Method for current sampling error correction and low frequency torque ripple suppression

Abstract

A method for correcting current sampling errors and suppressing low-frequency torque ripples belongs to the field of motor control technology. This invention includes calculating the first... k Second-rate dq In shaft current feedback, the corresponding amplitudes and initial phase angles at the second and first harmonic fundamental frequencies are compared. k= 1, 2, 3…; Based on the calculation results, reconstruct the current pulsation information, and then… dq The shaft current feedback is used to correct sampling errors; it is then determined whether the current sampling error correction result meets the preset stopping condition, thus completing the online correction of the current sampling bias error and proportional error and suppressing the resulting low-frequency torque pulsation. During the current sampling error correction process, it is not necessary to know the controlled motor's moment of inertia and motor body parameters in advance, improving the compatibility and intelligence of the motor control system.

Description

Technical Field

[0001] This invention belongs to the field of motor control technology. Background Technology

[0002] Permanent magnet synchronous motors (PMSMs) use permanent magnets for excitation, resulting in a simpler motor structure and reduced processing and assembly costs, thus leading to their widespread application. However, during operation, temperature drift and zero drift in the hardware sampling circuit, along with proportional errors in the conditioning circuit, can cause errors in current sampling. These errors can be categorized into two types: current sampling bias error and proportional error. The current sampling bias error causes periodic pulsations in the dq-axis current feedback at a frequency equal to one times the fundamental frequency of the electrical circuit, while the current sampling proportional error causes periodic pulsations in the dq-axis current feedback at a frequency equal to two times the fundamental frequency. Since the q-axis current is directly proportional to the motor output torque, the motor torque will also experience torque pulsations, further leading to speed fluctuations at the same frequency, thus affecting the operational performance of the PMSM control system.

[0003] Currently, some existing current sampling error correction methods require the use of information such as the motor's moment of inertia, viscosity coefficient, or body parameters to solve for the current sampling error at a fixed speed in order to complete the sampling error correction.

[0004] In the process of implementing the embodiments of this disclosure, at least the following problems were found in the related art: In existing technologies, the process of solving current sampling errors requires prior knowledge of parameters such as the moment of inertia, viscosity coefficient, or stator resistance of the controlled motor. This requirement may not be met in practical industrial applications, especially in motor control systems with unknown parameters and time-varying inertia. Therefore, the targeted development of highly robust current sampling error correction technology has significant theoretical and practical value. Summary of the Invention

[0005] The purpose of this invention is to improve the applicability of low-frequency torque ripple suppression strategies under current sampling errors in permanent magnet synchronous motor control systems, and to provide a method for current sampling error correction and low-frequency torque ripple suppression.

