vibration and noise that migrates with the operating conditions of a permanent magnet synchronous motor in real time, accurately and effectively. Summary of the Invention

[0004] The purpose of this invention is to at least partially solve one of the technical problems existing in the prior art.

[0005] To achieve the above objectives, the present invention provides a vibration and noise suppression control method for a permanent magnet synchronous motor, comprising: Acquire real-time operating status data of permanent magnet synchronous motor and extract characteristic spatial order electromagnetic force wave components that cause stator deformation; Based on the characteristic space order electromagnetic force wave components and the preset stator mechanical admittance model, the multi-frequency compensation current command for generating the reverse canceling force wave is calculated. The multi-frequency compensation current command is injected into the base current control loop, and a suppressive voltage waveform is generated by space vector pulse width modulation to drive the permanent magnet synchronous motor.

[0006] Furthermore, real-time operating status data of the permanent magnet synchronous motor is acquired, and characteristic spatial order electromagnetic force wave components causing stator deformation are extracted, including: The three-phase stator current and rotor position angle of the permanent magnet synchronous motor are collected as the real-time operating status data; The three-phase stator current is subjected to synchronous rotating coordinate transformation. The fundamental component is removed from the coordinate transformation result, and the remaining alternating harmonic components are subjected to adaptive notch filtering to obtain a broadband current residual signal. Based on the rotor position angle, the broadband current residual signal is subjected to order tracking analysis to separate specific harmonic components that change synchronously with the motor speed. Based on a preset current harmonic-electromagnetic force wave mapping relationship, the specific harmonic component is converted into the characteristic spatial order electromagnetic force wave component that causes stator deformation.

[0007] Furthermore, a synchronous rotating coordinate transformation is performed on the three-phase stator current, the fundamental component is removed from the coordinate transformation result, and the remaining alternating harmonic components are subjected to adaptive notch filtering to obtain a broadband current residual signal, including: The electrical angle is calculated using the rotor position angle, and the Parker transformation is performed on the three-phase stator current using the electrical angle as the phase reference to obtain the original AC and DC axis currents containing AC and DC axis DC components and alternating harmonic components. The DC component of the quadrature axis is extracted from the original quadrature axis current as the characterization value of the fundamental component, and the difference between the original quadrature axis current and the DC component of the quadrature axis is calculated to remove the fundamental component and obtain the alternating harmonic component; The alternating harmonic components are input into an adaptive notch filter to filter out the low-frequency pulsation interference remaining during the coordinate transformation process, thereby obtaining the pure broadband current residual signal.

[0008] Furthermore, the alternating harmonic components are input into an adaptive notch filter to filter out low-frequency pulsation interference remaining during the coordinate transformation process, thereby obtaining a pure broadband current residual signal, including: The rotor position angle is differentiated to obtain the real-time mechanical speed of the permanent magnet synchronous motor; Based on the real-time mechanical speed and the preset rotor eccentricity characteristic order, the dynamic center frequency of the low-frequency pulsation interference remaining during the coordinate transformation process is calculated. The dynamic center frequency is updated in the transfer function of the adaptive notch filter, and notch attenuation operation is performed on the alternating harmonic components with the dynamic center frequency as the stopband center to filter out the low-frequency pulsation interference contained in the alternating harmonic components. The output signal after filtering out the low-frequency pulsation interference is extracted as the pure broadband current residual signal.

[0009] Furthermore, based on the real-time mechanical speed and the preset rotor eccentricity characteristic order, the dynamic center frequency of the low-frequency pulsation interference remaining during the coordinate transformation process is calculated, including: The transient angular acceleration of the permanent magnet synchronous motor is obtained by performing a differential operation on the real-time mechanical rotation speed. Based on the preset rotor inertia identification model and the transient angular acceleration, the transient frequency offset compensation amount caused by the sudden change in rotational speed is calculated. Multiplying the real-time mechanical speed by the preset rotor eccentricity characteristic order yields the steady-state fundamental disturbance frequency. The steady-state basic interference frequency is superimposed and corrected with the transient frequency offset compensation amount to calculate the dynamic center frequency of the low-frequency pulsation interference remaining during the coordinate transformation process.

[0010] Furthermore, based on the characteristic space order electromagnetic force wave components and the preset stator mechanical admittance model, a multi-frequency compensation current command for generating the reverse cancelling force wave is calculated, including: Invert the amplitude of each frequency component in the characteristic space order electromagnetic force wave component to obtain the target cancellation force wave component corresponding to each frequency component. Each target cancelling force wave component is input into the preset stator mechanical admittance model to extract the mechanical resonance phase shift angle at each corresponding frequency; Based on the phase shift angle of each mechanical resonance, the phase of the target canceling force wave component at the corresponding frequency is corrected to obtain the corrected canceling force wave component corresponding to each frequency component. Each corrected and canceled force wave component is input into a preset force-current conversion matrix to calculate the compensation current command corresponding to each frequency component. The compensation current commands corresponding to each frequency component are superimposed to obtain the multi-frequency compensation current command used to generate the reverse canceling force wave.

[0011] Furthermore, the expression for calculating the mechanical resonance phase shift angle in the stator mechanical admittance model is as follows: ; In the formula, For the corresponding frequency The mechanical resonance phase shift angle under the given conditions, in rad; The frequency of the force wave component that the target cancels out, in Hz; The stator modal damping coefficient is pre-calibrated and its unit is N·s / m; The pre-calibrated stator modal stiffness is expressed in N / m. The mass of the pre-calibrated stator mode is expressed in kg.

[0012] Furthermore, each corrected and canceled force wave component is input into a preset force-current conversion matrix to calculate the compensation current command corresponding to each frequency component, including: The real-time AC and DC axis currents of the permanent magnet synchronous motor are obtained, and the AC and DC axis inductance saturation characterization value reflecting the degree of magnetic circuit saturation is calculated based on the real-time AC and DC axis currents. Based on the cross-axis and direct-axis inductance saturation characterization values, the cross-coupling coefficients in the preset reference force-current conversion matrix are nonlinearly interpolated and corrected to obtain the dynamic force-current conversion matrix. The compensation current command corresponding to each frequency component, including the quadrature axis compensation component and the direct axis compensation component, is calculated by performing matrix multiplication on each of the corrected and canceled force wave components and the dynamic force-current conversion matrix.

[0013] Furthermore, the multi-frequency compensation current command is injected into the base current control loop, and a suppressive voltage waveform is generated through space vector pulse width modulation to drive the permanent magnet synchronous motor, including: Based on the sampling period and algorithm execution period of the control system, the digital delay phase angle of the basic current control loop is calculated for each frequency component in the multi-frequency compensation current command. The digital delay phase angle is used to perform phase lead compensation on the corresponding frequency component in the multi-frequency compensation current command, and the compensated command is injected into the basic current control loop to obtain a composite voltage vector. The composite voltage vector is input to the space vector pulse width modulation module to generate the suppressive voltage waveform, and the suppressive voltage waveform is output to the inverter to drive the permanent magnet synchronous motor.

[0014] Further, the composite voltage vector is input to the space vector pulse width modulation module to generate the suppressive voltage waveform, and the suppressive voltage waveform is output to the inverter to drive the permanent magnet synchronous motor, including: Obtain the real-time DC bus voltage of the inverter, and construct a space vector hexagonal dynamic voltage boundary based on the real-time DC bus voltage; The composite voltage vector is decomposed into a fundamental driving voltage component and a harmonic compensation voltage component for vibration suppression. Determine whether the composite voltage vector's amplitude and phase composite trajectory exceeds the space vector hexagonal dynamic voltage boundary; If it exceeds the limit, then, with the principle of keeping the fundamental driving voltage component unchanged, the harmonic compensation voltage component is projected and limited along the residual margin profile of the space vector hexagonal dynamic voltage boundary to reconstruct a safe composite voltage vector. The safety composite voltage vector is input into the space vector pulse width modulation module to generate the suppressive voltage waveform, and the suppressive voltage waveform is output to the inverter to drive the permanent magnet synchronous motor.

[0015] This invention provides a vibration and noise suppression control method for permanent magnet synchronous motors, comprising: acquiring real-time operating status data of the permanent magnet synchronous motor and extracting characteristic spatial order electromagnetic force wave components that cause stator deformation; calculating multi-frequency compensation current commands for generating reverse cancelling force waves based on the characteristic spatial order electromagnetic force wave components and a preset stator mechanical admittance model; injecting the multi-frequency compensation current commands into the basic current control loop, and generating a suppressive voltage waveform through space vector pulse width modulation to drive the permanent magnet synchronous motor. This method solves the technical problem of how to suppress electromagnetic vibration and noise that migrates with the operating conditions of a permanent magnet synchronous motor in real time, accurately and effectively, under dynamic changes in operating conditions. It achieves dynamic tracking and reverse cancellation of multi-frequency electromagnetic force waves, reduces stator forced deformation and its radiated noise, and balances the basic driving performance and control stability of the motor. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a schematic diagram of the steps of a vibration and noise suppression control method for a permanent magnet synchronous motor in one embodiment of the present invention; Figure 2 This is one embodiment of the present invention. Figure 1 A schematic diagram of step S1.

[0018] The objectives, features, and advantages of this invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0019] The embodiments of the present invention are described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention. The step numbers in the following embodiments are set only for ease of explanation, and there is no limitation on the order between the steps. The execution order of each step in the embodiments can be adaptively adjusted according to the understanding of those skilled in the art.

[0020] The vibration and noise suppression control method for a permanent magnet synchronous motor according to an embodiment of the present invention will be described in detail below with reference to the accompanying drawings.

[0021] like Figures 1-2 The present invention provides a vibration and noise suppression control method for a permanent magnet synchronous motor, comprising: S1: Obtain real-time operating status data of the permanent magnet synchronous motor and extract the characteristic spatial order electromagnetic force wave components that cause stator deformation.

[0022] S11, collect the three-phase stator current and rotor position angle of the permanent magnet synchronous motor as the real-time operating status data.

[0023] Specifically, current sensors are installed on the power branches corresponding to the three-phase windings of the permanent magnet synchronous motor, and the three-phase stator current is synchronously collected within the sampling window when the inverter switching state is stable to reduce the interference of switching transients on the sampling results. Before acquisition, bias correction and amplitude conversion are performed based on the sensor zero-point output and calibration gain, and abnormal pulses are then limited to obtain three-phase stator currents with uniform timestamps. The rotor mechanical position is detected by a rotary transformer or encoder, and the sensor output signal is decoded, zero-position compensated, and angle continuous processing is performed to obtain the rotor position angle. Subsequently, the rotor position angle and the three-phase stator current are aligned frame by frame according to the timestamp to form real-time operating status data. In this embodiment, the above scheme can improve the synchronization of current and rotor position acquisition, providing a reliable data foundation for subsequent electromagnetic force wave component extraction.

[0024] S12, perform synchronous rotating coordinate transformation on the three-phase stator current, remove the fundamental component from the coordinate transformation result, and perform adaptive notch filtering on the remaining alternating harmonic components to obtain the broadband current residual signal.

[0025] S121, calculate the electrical angle using the rotor position angle, and perform Parker transformation on the three-phase stator current using the electrical angle as the phase reference to obtain the original AC and DC axis current containing AC and DC axis DC components and alternating harmonic components; S122, extract the DC component of the quadrature axis from the original quadrature axis current as the characterization value of the fundamental component, and perform a difference operation between the original quadrature axis current and the DC component of the quadrature axis to strip away the fundamental component and obtain the alternating harmonic component.