[0006] The steps of this invention are: S11. The motor control system determines whether the system has entered a stable state; S12. The motor control system calculates the amplitude and initial phase angle at twice the fundamental frequency in the k-th dq-axis current feedback, k = 1, 2, 3...; S13. The motor control system calculates the amplitude and initial phase angle at the fundamental frequency of the first electric frequency in the k-th dq-axis current feedback, k = 1, 2, 3...; S14. The motor control system reconstructs the current pulsation information based on the calculation results and corrects the current sampling error. S15. The motor control system determines whether the current sampling error correction result meets the preset stop condition. S16. Calculate the amplitude at twice the fundamental frequency and the corresponding amplitude at the fundamental frequency in the k-th dq-axis current feedback of the motor control system according to formula (5): Where x can be d or q, I xk In the k-th dq-axis current feedback, the frequency is f base or 2f base The periodic pulsation amplitude at I Rek I represents the real part amplitude of the k-th dq-axis current feedback at a specific frequency. Imk This represents the imaginary part amplitude of the k-th dq-axis current feedback at a specific frequency. S17. Calculate the real part amplitude of the k-th dq-axis current feedback at a specific frequency according to equation (6): Where N is the total number of sampling points, i xm (i) represents the i-th dq-axis current feedback, and p is the frequency relative to the sampling frequency f. s The target frequency is an integer related to the number of sampling points N, where i∈N; S18. Calculate the imaginary amplitude of the k-th dq-axis current feedback at a specific frequency according to equation (7): Where N is the total number of sampling points, i xm (i) represents the i-th dq-axis current feedback, and p is the frequency relative to the sampling frequency f. s The target frequency is an integer related to the number of sampling points N, where i∈N; S19. Calculate the total number of sampling points N according to formula (8): N = pf s / nf base (8) Among them, f s The sampling frequency represents n times the fundamental frequency; S20. Calculate the initial phase angle at the target frequency in the k-th dq-axis current feedback of the motor control system according to equation (9): S21, according to I Rek The calculation results determine the initial phase angle at the target frequency, including: When I in equation (6) Re When the calculation result is less than 0, in equation (9) It should be modified by equation (10): The pulsation information at twice the fundamental frequency in the k-th dq-axis current feedback of the motor control system is calculated according to equation (11): in, and The frequency obtained from the k-th DFT reconstruction is 2f. base Periodic pulsations in the dq-axis current feedback, ω e Let t be the electric angular velocity of the motor, and t be the running time. and The k-th reconstruction frequency is 2f. base The periodic pulsating phase angle of the dq-axis current; S22. Based on the DFT signal reconstruction principle, calculate the pulsating phase angle, including: Guarantee the instantaneous removal of pulsating phase angle during reconstruction The remaining phase angles are 0 rad, that is It should be defined by equation (12): Where, θ e_start This represents the instantaneous electrical angle value at the start of the DFT reconstruction. S23. Calculate the pulsation information at the fundamental frequency (one times the electrical frequency) in the k-th dq-axis current feedback of the motor control system according to equation (13): in, and f represents the frequency obtained from the k-th DFT reconstruction. base Periodic pulsations in the dq-axis current feedback, ω e Let t be the electric angular velocity of the motor, and t be the running time. and The frequency of the k-th reconstruction is f. base The periodic pulsating phase angle of the dq-axis current; S24. Based on the DFT signal reconstruction principle, calculate the pulsating phase angle, including: Guarantee the instantaneous removal of pulsating phase angle during reconstruction The remaining phase angles are 0 rad, that is It should be defined by equation (14): Where, θ e_start This represents the instantaneous electrical angle value at the start of the DFT reconstruction. S25. Based on the reconstructed results of the pulsation of the second-order and first-order fundamental frequencies, correct the current sampling ratio error and bias error, including: The k-th dq-axis current feedback is corrected according to equation (15): Among them, i dmk with i qmk Represents the k-th current feedback, i dmk_new with i qmk_new This represents the corrected k-th current feedback. S26. When there are current sampling bias error and proportional error, the sampled current of phases abc is expressed as: Among them, i am i bm with i cm k represents the sampling current of phases abc. a With k b i represents the sampling ratio coefficient of phase ab current. ar with i br Represents the actual phase current of ab, Δi a With Δi b This represents the sampling bias error of the current in phases a and b; The dq-axis current feedback obtained by Clark and Park coordinate transformation of the sampled currents of phases S27 and abc is expressed as follows: Among them, i dm with i qm Represents dq-axis current feedback, i dr with i qr The dq-axis current, Δi, represents the current obtained from the actual phase current through Clark and Park coordinate transformation. d_err_mea With Δi q_err_mea S28 represents the dq-axis current sampling error caused by the phase current sampling bias error; when only the current sampling bias error exists, combine equations (16) and (17) and substitute k a =k b =1, therefore: The above formula simplifies to: Where, the expression for β is: S29. When only the current sampling ratio error exists, the phase current feedback is written as follows: Among them, Ia I b This represents the peak value of the ab phase current measured by the controller; According to equation (21), the expression for the actual current in phases ab is: According to Kirchhoff's current law, the c-phase sampling current i cm With the actual current i in phase c cr for: S30. Combining equations (21) and (23), we obtain the result of the three-phase sampled current after Clark and Park transformation under the current sampling ratio error: S31. Combining equations (22) and (23), we obtain the following result for the three-phase current after Clark and Park transformations under the current sampling ratio error: S32. The following relationship is obtained: The expressions for k1 to k5 are as follows: k1=I a k b k2 = I b k a k3 = I a k a k b k4 = I b k a k b k5 = k a k b (27).

[0007] This invention enables online correction of current sampling bias error and proportional error, and suppresses the resulting low-frequency torque ripple. During the current sampling error correction process, it is not necessary to know the controlled motor's moment of inertia and other motor parameters in advance, thus improving the compatibility and intelligence of the motor control system. Attached Figure Description