[0026] Specifically, in practice, the rotor position angle of the permanent magnet synchronous motor is read, and combined with the number of pole pairs and the calibration zero position of the rotor position sensor, the mechanical position is converted into electrical angles. The converted electrical angles are cyclically normalized over one electrical cycle to eliminate angle accumulation caused by continuous rotor rotation. Subsequently, using the electrical angles as the phase reference of the synchronous rotating coordinate system, the three-phase stator currents at the same sampling moment are projected onto the quadrature and direct axes to obtain the original quadrature and direct axis currents. When the electrical angle tracking is accurate, the fundamental positive sequence component of the three-phase stator current remains relatively stationary with the synchronous rotating coordinate system, and therefore appears as slowly changing quadrature and direct axis DC components in the original quadrature and direct axis currents. The harmonic components generated by inverter nonlinearity, air gap magnetic field distortion, and current modulation appear as alternating harmonic components that change with time. Sliding extraction channels are set for the original quadrature axis current and the original direct axis current, and the sampling window is updated according to a complete electrical cycle. The data within the window is averaged to obtain the corresponding quadrature axis DC components and direct axis DC components, which together serve as the characterization value of the fundamental component. The quadrature-axis DC component is then subtracted from the original quadrature-axis current at each sampling time, and the direct-axis DC component is subtracted from the original direct-axis current. The difference between the two results constitutes the alternating harmonic component. The window length is adjusted according to the rate of change of the electrical angle and always covers the complete electrical cycle to avoid deviations in DC component extraction caused by changes in rotational speed. In this embodiment, the above scheme can separate the fundamental component and the alternating harmonic component in a synchronously rotating coordinate system, providing accurate input for subsequent identification of electromagnetic force wave components of the characteristic space order.

[0027] S123, the alternating harmonic components are input into an adaptive notch filter to filter out the low-frequency pulsation interference remaining during the coordinate transformation process, thereby obtaining the pure broadband current residual signal.

[0028] S1231, perform differential calculation on the rotor position angle to obtain the real-time mechanical speed of the permanent magnet synchronous motor.

[0029] Specifically, in implementation, the rotor position angle is continuously read according to a unified sampling period. Data crossing the angular period boundary is expanded based on the rotation direction of adjacent sampling points to eliminate abrupt changes caused by the rotor position angle jumping back from its maximum value to zero. The angle increment between the current sampling point and the previous sampling point is calculated, and discrete differentiation is performed based on the corresponding sampling time interval to obtain the mechanical angular velocity. Then, according to the time unit and rotational speed representation requirements, the mechanical angular velocity is converted into real-time mechanical rotational speed. To address rotational speed fluctuations that may be caused by quantization errors in the position sensor, the differentiation results are smoothed with low latency while preserving the rotational speed rise / fall trend and rotational direction. In this embodiment, the above scheme can avoid sudden changes in rotational speed caused by angle retracement and sampling noise, improving the continuity and accuracy of real-time mechanical rotational speed.

[0030] S1232, Based on the real-time mechanical speed and the preset rotor eccentricity characteristic order, calculate the dynamic center frequency of the low-frequency pulsation interference remaining during the coordinate transformation process.

[0031] S12321, perform differential calculation on the real-time mechanical rotation speed to obtain the transient angular acceleration of the permanent magnet synchronous motor; S12322, Based on the preset rotor inertia identification model and the transient angular acceleration, calculate the transient frequency offset compensation amount caused by the sudden change in rotational speed; S12323, Multiply the real-time mechanical speed by the preset rotor eccentricity characteristic order to obtain the steady-state basic disturbance frequency; S12324, The steady-state basic interference frequency and the transient frequency offset compensation are superimposed and corrected to calculate the dynamic center frequency of the low-frequency pulsation interference remaining during the coordinate transformation process.

[0032] Specifically, in practice, following the same sampling sequence as the real-time mechanical rotation speed, the change in rotation speed between the current sampling time and the previous sampling time is calculated. Then, discrete differentiation is performed using the corresponding sampling time interval as a reference to obtain the transient angular acceleration. To avoid amplifying the noise from the rotation speed quantization after differentiation, the real-time mechanical rotation speed is first smoothed within a local time window that does not cross the range of sudden speed changes before performing the differentiation operation. The transient angular acceleration retains its positive and negative directions, with positive values ​​representing acceleration and negative values ​​representing deceleration.

[0033] The preset rotor inertia identification model is established through the torque and speed responses during the motor calibration phase. The model records the equivalent rotor inertia, equivalent mechanical damping, and response delay of the speed detection link of the permanent magnet synchronous motor. During operation, transient angular acceleration is input into the rotor inertia identification model. The time scale of the mechanical response is determined by the equivalent rotor inertia and equivalent mechanical damping, and the effective compensation duration is determined by combining the response delay. Then, based on the speed change trend within the effective compensation duration, the transient angular acceleration is converted into a directional transient speed change, and then into a transient frequency offset compensation amount expressed in Hertz.

[0034] When the motor accelerates, the transient frequency offset compensation is positive; when the motor decelerates, it is negative; when the transient angular acceleration approaches zero, the compensation synchronously decays to zero. Then, the real-time mechanical speed is uniformly converted to revolutions per second (RPM) to avoid dimensional inconsistencies caused by direct calculation between RPM and Hertz. This is then multiplied by a dimensionless preset rotor eccentricity characteristic order to obtain the steady-state fundamental interference frequency in Hertz. The rotor eccentricity characteristic order is predetermined based on the eccentricity type of the permanent magnet synchronous motor and the offline spectrum calibration results, and is used to characterize the harmonic relationship between low-frequency pulsating interference and the mechanical rotation frequency.

[0035] The steady-state base interference frequency is superimposed and corrected according to the positive and negative directions of the transient frequency offset compensation, and the correction result is limited to an effective range above zero frequency and below the Nyquist frequency of the sampling system. The resulting frequency is the dynamic center frequency of the low-frequency pulsating interference. In this embodiment, the above scheme enables the dynamic center frequency to continuously shift with the acceleration and deceleration of the motor, improving the accuracy of low-frequency pulsating interference localization.

[0036] S1233, update the dynamic center frequency to the transfer function of the adaptive notch filter, and perform notch attenuation operation on the alternating harmonic component with the dynamic center frequency as the stopband center to filter out the low-frequency pulsation interference contained in the alternating harmonic component. S1234, extract the output signal after filtering out the low-frequency pulsation interference, as the pure broadband current residual signal.

[0037] Specifically, an adaptive notch filter is constructed using a second-order infinite impulse response structure, and the dynamic center frequency is converted into a digital domain angular frequency according to the sampling frequency of the alternating harmonic components. The phase positions of the zeros and poles in the transfer function are synchronously adjusted using this digital domain angular frequency, so that the center of the notch stopband always follows the frequency shift of the low-frequency pulsating interference; the pole radius is determined according to the preset stopband width, which covers the finite frequency broadening caused by speed fluctuations, while avoiding adjacent broadband components to be retained.

[0038] To prevent sudden changes in the dynamic center frequency from causing a jump in the filter coefficients, parameter updates are initiated only when the difference in center frequencies between adjacent control cycles exceeds the update threshold, and a continuous transition is made between the old and new parameters, keeping the internal state of the filter constant.

[0039] The quadrature-axis and direct-axis components of the alternating harmonic components are input into an adaptive notch filter with identical parameters. This filter attenuates low-frequency pulsating interference near the dynamic center frequency, while allowing broadband harmonic components outside the stopband to pass through. After the two filtering channels have completed data processing at the same sampling time, the corresponding outputs are extracted and recombined according to the original timestamps to form the quadrature-axis and direct-axis current residuals after filtering out low-frequency pulsating interference. This residual signal serves as the pure broadband current residual signal.

[0040] In this embodiment, the above scheme can track and suppress low-frequency pulsation interference migrating under variable speed conditions, while retaining broadband current information for electromagnetic force wave analysis.

[0041] S13, combined with the rotor position angle, perform order tracking analysis on the broadband current residual signal to separate specific harmonic components that change synchronously with the motor speed; S14, based on the preset current harmonic-electromagnetic force wave mapping relationship, the specific subharmonic component is converted into the characteristic spatial order electromagnetic force wave component that causes stator deformation.

[0042] Specifically, the continuously acquired rotor position angles are first periodically expanded to form a monotonically continuous mechanical angle sequence. Then, based on this sequence, the broadband current residual signal is converted from equal time interval sampling to equal rotation angle interval sampling, so that the frequency drift caused by speed increase or decrease is expressed as a fixed order in the angle domain.

[0043] Order spectrum analysis was performed on the resampled quadrature-axis current residuals and direct-axis current residuals to obtain the amplitude and phase corresponding to each order, and the bandwidth was selected according to the preset target order range. Spectral components that fall within the corresponding order bandwidth and change synchronously with the motor speed within the continuous analysis window were retained, and then the signal was reconstructed based on their amplitude and phase to obtain the specific harmonic components.

[0044] The preset current harmonic-electromagnetic force wave mapping relationship is established in advance by the slot pole coordination, winding distribution, air gap magnetic field and stator structure parameters of the permanent magnet synchronous motor. The spatial order, amplitude gain and phase correction of the electromagnetic force wave corresponding to different current harmonic orders are recorded.

[0045] The order, amplitude, and phase of a specific harmonic component are input into the mapping relationship to match the corresponding spatial order. The gain of the amplitude is converted and the phase is synchronously corrected. When multiple specific harmonic components act on the same spatial order, they are vector synthesized according to the phase relationship to obtain the characteristic spatial order electromagnetic force wave component that causes stator deformation.

[0046] In this embodiment, the above-mentioned scheme can stably separate the target current harmonics under variable speed conditions and accurately characterize the electromagnetic force waves related to stator deformation.

[0047] S2, based on the characteristic space order electromagnetic force wave components and the preset stator mechanical admittance model, calculate the multi-frequency compensation current command used to generate the reverse canceling force wave.

[0048] S21, the amplitude of each frequency component in the characteristic space order electromagnetic force wave component is inverted to obtain the target cancellation force wave component corresponding to each frequency component.

[0049] Specifically, in implementation, the characteristic spatial order electromagnetic force wave components are read item by item according to frequency and spatial order. The center frequency, spatial order, and absolute value of amplitude of each frequency component remain unchanged, and their signed amplitudes are adjusted to the opposite direction. For frequency components represented by amplitude and phase, inverting the amplitude is equivalent to adding half a cycle to the original phase, so that the generated component is in the same frequency, spatial order, and opposite phase to the original electromagnetic force wave. After the conversion is completed, the reverse components are combined according to frequency order to obtain the corresponding target canceling force wave components. In this embodiment, the above scheme can clearly define the frequency, spatial order, and reverse relationship of each target canceling force wave component, providing a target basis for subsequent compensation current calculation.

[0050] S22, input the target cancellation force wave components into the preset stator mechanical admittance model, and extract the mechanical resonance phase shift angle at each corresponding frequency.

[0051] The expression for calculating the mechanical resonance phase shift angle in the stator mechanical admittance model is as follows: ; S221, where, For the corresponding frequency The mechanical resonance phase shift angle under the given conditions, in rad; The frequency of the force wave component that the target cancels out, in Hz; The stator modal damping coefficient is pre-calibrated and its unit is N·s / m; The pre-calibrated stator modal stiffness is expressed in N / m. The mass of the pre-calibrated stator mode is expressed in kg.