[0008] Figure 1 This is a method for correcting current sampling errors and a control block diagram for suppressing low-frequency torque ripple provided in this disclosure embodiment; Figure 2 This is a flowchart of a method for correcting current sampling errors and suppressing low-frequency torque ripple, provided in an embodiment of this disclosure. Figure 3This is a flowchart of another method for suppressing double-frequency torque ripple under the coexistence of multiple influencing factors, provided in an embodiment of this disclosure; Figure 4(a) shows the dq axis current feedback waveforms before and after sampling error correction for a motor operating at 300 r / min and 1 N·m load, under the current sampling ratio error provided in this embodiment of the present disclosure. Figure 4(b) shows the speed feedback waveforms of the motor before and after sampling error correction under the current sampling ratio error provided in the embodiment of this disclosure, when the motor operates at a speed of 300 r / min and a load of 1 N·m. Figure 4(c) shows the current feedback waveforms of phases abc before and after sampling error correction for a motor operating at 300 r / min and 1 N·m load under the current sampling ratio error provided in the embodiment of this disclosure. Figure 5(a) and Figure 5(b) show the FFT analysis results of the q-axis current before and after sampling error correction for the motor at a speed of 300 r / min and a load of 1 N·m under the current sampling ratio error. Figure 6(a) and Figure 6(b) show the FFT analysis results of the speed feedback before and after the sampling error correction for the motor at a speed of 300 r / min and a load of 1 N·m, under the current sampling proportional error. Figure 7(a) shows the dq axis current feedback waveforms before and after sampling error correction for a motor operating at 300 r / min and 1 N·m load under the current sampling bias error provided in this embodiment of the present disclosure. Figure 7(b) shows the speed feedback waveforms of the motor before and after sampling error correction under the current sampling bias error provided in the embodiments of this disclosure, with the motor operating at a speed of 300 r / min and a load of 1 N·m. Figure 7(c) shows the current feedback waveforms of phases abc before and after sampling error correction under the current sampling bias error provided in the embodiments of this disclosure, when the motor operates at a speed of 300 r / min and a load of 1 N·m. Figure 8(a) and Figure 8(b) show the FFT analysis results of the q-axis current before and after sampling error correction under the current sampling bias error, with the motor operating at a speed of 300 r / min and a load of 1 N·m. Figure 9(a) and Figure 9(b) show the FFT analysis results of the speed feedback before and after the sampling error correction of the motor under the current sampling bias error, with a speed of 300 r / min and a load of 1 N·m. Figure 10(a) shows the dq axis current feedback waveforms before and after sampling error correction when the current sampling ratio and bias error coexist with the motor at a speed of 300 r / min and a load of 1 N·m, according to an embodiment of this disclosure. Figure 10(b) shows the speed feedback waveforms before and after sampling error correction when the current sampling ratio and bias error coexist with the motor at a speed of 300 r / min and a load of 1 N·m, according to an embodiment of this disclosure. Figure 10(c) shows the current feedback waveforms of phases abc before and after sampling error correction when the current sampling ratio and bias error coexist with the motor at a speed of 300 r / min and a load of 1 N·m, according to an embodiment of this disclosure. Figures 11(a), 11(b), and 11(c) show the FFT analysis results of the q-axis current before sampling error correction, after proportional error correction, and after proportional error and bias error correction, provided by the embodiments of this disclosure, under the condition that the current sampling ratio and bias error coexist, and the motor operates at a speed of 300 r / min and a load of 1 N·m. Figures 12(a), 12(b), and 12(c) show the FFT analysis results of the speed feedback when the current sampling ratio and bias error coexist, and the motor is operating at a speed of 300 r / min with a load of 1 N·m, before the sampling error is corrected, after the ratio error is corrected, and after the ratio error and bias error are corrected, provided in the embodiments of this disclosure. Figure 13 This is a schematic diagram of a device for suppressing torque / speed pulsation provided in an embodiment of this disclosure. Detailed Implementation

[0009] This invention discloses a method for correcting current sampling errors and suppressing low-frequency torque ripple, comprising: calculating the amplitude and initial phase angle at the second and first fundamental frequencies in the k-th dq-axis current feedback of the motor control system, where k = 1, 2, 3…; reconstructing current ripple information based on the calculation results, and correcting the sampling error of the dq-axis current feedback; determining whether the current sampling error correction result meets the preset stopping condition, thereby completing the online correction of the current sampling bias error and proportional error and suppressing the resulting low-frequency torque ripple. During the current sampling error correction process, it is not necessary to know the moment of inertia and motor body parameters of the controlled motor in advance, improving the compatibility and intelligence of the motor control system. This application also discloses a device and a motor control system for suppressing torque ripple caused by current sampling errors.

[0010] This disclosure provides a method and apparatus for correcting current sampling errors and suppressing low-frequency torque ripple, as well as a motor control system, to improve the applicability of the strategy for suppressing low-frequency torque ripple under current sampling errors in a permanent magnet synchronous motor control system.

[0011] In some embodiments, the method for correcting current sampling errors and suppressing low-frequency torque ripple includes: calculating the amplitude and initial phase angle at the second and first fundamental frequencies in the k-th dq-axis current feedback of the motor control system, where k = 1, 2, 3...; Based on the calculation results, the periodic pulsations of the second and first fundamental frequencies in the k-th dq-axis current feedback are reconstructed, k = 1, 2, 3...; Based on the reconstruction results, the dq-axis current feedback is corrected to complete the current sampling error correction; Determine whether the current sampling error correction result meets the preset stopping condition.

[0012] In some embodiments, the apparatus for correcting current sampling errors includes: The processor and a memory storing program instructions, the processor being configured to, when executing the program instructions, perform the method described above for suppressing torque / speed pulsations.

[0013] In some embodiments, the motor control system includes: means for current sampling error correction as described above.

[0014] The method, apparatus, and motor control system for current sampling error correction provided in this invention can achieve the following technical effects: By calculating the amplitude and initial phase angle at the second and first harmonic frequencies in the dq-axis current feedback of the running motor, the pulsation components at these frequencies are reconstructed. This process corrects the current sampling error (sampling bias error and proportional error) and suppresses torque pulsation caused by the current sampling error. In the process of current sampling error correction and low-frequency torque pulsation suppression, it is not necessary to know the controlled motor's moment of inertia and viscosity coefficient in advance; instead, the judgment is made based on the feedback information from the motor control system. Therefore, the compatibility and intelligence of the motor control system's torque / speed pulsation suppression strategy are improved.

[0015] Due to factors such as zero drift and temperature drift, the current sampling stage of permanent magnet synchronous motors (PMSMs) suffers from numerous non-ideal factors. To simplify analysis, these non-ideal factors can be categorized into two types: current sampling bias error and current sampling ratio error. Current sampling bias error leads to torque pulsation at one times the electrical frequency, while current sampling ratio error leads to torque pulsation at twice the electrical frequency, affecting motor control performance. Many scholars have conducted in-depth research on current sampling error correction strategies. The IEEE Transactions on Industry Applications, in its published paper "Analysis and Compensation of Current Measurement Error in Vector-Controlled AC Motor Drives," provides a detailed analysis of the impact of different types of current sampling errors, presents the amplitude relationship of current sampling errors in open-loop operation, and proposes corresponding current sampling error correction strategies. However, this method only corrects the q-axis current feedback and requires information such as the load motor's rotational inertia and the permanent magnet flux linkage. Similarly, the method proposed in the IEEE Transactions on Industry Applications, in its published paper "Compensation of Current Measurement Error for Current-Controlled PMSM Drives," also requires parameters such as stator resistance and dq-axis inductance. While the above methods can achieve ideal low-frequency torque ripple suppression performance, they cannot complete current sampling error correction when the motor parameters are unknown. In addition, the above methods only analyze the influence of ripple characterization under open-loop conditions, but do not conduct in-depth analysis of the influence of closed-loop control on ripple amplitude.

[0016] Considering that actual operating conditions may include situations where motor parameter information cannot be obtained and the motor control system is operating in a closed-loop control state, it is necessary to optimize the existing current sampling error correction algorithm, reduce parameter dependence, and consider the actual impact of closed-loop control on the algorithm.

[0017] Combination Figure 1 The diagram shows a control block diagram of a method for current sampling error correction and low-frequency torque ripple suppression provided in this disclosure. In this embodiment, the permanent magnet synchronous motor (PMSM) has a d-axis command current. With q-axis command current p0 represents the number of pole pairs of the motor, k a k b i represents the sampling ratio coefficient of phase ab current. ar ibr Δi represents the actual current in phases ab. a With Δi b i represents the sampling bias error of the two-phase currents a and b. am i bm This represents the sampled current of phases ab. The sampled phase current value is used in conjunction with the electrical angle feedback θ. e Determine the d-axis current feedback i dm With q-axis current feedback i qm The current loop PI controller (Proportional Integral Controller) outputs a voltage command. and After determining the speed and achieving steady-state operation under load, a Discrete Fourier Transform (DFT) is performed on the dq-axis current feedback to obtain the corresponding amplitudes and initial phase angles at the second and first harmonic frequencies of the sampled dq-axis current. Based on these results, the sampling error of the dq-axis current feedback is corrected, and torque ripple caused by the current sampling error is suppressed. The corrected dq-axis current feedback is i dm_new with i dm_new .

[0018] Combination Figure 2 As shown, this disclosure provides a method for suppressing torque / speed pulsation, comprising: S01, calculate the amplitude and initial phase angle at twice the fundamental frequency in the k-th dq-axis current feedback, k = 1, 2, 3...

[0019] S02, calculate the amplitude and initial phase angle at the fundamental frequency (one times the electrical frequency) in the k-th dq-axis current feedback, k = 1, 2, 3...

[0020] In this technical solution, the dq axis current feedback of the motor control system is obtained by sampling the phase current and transforming it using Clark and Park methods.

[0021] S03, the motor control system reconstructs the pulsation components at the second and first electric frequency base frequencies in the k-th dq axis current feedback according to the calculation results, k = 1, 2, 3...

[0022] S04. Based on the reconstructed information, the sampling error of the dq axis current feedback is corrected, and the torque pulsation caused by the current sampling error is suppressed.

[0023] S05, determine whether the current sampling error correction result meets the preset stop condition.

[0024] The method for current sampling error correction and low-frequency torque ripple suppression provided in this disclosure calculates the amplitude and initial phase angle corresponding to twice the electrical frequency fundamental frequency in the dq-axis current feedback of the motor control system. This reconstructs the current ripple component during current sampling error correction, thereby correcting the current sampling error and suppressing low-frequency torque ripple caused by the error. During the current sampling error correction process, it is not necessary to know the controlled motor's moment of inertia and viscosity coefficient in advance; instead, the judgment is made based on the feedback information from the motor control system, thus improving the compatibility and intelligence of the correction algorithm.

[0025] Optionally, the amplitude at twice the fundamental frequency in the k-th dq-axis current feedback of the motor control system is calculated, including: Calculate the i-th dq-axis current feedback of the motor control system; based on the i-th dq-axis current feedback, calculate the corresponding amplitudes at the second and first times the fundamental frequency of the electric frequency in the k-th dq-axis current feedback of the motor control system.

[0026] Optionally, the calculation of the i-th dq-axis current feedback of the motor control system includes: Obtain the three-phase current samples (abc); based on the sampled currents of the three-phase currents (abc), calculate the i-th dq-axis current feedback of the motor control system.

[0027] In practical applications, the sampling current of phase a and phase b can be obtained through a current sensor, and then the sampling current of phase c can be determined according to equation (1): i cm =-(i am +i bm (1).

[0028] Optionally, the i-th dq-axis current feedback of the motor control system is calculated according to equation (2): Where, θ e The value is the electrical angle feedback value, π is the mathematical constant pi, which has a value of 3.1415926, and i is an integer greater than or equal to zero.