[0052] Specifically, This represents the dynamic damping term of the stator at the target frequency, used to characterize the degree of energy dissipation caused by structural damping during stator vibration. As the frequency of the target canceling force wave component increases... As the dynamic damping term increases, the influence of damping on the stator vibration phase also increases. This represents the inertial stiffness term resulting from the stator modal mass at the corresponding frequency. This represents the dynamic stiffness term formed by the combined effect of stator modal stiffness and inertial stiffness. Both the dynamic damping term and the dynamic stiffness term are in N / m, and their ratio is dimensionless. After arctangent calculation, the mechanical resonance phase shift angle in rad is obtained.

[0053] When the frequency of the target canceling force wave component is lower than the natural frequency of the corresponding stator mode, the dynamic stiffness term is positive and mainly determined by the stator modal stiffness, resulting in a relatively small mechanical resonance phase shift angle. As the frequency approaches the natural frequency, the inertial stiffness term gradually approaches the stator modal stiffness, causing the dynamic stiffness term to approach zero, and the mechanical resonance phase shift angle to approach half a circumference angle, indicating a significant resonant phase lag in the stator mechanical response. When the frequency is higher than the natural frequency, the inertial stiffness term dominates, the dynamic stiffness term turns negative, and the mechanical response enters the inertial control region. To avoid errors in phase quadrant determination caused by changes in the sign of the dynamic stiffness term, quadrant correction is performed by combining the signs of the dynamic damping and dynamic stiffness terms during actual calculations, thereby obtaining continuous mechanical resonance phase shift angles.

[0054] S23, based on the phase shift angle of each mechanical resonance, the phase of the target canceling force wave component at the corresponding frequency is corrected to obtain the corrected canceling force wave component corresponding to each frequency component.

[0055] Specifically, each mechanical resonance phase shift angle is matched one by one with the target canceling force wave component of the corresponding frequency according to the frequency index. Using the current phase of the target canceling force wave component as a reference, phase compensation is applied according to the hysteresis direction of the stator mechanical response relative to the electromagnetic excitation. When the mechanical response exhibits phase hysteresis, the target canceling force wave component is corrected in advance according to the corresponding mechanical resonance phase shift angle; otherwise, it is corrected in hysteresis, while maintaining its frequency, spatial order, and amplitude unchanged. The corrected components are combined in the original frequency order to obtain the corrected canceling force wave component. In this embodiment, the above scheme can compensate for the phase shift caused by stator mechanical resonance, ensuring that the canceling effect maintains an anti-phase relationship at the stator deformation response end.

[0056] S24: Input each corrected and canceled force wave component into the preset force-current conversion matrix to calculate the compensation current command corresponding to each frequency component.

[0057] S241, Obtain the real-time AC and DC axis current of the permanent magnet synchronous motor, and calculate the AC and DC axis inductance saturation characterization value reflecting the degree of magnetic circuit saturation based on the real-time AC and DC axis current; S242, Based on the cross-axis and direct-axis inductance saturation characterization values, nonlinear interpolation correction is performed on the cross-coupling coefficients in the preset reference force-current conversion matrix to obtain the dynamic force-current conversion matrix; S243, perform matrix multiplication on each of the corrected and canceled force wave components and the dynamic force-current conversion matrix to calculate the compensation current command corresponding to each frequency component, which includes the quadrature axis compensation component and the direct axis compensation component.

[0058] Specifically, the real-time quadrature-axis current and real-time direct-axis current corresponding to the current control cycle are read from the basic current control loop, and the two currents are smoothed with low delay to eliminate the influence of switching ripple on the judgment of magnetic circuit saturation state. A two-dimensional data table is established in advance through the magnetization characteristics calibration of the permanent magnet synchronous motor, relating the quadrature-axis current, direct-axis current, quadrature-axis incremental inductance, and direct-axis incremental inductance. The incremental inductance represents the flux linkage change rate corresponding to the current change near the operating point. During operation, the current operating point is determined by the real-time quadrature-axis and direct-axis currents. The incremental inductance of adjacent calibration nodes is read from the two-dimensional data table, and the current quadrature-axis incremental inductance and direct-axis incremental inductance are obtained using monotonic two-dimensional interpolation. The current incremental inductance is compared with the reference inductance in the corresponding unsaturated state. The greater the decrease in inductance, the higher the saturation degree of the corresponding magnetic circuit, thus forming dimensionless quadrature-axis inductance saturation characterization values ​​and direct-axis inductance saturation characterization values. For operating points exceeding the calibration data boundaries, the value is taken from the closest boundary node to avoid extrapolation leading to saturation distortion.

[0059] The preset reference force-current conversion matrix is ​​established based on the motor slot-pole structure, winding connection method, and electromagnetic field calibration results of the unsaturated operating region. Each row of the matrix corresponds to the quadrature-axis compensation component and the direct-axis compensation component, respectively, and each column corresponds to the correction and cancellation force wave components of different spatial orders. The matrix elements represent the current conversion coefficients required to generate a unit force wave amplitude. Considering the mutual influence between the quadrature-axis magnetic circuit and the direct-axis magnetic circuit in the saturated state, a nonlinear correction data surface is pre-established for each cross-coupling coefficient, which varies with the quadrature-axis inductance saturation characterization value and the direct-axis inductance saturation characterization value. The current two inductance saturation characterization values ​​are located on this data surface, and two-dimensional piecewise interpolation is performed on surrounding calibration nodes to obtain the cross-coupling coefficients at the current operating point, and the corresponding elements in the reference force-current conversion matrix are replaced. Matrix elements not affected by cross-saturation retain the reference values, thereby generating a dynamic force-current conversion matrix. The matrix update is synchronized with the current sampling period, and the coefficient changes of adjacent periods are continuously processed to prevent command jumps when the operating point crosses the interpolation interval.