[0029] Optionally, the electrical angle feedback value is calculated according to equation (3): θ e =2πf base t (3) Among them, f base t is the fundamental frequency of the electrical circuit, and t is the running time. Alternatively, the fundamental frequency can be determined according to equation (4): f base =speed_ref*p0 / 60 (4) Where speed_ref is the current motor speed command, and p0 represents the number of motor pole pairs.

[0030] In some alternative implementations, combined with Figure 3 As shown, this disclosure provides another method for current sampling error correction and low-frequency torque ripple suppression, including: S11, The motor control system determines whether the system has entered a stable state; S12, The motor control system calculates the amplitude and initial phase angle at twice the fundamental frequency in the k-th dq-axis current feedback, k = 1, 2, 3...; S13, The motor control system calculates the amplitude and initial phase angle at the one-time fundamental frequency in the k-th dq axis current feedback, k = 1, 2, 3...; S14, The motor control system reconstructs the current pulsation information based on the calculation results and corrects the current sampling error; S15, the motor control system determines whether the current sampling error correction result meets the preset stop condition.

[0031] Optionally, determining whether the system has entered a stable state includes: If the current speed of the motor in the motor control system remains unchanged within a preset time, the motor control system is considered to have entered a stable state; or, If the average value of the q-axis current feedback of the motor control system remains unchanged within a preset time, the motor control system is considered to have entered a stable state.

[0032] In practical applications, the current speed of the motor in the motor control system is an obtainable value, and the dq axis current feedback of the motor control system is determined by the above equations (1), (2), (3), and (4). Optionally, the amplitudes at twice the fundamental frequency and at the fundamental frequency in the k-th dq-axis current feedback of the motor control system are calculated according to equation (5): Where x can be d or q, I xk In the k-th dq-axis current feedback, the frequency is f base or 2f base The periodic pulsation amplitude at I Rek This represents the k-th dq-axis current feedback at a specific frequency (f). base or 2f base The real part amplitude under ) I Imk This represents the k-th dq-axis current feedback at a specific frequency (f). base or 2f base The imaginary amplitude under ().

[0033] Optionally, the k-th dq-axis current feedback at a specific frequency (f) is calculated according to equation (6). base or 2f base The real part amplitude under: Where N is the total number of sampling points, i xm (i) represents the i-th dq-axis current feedback, and p is the frequency relative to the sampling frequency f. s Target frequency (f) base or 2f base (i) is an integer related to the number of sampling points N, where i∈N.

[0034] Optionally, the k-th dq-axis current feedback at a specific frequency (f) is calculated according to equation (7). base or 2f base The imaginary part amplitude under () Where N is the total number of sampling points, i xm (i) represents the i-th dq-axis current feedback, and p is the frequency relative to the sampling frequency f. s Target frequency (f) base or 2f base (i) is an integer related to the number of sampling points N, where i∈N.

[0035] Optionally, the total number of sampling points N can be calculated according to equation (8): N = pf s / nf base (8) Among them, f s The sampling frequency is n times the fundamental frequency. When there is a phase current sampling bias error, n = 1; when there is a phase current proportional bias, n = 2.

[0036] In practical applications, the preset period p should theoretically take as many values ​​as possible to ensure more accurate calculations of the amplitudes of the real and imaginary parts. However, in actual operating conditions, abrupt changes caused by load or other objective factors need to be considered. Therefore, in this technical solution, the value of the preset period p is taken as a positive integer less than 5.

[0037] Optionally, the target frequency (f) in the k-th dq-axis current feedback of the motor control system is calculated according to equation (9). base or 2f base Initial phase angle at ()

[0038] Optionally, according to I Rek The calculation results are for the target frequency (f) base or 2f base The initial phase angle at point () is determined, including: Determine the initial phase angle for DFT reconstruction; since the range of the arctan function is... When I in equation (6) Re When the calculation result is less than 0, in equation (9) It should be modified by equation (10):

[0039] Optionally, the pulsation information at twice the fundamental frequency in the k-th dq-axis current feedback of the motor control system is calculated according to equation (11): in, and The frequency obtained from the k-th DFT reconstruction is 2f. base Periodic pulsations in the dq-axis current feedback, ω e Let t be the electric angular velocity of the motor, and t be the running time. and The k-th reconstruction frequency is 2f. base The dq-axis current periodic pulsation phase angle.

[0040] Optionally, the pulsating phase angle is calculated based on the DFT signal reconstruction principle, including: Guarantee the instantaneous removal of pulsating phase angle during reconstruction The remaining phase angles are 0 rad, that is It should be defined by equation (12): Where, θ e_start This represents the instantaneous electrical angle value at the start of the DFT reconstruction.

[0041] Optionally, the pulsation information at one frequency (fundamental frequency) in the k-th dq-axis current feedback of the motor control system is calculated according to equation (13): in, and f represents the frequency obtained from the k-th DFT reconstruction. base Periodic pulsations in the dq-axis current feedback, ω e Let t be the electric angular velocity of the motor, and t be the running time. and The frequency of the k-th reconstruction is f. base The dq-axis current periodic pulsation phase angle.

[0042] Optionally, the pulsating phase angle is calculated based on the DFT signal reconstruction principle, including: Guarantee the instantaneous removal of pulsating phase angle during reconstruction The remaining phase angles are 0 rad, that is It should be defined by equation (14): Where, θ e_start This represents the instantaneous electrical angle value at the start of the DFT reconstruction.

[0043] Optionally, based on the reconstructed results of the pulsations of the second and first harmonic fundamental frequencies (Equations (11) and (13)), the current sampling ratio error and bias error are corrected, including: Based on the amplitude and pulsating phase angle at the second and first fundamental frequencies in the k-th dq-axis current feedback, the current sampling error is corrected in the rotating coordinate system; based on the corrected current feedback harmonic amplitude, it is determined whether the correction result meets the preset stopping condition.

[0044] Optionally, the k-th dq-axis current feedback can be corrected according to equation (15): Among them, i dmk with i qmk Represents the k-th current feedback, i dmk_new with i qmk_new This represents the corrected k-th current feedback.

[0045] In practical applications, it should be understood that the correction of current sampling error is continuous even if the correction result does not meet the preset stopping condition. During this process, the frequency f in the dq-axis current feedback is repeatedly extracted. base or 2f base The periodic pulsation is corrected by using Equation (15) to correct the dq axis current feedback.

[0046] In this technical solution, when there are current sampling bias error and proportional error, the sampling current of phases abc can be expressed as: Among them, i am i bm with i cm k represents the sampling current of phases abc. a With k b i represents the sampling ratio coefficient of phase ab current. ar with i br Represents the actual phase current of ab, Δi a With Δi b This represents the sampling bias error of the current in phases a and b.

[0047] Furthermore, the dq-axis current feedback obtained by transforming the sampled currents of phases abc through Clark and Park coordinates can be expressed as: Among them, i dm with i qm Represents dq-axis current feedback, i dr with i qr The dq-axis current, Δi, represents the current obtained from the actual phase current through Clark and Park coordinate transformation. d_err_mea With Δi q_err_mea This represents the dq-axis current sampling error caused by the phase current sampling bias error.

[0048] When only current sampling bias error exists, combine equations (16) and (17) and substitute k. a =k b =1, therefore: The above formula can be further simplified to: Where, the expression for β is:

[0049] Therefore, the phase current sampling bias error will result in the frequency being the fundamental frequency of the motor's operating frequency (f). base The periodic torque pulsation will cause periodic speed pulsation of the same frequency.

[0050] When only the current sampling ratio error exists, the phase current feedback can be written as follows: Among them, I a I b This represents the peak value of phase ab current measured by the controller.

[0051] According to equation (21), the expression for the actual current in phases ab is:

[0052] Since the three-phase windings of the experimental surface-mount permanent magnet synchronous motor are star-connected, according to Kirchhoff's current law, the c-phase sampling current i can be obtained. cm With the actual current i in phase c cr for:

[0053] Combining equations (21) and (23), we can obtain the result of the three-phase sampled current after Clark and Park transformation under the current sampling ratio error:

[0054] Combining equations (22) and (23), we can obtain the result of the three-phase current after Clark and Park transformation under the current sampling ratio error:

[0055] The following relationship can be obtained: The expressions for k1 to k5 are as follows: k1=I a k b k2 = I b k a k3 = I a k a k b k4 = I b k a k b k5 = k a k b (27).

[0056] Therefore, the phase current sampling ratio error will result in a frequency that is twice the fundamental frequency of the motor's operating frequency (f). base The periodic torque pulsation will cause periodic speed pulsation of the same frequency.

[0057] In some embodiments, the method for current sampling error correction provided in this disclosure is verified. Table 1 shows some parameter information of the permanent magnet synchronous motor used in the experiment:

[0058] In practical applications, when there is a current sampling ratio error (k) a =0.8, k b =1.2), when the motor is running at 300 r / min (electric frequency base frequency 20 Hz) and under a load of 1 N·m, the dq axis current feedback before and after current sampling ratio error correction is shown in Figure 4(a), the speed feedback is shown in Figure 4(b), and the abc phase current feedback is shown in Figure 4(c). It can be seen that when current sampling ratio error occurs, the dq axis current feedback and speed feedback show obvious periodic pulsation, and the three-phase current shows obvious asymmetry. After the first sampling error correction, the periodic inertial pulsation of the dq axis current feedback is suppressed and the asymmetry of the three-phase current is improved. After the second correction, the periodic pulsation of the dq axis current feedback reaches the preset stopping condition, and the sampling ratio error correction ends.

[0059] Figures 5(a) and 5(b) show the FFT analysis results of the q-axis current feedback before and after the current sampling ratio error correction. It can be seen that when there is a ratio error, the q-axis current feedback ripple frequency is twice the fundamental frequency (40Hz), and its harmonic amplitude is 0.4092A. After adopting the method in the embodiment of this disclosure, the current ripple amplitude at twice the fundamental frequency is reduced to 0.0176A, which is 4.3% of that before the sampling ratio error correction.