[0060] Subsequently, the corrected and canceled force wave components are arranged into force wave input vectors according to frequency and spatial order, while their amplitude, phase, and direction are preserved. For each frequency, the force wave input vector is multiplied by the dynamic force-current conversion matrix, with the matrix conversion coefficients using the current dimension corresponding to each unit force wave amplitude, thus ensuring the output is a current quantity. After the calculation, the quadrature-axis compensation components and direct-axis compensation components that cooperate at the same frequency are obtained, and the frequency and phase information corresponding to each component is preserved, combining to form the compensation current command corresponding to each frequency component. In this embodiment, the above scheme can correct the quadrature-axis and direct-axis coupling relationship based on the real-time magnetic circuit saturation degree, improving the matching accuracy between the compensation current command and the target canceled force wave.

[0061] S25, superimpose the compensation current commands corresponding to each frequency component to obtain the multi-frequency compensation current command used to generate the reverse canceling force wave.

[0062] Specifically, based on the frequency, amplitude, and phase carried by each compensation current command, corresponding quadrature-axis compensation current sequences and direct-axis compensation current sequences are generated under a unified control clock to ensure that different frequency components maintain a predetermined phase relationship at the same sampling time. Subsequently, the quadrature-axis compensation components of each frequency are superimposed point by point, and the direct-axis compensation components of each frequency are superimposed point by point to form a quadrature-direct-axis multi-frequency composite current. After superposition, the peak value of the composite current is verified; if it exceeds the preset compensation current limit, all components are scaled proportionally to maintain the frequency composition and phase relationship, ultimately obtaining the multi-frequency compensation current command. In this embodiment, the above scheme can coordinate the output of compensation currents of multiple frequencies and avoid the superimposed peak value from exceeding the allowable range of the motor.

[0063] S3, the multi-frequency compensation current command is injected into the basic current control loop, and a suppressive voltage waveform is generated by space vector pulse width modulation to drive the permanent magnet synchronous motor.

[0064] S31, based on the sampling period and algorithm execution period of the control system, calculate the digital delay phase angle of the basic current control loop for each frequency component in the multi-frequency compensation current command; S32, the digital delay phase angle is used to perform phase advance compensation on the corresponding frequency component in the multi-frequency compensation current command, and the compensated command is injected into the basic current control loop to obtain a composite voltage vector.

[0065] Specifically, in implementation, the current sampling period of the control system and the algorithm execution period required to complete one compensation calculation and update the control command are read to determine the equivalent digital delay from the current sampling moment to the actual entry of the compensation command into the basic current control loop. The equivalent digital delay includes the waiting time caused by sampling and holding and the command update time caused by algorithm calculation. The order of these two times is calibrated using the controller timestamp to avoid double counting of overlapping periods. For each frequency component in the multi-frequency compensation current command, the digital delay phase angle is calculated based on the proportion of the period experienced by that frequency within the equivalent digital delay; the higher the frequency, the greater the phase lag corresponding to the same digital delay. Subsequently, while keeping the amplitude of each frequency component constant, the corresponding phase is shifted forward according to the digital delay phase angle, so that the compensation current command returns to the target phase after reaching the actual control action moment. The quadrature-axis compensation component and the direct-axis compensation component that have completed phase lead compensation are superimposed on the quadrature-axis current setpoint and the direct-axis current setpoint of the basic current control loop, respectively, to form a composite current setpoint. The basic current control loop adjusts based on the deviation between the composite current setpoint and the real-time quadrature and direct-axis currents, and generates quadrature-axis voltage commands and direct-axis voltage commands in conjunction with the quadrature-direct-axis decoupling term. These two commands together constitute the composite voltage vector. In this embodiment, the above scheme can compensate for the frequency-dependent phase lag caused by the digital control link, improving the phase accuracy of the reverse cancelling force wave.

[0066] S33, the composite voltage vector is input to the space vector pulse width modulation module to generate the suppressive voltage waveform, and the suppressive voltage waveform is output to the inverter to drive the permanent magnet synchronous motor.

[0067] S331, Obtain the real-time DC bus voltage of the inverter, and construct a space vector hexagonal dynamic voltage boundary based on the real-time DC bus voltage; S332, decompose the composite voltage vector into a fundamental driving voltage component and a harmonic compensation voltage component for vibration suppression; S333, determine whether the composite voltage vector's amplitude and phase synthesis trajectory exceeds the space vector hexagonal dynamic voltage boundary; S334, if it exceeds the limit, then based on the principle of keeping the fundamental driving voltage component unchanged, the harmonic compensation voltage component is projected and limited along the residual margin profile of the space vector hexagonal dynamic voltage boundary to reconstruct a safe composite voltage vector. S335, the safe composite voltage vector is input to the space vector pulse width modulation module to generate the suppressive voltage waveform, and the suppressive voltage waveform is output to the inverter to drive the permanent magnet synchronous motor.

[0068] Specifically, the real-time DC bus voltage between the positive and negative DC buses of the inverter is acquired in each control cycle, and the sampled values ​​are smoothed briefly to suppress switching ripple while retaining the bus voltage changes caused by load surges. Based on the voltage vector directions corresponding to the six effective switching states of the two-level inverter in the stationary coordinate system, the positions of the six vertices are determined using the real-time DC bus voltage. Simultaneously, the on-state voltage drop of the power devices, dead-time effects, and reserved modulation safety margins are deducted, and adjacent vertices are connected to form a dynamic hexagonal space vector voltage boundary that scales synchronously with the real-time DC bus voltage. When the DC bus voltage decreases, this boundary contracts inward; when the DC bus voltage increases, the boundary expands accordingly, but does not exceed the inverter's allowable modulation range.