[0060] Figures 6(a) and 6(b) show the FFT analysis results of the speed feedback before and after the current sampling ratio error correction. It can be seen that when there is a ratio error, the speed feedback pulsation frequency is twice the electrical frequency (40Hz) fundamental frequency, and its harmonic amplitude is 2.446 r / min. After adopting the method in the embodiment of this disclosure, the speed pulsation amplitude at twice the electrical frequency is reduced to 0.138 r / min, which is 5.6% of that before the sampling ratio error correction.

[0061] In practical applications, when there is a current sampling bias error (Δi) a =0.1A, Δi b =-0.25A), when the motor is running at 300 r / min (electrical frequency 20 Hz) and under a load of 1 N·m, the dq-axis current feedback before and after current sampling bias error correction is shown in Figure 7(a), the speed feedback is shown in Figure 7(b), and the abc-phase current feedback is shown in Figure 7(c). It can be seen that when current sampling bias error occurs, the dq-axis current feedback and speed feedback show obvious periodic pulsation, and the three-phase current shows obvious asymmetry. After current sampling bias error correction, the periodic inertial pulsation of the dq-axis current feedback is suppressed and the three-phase current asymmetry is significantly improved. Because the corrected current pulsation reaches the preset stopping condition, the sampling bias error correction ends.

[0062] Figures 8(a) and 8(b) show the FFT analysis results of the q-axis current feedback before and after the current sampling bias error correction. It can be seen that when there is a bias error, the q-axis current feedback ripple frequency is the fundamental frequency of the electrical frequency (20Hz), and its harmonic amplitude is 0.2268A. After adopting the method in the embodiment of this disclosure, the current ripple amplitude at the electrical frequency is reduced to 0.0212A, which is 9.3% of that before the sampling bias error correction.

[0063] Figures 9(a) and 9(b) show the FFT analysis results of the speed feedback before and after the current sampling bias error correction. It can be seen that when there is a bias error, the speed feedback pulsation frequency is the fundamental frequency of the first electrical frequency (20Hz), and its harmonic amplitude is 1.322 r / min. After adopting the method in the embodiment of this disclosure, the speed pulsation amplitude at the first electrical frequency is reduced to 0.0867 r / min, which is 6.5% of that before the sampling bias error correction.

[0064] In practical applications, when there is a current sampling ratio and a bias error (k) a =1.1, k b =0.9, Δi a =0.2A, Δi b =0.1A), when the motor is running at 300 r / min (electric frequency base frequency 20 Hz) and under a load of 1 N·m, the dq axis current feedback before and after current sampling error correction is shown in Figure 10(a), the speed feedback is shown in Figure 10(b), and the abc phase current feedback is shown in Figure 10(c). It can be seen that when there is a current sampling error, the dq axis current feedback and speed feedback show obvious periodic pulsation, and the three-phase current shows obvious asymmetry. After correcting the current sampling ratio error and bias error in sequence, the current and speed pulsation are effectively suppressed, and the phase current symmetry is significantly improved.

[0065] Figures 11(a), 11(b), and 11(c) show the q-axis current feedback FFT analysis results before current sampling error correction, after current sampling ratio error correction, and after current sampling bias error correction under the operating condition shown in Figure 10. When ratio and bias sampling errors coexist, there are pulsations at one and two times the electrical frequency in the current feedback, with amplitudes of 0.402A and 0.175A, respectively. After ratio error correction, the amplitude at two times the electrical frequency decreases to 0.0162A, which is 9.2% of the original value. After bias error correction, the amplitude at one times the electrical frequency decreases to 0.0213A, which is 5.3% of the original value.

[0066] Figures 12(a), 12(b), and 12(c) show the FFT analysis results of the speed feedback before, after, and after the current sampling ratio error correction and current sampling bias error correction under the operating condition shown in Figure 10. When the ratio and bias sampling errors coexist, there are pulsations at one and two times the electrical frequency in the speed feedback, with amplitudes of 2.355 r / min and 1.163 r / min, respectively. After the ratio error correction, the amplitude at two times the electrical frequency decreases to 0.1075 r / min, which is 4.5% of the original value. After the bias error correction, the amplitude at one times the electrical frequency decreases to 0.103 r / min, which is 8.8% of the original value.

[0067] Combination Figure 13As shown, this disclosure provides an apparatus for suppressing torque / speed pulsation, including a processor 100 and a memory 101. Optionally, the apparatus may further include a communication interface 102 and a bus 103. The processor 100, communication interface 102, and memory 101 can communicate with each other via the bus 103. The communication interface 102 can be used for information transmission. The processor 100 can call logical instructions in the memory 101 to execute the method for suppressing torque / speed pulsation described in the above embodiment.

[0068] Furthermore, the logic instructions in the aforementioned memory 101 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium.

[0069] The memory 101, as a computer-readable storage medium, can be used to store software programs and computer-executable programs, such as program instructions / modules corresponding to the methods in the embodiments of this disclosure. The processor 100 executes functional applications and data processing by running the program instructions / modules stored in the memory 101, that is, it implements the method for suppressing torque / speed pulsation in the above embodiments.

[0070] The memory 101 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the terminal device. Furthermore, the memory 101 may include high-speed random access memory and may also include non-volatile memory.