[0069] The composite voltage vector consists of the fundamental drive voltage component generated by the basic current control loop and the harmonic compensation voltage component generated by the compensation current regulation. Based on the rotor position angle, the two voltage components are transformed from the quadrature-direction rotating coordinate system to the same stationary coordinate system, and phasors are superimposed according to a unified timestamp to preserve their respective amplitude and phase relationships. The fundamental drive voltage component is responsible for the torque and flux linkage control of the permanent magnet synchronous motor, while the harmonic compensation voltage component is responsible for generating the reverse canceling force wave. The two components remain distinguishable in their origins during subsequent limiting. The spatial vector sector where the composite voltage vector currently resides is then identified, and its composite trajectory is compared with the two hexagonal boundaries corresponding to that sector. When the trajectory remains within each boundary, it indicates no overmodulation risk, and the composite voltage vector remains unchanged.

[0070] When the composite voltage vector's trajectory exceeds the hexagonal dynamic voltage boundary of the space vector, first confirm that the fundamental driving voltage component is within the boundary. Then, using the endpoint of this fundamental driving voltage component as the starting point, construct the remaining voltage margin profile after deducting the fundamental occupancy from the hexagonal boundary. Search for the intersection point between the harmonic compensation voltage component and the remaining voltage margin profile along the original direction of the harmonic compensation voltage component, and determine the maximum allowable harmonic compensation amplitude using this intersection point. The portion exceeding the maximum amplitude is compressed, while the portion not exceeding it remains unchanged. Amplification only applies to the harmonic compensation voltage component and does not change the fundamental driving voltage component. When the fundamental driving voltage component approaches the boundary, the available margin of the harmonic compensation voltage component decreases synchronously; when the fundamental driving voltage component reaches the boundary, the harmonic compensation voltage margin in the corresponding direction drops to zero. Re-superimpose the limited harmonic compensation voltage component with the original fundamental driving voltage component to obtain a safe composite voltage vector located within the hexagonal dynamic voltage boundary of the space vector.

[0071] Finally, the safety composite voltage vector is input to the space vector pulse width modulation module. Based on its sector, two adjacent effective switching vectors and a zero vector are selected. The activation time of each switching vector is allocated according to the projection relationship of the safety composite voltage vector within one pulse width modulation cycle, and a symmetrical switching sequence is used to generate inverter bridge arm drive pulses. The inverter modulates the real-time DC bus voltage into a suppressive voltage waveform based on the drive pulses and applies it to the three-phase stator windings of the permanent magnet synchronous motor. The above projection limiting is a continuous voltage vector constraint, while space vector pulse width modulation is a discrete switching timing generation; the two are executed sequentially. In this embodiment, the above scheme can ensure the fundamental wave drive requirements while utilizing the remaining voltage margin to output harmonic compensation voltage, avoiding overmodulation and achieving vibration and noise suppression.

Claims

1. A permanent magnet synchronous motor vibration noise suppression control method, characterized by, include: Acquire real-time operating status data of permanent magnet synchronous motor and extract characteristic spatial order electromagnetic force wave components that cause stator deformation; Based on the characteristic space order electromagnetic force wave components and the preset stator mechanical admittance model, the multi-frequency compensation current command for generating the reverse canceling force wave is calculated. The multi-frequency compensation current command is injected into the base current control loop, and a suppressive voltage waveform is generated by space vector pulse width modulation to drive the permanent magnet synchronous motor.

2. The vibration noise suppression control method of a permanent-magnet synchronous motor according to claim 1, characterized by, Acquire real-time operating status data of the permanent magnet synchronous motor and extract the characteristic spatial order electromagnetic force wave components that cause stator deformation, including: The three-phase stator current and rotor position angle of the permanent magnet synchronous motor are collected as the real-time operating status data; The three-phase stator current is subjected to synchronous rotating coordinate transformation. The fundamental component is removed from the coordinate transformation result, and the remaining alternating harmonic components are subjected to adaptive notch filtering to obtain a broadband current residual signal. Based on the rotor position angle, the broadband current residual signal is subjected to order tracking analysis to separate specific harmonic components that change synchronously with the motor speed. Based on a preset current harmonic-electromagnetic force wave mapping relationship, the specific harmonic component is converted into the characteristic spatial order electromagnetic force wave component that causes stator deformation.

3. The vibration noise suppression control method of a permanent-magnet synchronous motor according to claim 2, characterized by, The three-phase stator current is subjected to synchronous rotating coordinate transformation. The fundamental component is extracted from the coordinate transformation result, and the remaining alternating harmonic components are subjected to adaptive notch filtering to obtain a broadband current residual signal, including: The electrical angle is calculated using the rotor position angle, and the Parker transformation is performed on the three-phase stator current using the electrical angle as the phase reference to obtain the original AC and DC axis currents containing AC and DC axis DC components and alternating harmonic components. The DC component of the quadrature axis is extracted from the original quadrature axis current as the characterization value of the fundamental component, and the difference between the original quadrature axis current and the DC component of the quadrature axis is calculated to remove the fundamental component and obtain the alternating harmonic component; The alternating harmonic components are input into an adaptive notch filter to filter out the low-frequency pulsation interference remaining during the coordinate transformation process, thereby obtaining the pure broadband current residual signal.

4. The vibration noise suppression control method of a permanent-magnet synchronous motor according to claim 3, characterized by, The alternating harmonic components are input into an adaptive notch filter to filter out low-frequency pulsation interference remaining during the coordinate transformation process, thereby obtaining the pure broadband current residual signal, including: The rotor position angle is differentiated to obtain the real-time mechanical speed of the permanent magnet synchronous motor; Based on the real-time mechanical speed and the preset rotor eccentricity characteristic order, the dynamic center frequency of the low-frequency pulsation interference remaining during the coordinate transformation process is calculated. The dynamic center frequency is updated in the transfer function of the adaptive notch filter, and notch attenuation operation is performed on the alternating harmonic components with the dynamic center frequency as the stopband center to filter out the low-frequency pulsation interference contained in the alternating harmonic components. The output signal after filtering out the low-frequency pulsation interference is extracted as the pure broadband current residual signal.

5. The vibration noise suppression control method of a permanent-magnet synchronous motor according to claim 4, characterized by, Based on the real-time mechanical speed and the preset rotor eccentricity characteristic order, the dynamic center frequency of the low-frequency pulsation interference remaining during the coordinate transformation is calculated, including: The transient angular acceleration of the permanent magnet synchronous motor is obtained by performing a differential operation on the real-time mechanical rotation speed. Based on the preset rotor inertia identification model and the transient angular acceleration, the transient frequency offset compensation amount caused by the sudden change in rotational speed is calculated. Multiplying the real-time mechanical speed by the preset rotor eccentricity characteristic order yields the steady-state fundamental disturbance frequency. The steady-state basic interference frequency is superimposed and corrected with the transient frequency offset compensation amount to calculate the dynamic center frequency of the low-frequency pulsation interference remaining during the coordinate transformation process.

6. The vibration noise suppression control method of a permanent-magnet synchronous motor according to claim 1, characterized by, Based on the characteristic spatial order electromagnetic force wave components and the preset stator mechanical admittance model, the multi-frequency compensation current command for generating the reverse cancelling force wave is calculated, including: Invert the amplitude of each frequency component in the characteristic space order electromagnetic force wave component to obtain the target cancellation force wave component corresponding to each frequency component. Each target cancelling force wave component is input into the preset stator mechanical admittance model to extract the mechanical resonance phase shift angle at each corresponding frequency; Based on the phase shift angle of each mechanical resonance, the phase of the target canceling force wave component at the corresponding frequency is corrected to obtain the corrected canceling force wave component corresponding to each frequency component. Each corrected and canceled force wave component is input into a preset force-current conversion matrix to calculate the compensation current command corresponding to each frequency component. The compensation current commands corresponding to each frequency component are superimposed to obtain the multi-frequency compensation current command used to generate the reverse canceling force wave.

7. The vibration noise suppression control method of a permanent-magnet synchronous motor according to claim 6, characterized by, The expression for calculating the mechanical resonance phase shift angle in the stator mechanical admittance model is as follows: ; In the formula, is the phase shift angle of the mechanical resonance at the corresponding frequency , unit: rad; is the frequency of the target cancellation force wave component, unit: Hz; is the pre-calibrated stator modal damping coefficient, unit: N·s / m; is the pre-calibrated stator modal stiffness, unit: N / m; is the pre-calibrated stator modal mass, unit: kg.

8. The vibration noise suppression control method of a permanent-magnet synchronous motor according to claim 6, characterized by, Each corrected and canceled force wave component is input into a preset force-current conversion matrix to calculate the compensation current command corresponding to each frequency component, including: The real-time AC and DC axis currents of the permanent magnet synchronous motor are obtained, and the AC and DC axis inductance saturation characterization value reflecting the degree of magnetic circuit saturation is calculated based on the real-time AC and DC axis currents. Based on the cross-axis and direct-axis inductance saturation characterization values, the cross-coupling coefficients in the preset reference force-current conversion matrix are nonlinearly interpolated and corrected to obtain the dynamic force-current conversion matrix. The compensation current command corresponding to each frequency component, including the quadrature axis compensation component and the direct axis compensation component, is calculated by performing matrix multiplication on each of the corrected and canceled force wave components and the dynamic force-current conversion matrix.

9. The vibration noise suppression control method of a permanent-magnet synchronous motor according to claim 1, characterized by, Injecting the multi-frequency compensation current command into the base current control loop, and generating a suppressive voltage waveform through space vector pulse width modulation to drive the permanent magnet synchronous motor, including: Based on the sampling period and algorithm execution period of the control system, the digital delay phase angle of the basic current control loop is calculated for each frequency component in the multi-frequency compensation current command. The digital delay phase angle is used to perform phase lead compensation on the corresponding frequency component in the multi-frequency compensation current command, and the compensated command is injected into the basic current control loop to obtain a composite voltage vector. The composite voltage vector is input to the space vector pulse width modulation module to generate the suppressive voltage waveform, and the suppressive voltage waveform is output to the inverter to drive the permanent magnet synchronous motor.

10. The vibration noise suppression control method of a permanent-magnet synchronous motor according to claim 9, characterized by, The composite voltage vector is input to the space vector pulse width modulation module to generate the suppressive voltage waveform, and the suppressive voltage waveform is output to the inverter to drive the permanent magnet synchronous motor, including: Obtain the real-time DC bus voltage of the inverter, and construct a space vector hexagonal dynamic voltage boundary based on the real-time DC bus voltage; The composite voltage vector is decomposed into a fundamental driving voltage component and a harmonic compensation voltage component for vibration suppression. Determine whether the composite voltage vector's amplitude and phase composite trajectory exceeds the space vector hexagonal dynamic voltage boundary; If it exceeds the limit, then, with the principle of keeping the fundamental driving voltage component unchanged, the harmonic compensation voltage component is projected and limited along the residual margin profile of the space vector hexagonal dynamic voltage boundary to reconstruct a safe composite voltage vector. The safety composite voltage vector is input into the space vector pulse width modulation module to generate the suppressive voltage waveform, and the suppressive voltage waveform is output to the inverter to drive the permanent magnet synchronous motor.