Claims

1. A method for correcting current sampling errors and suppressing low-frequency torque ripple, characterized in that: The steps are as follows: S1. The motor control system determines whether the system has entered a stable state; S2, Motor control system calculation Second-rate The amplitude and initial phase angle at twice the fundamental frequency in shaft current feedback. ; S3, Motor control system calculation Second-rate The amplitude and initial phase angle at one times the fundamental frequency in shaft current feedback. ; S4. The motor control system reconstructs the current pulsation information based on the calculation results and corrects the current sampling error. S5. The motor control system determines whether the current sampling error correction result meets the preset stop condition. Among them, the first step of the motor control system is calculated according to equation (9). Second-rate Initial phase angle at the target frequency in shaft current feedback: （9） In the formula, Representing the Second-rate The real part amplitude of the shaft current feedback at a specific frequency. Representing the Second-rate The imaginary part amplitude of the shaft current feedback at a specific frequency. middle for or ; according to The calculation results determine the initial phase angle at the target frequency, including: when When the calculation result is less than 0, in equation (9) The following corrections are made using equation (10): （10） Calculate the first step of the motor control system according to formula (11). Second-rate The pulsation information at twice the fundamental frequency in the shaft current feedback: （11） in, and Representing the The frequency obtained from the DFT reconstruction is of Periodic pulsation in shaft current feedback The electric angular velocity of the motor. For runtime, and Representing the The frequency of reconstruction is of Phase angle of periodic pulsation of shaft current. , The first Second-rate shaft and The frequency in shaft current feedback is or The periodic pulsation amplitude at that location; Based on the DFT signal reconstruction principle, the pulsating phase angle is calculated, including: Guarantee the instantaneous removal of pulsating phase angle during reconstruction The remaining phase angles are 0 rad, that is Defined by equation (12): （12） in, This represents the instantaneous electrical angle value at the start of the DFT reconstruction. Calculate the first step of the motor control system according to formula (13). Second-rate The pulsation information at one times the fundamental frequency in the shaft current feedback: （13） in, and Representing the The frequency obtained from the DFT reconstruction is of Periodic pulsation in shaft current feedback The electric angular velocity of the motor. For runtime, and Representing the The frequency of reconstruction is of Phase angle of periodic pulsation of shaft current; Based on the DFT signal reconstruction principle, the pulsating phase angle is calculated, including: Guarantee the instantaneous removal of pulsating phase angle during reconstruction The remaining phase angles are 0 rad, that is Defined by equation (14): （14） in, This represents the instantaneous electrical angle value at the start of the DFT reconstruction. Based on the reconstructed results of the pulsations of the second and first harmonic fundamental frequencies, the current sampling ratio error and bias error are corrected, including: According to formula (15) for the first Second-rate Shaft current feedback is used for correction: （15） in, and Representing the Secondary current feedback and Represents the revised first Secondary current feedback.

2. The method for current sampling error correction and low-frequency torque ripple suppression according to claim 1, characterized in that: Calculate the first step of the motor control system according to equation (5). Second-rate The amplitude at twice the fundamental frequency and the corresponding amplitude at the fundamental frequency in shaft current feedback: （5） in, Can be or ; Calculate the first according to formula (6) Second-rate The real part amplitude of the shaft current feedback at a specific frequency: （6） in, This represents the total number of sampling points. For the first Second-rate Shaft current feedback, To match the sampling frequency Target frequency and number of sampling points Relevant integers, ; Calculate the first according to formula (7) Second-rate The imaginary amplitude of the shaft current feedback at a specific frequency: （7） in, This represents the total number of sampling points. For the first Second-rate Shaft current feedback, To match the sampling frequency Target frequency and number of sampling points Relevant integers, ; Calculate the total number of sampling points N according to formula (8): （8） in, Sampling frequency, representing Doubling the fundamental frequency.

3. The method for current sampling error correction and low-frequency torque ripple suppression according to claim 1, characterized in that: When current sampling bias error and proportional error exist The phase sampling current is expressed as: （16） in, , and represent Phase sampling current, and represent Phase current sampling ratio, and represent Real phase current, and represent Two-phase current sampling bias error; Phase sampling current obtained by Clark and Park coordinate transformation Shaft current feedback is expressed as: （17） in, and represent Shaft current feedback, and The representation is obtained by transforming the actual phase currents using Clark and Park coordinates. shaft current, and This represents the error caused by phase current sampling bias. Shaft current sampling error; When only current sampling bias error exists, combine equations (16) and (17) and substitute them into the equation. ,have: （18） In the formula, This is the electrical angle feedback value; The above formula simplifies to: （19） in, The expression is: （20）； When only the current sampling ratio error exists, the phase current feedback is written as follows: （21） in, , The data obtained by the controller Peak phase current; According to equation (21). The expression for the actual two-phase current is: （22） According to Kirchhoff's current law, we have Phase sampling current and Phase actual current for: （23）； Combining equations (21) and (23), we obtain the result of the three-phase sampled current after Clark and Park transformation under the current sampling ratio error: （24）； Combining equations (22) and (23), we obtain the following result for the three-phase current after Clark and Park transformations under the current sampling ratio error: （25）； The following relationship is obtained: （26） in, The expression is as follows: （27）。

Images