A surface-mounted permanent magnet synchronous motor rotor position and speed information detection method

Abstract

The application discloses a surface-mounted permanent magnet synchronous motor rotor position and rotating speed information detection method, injects a constant-amplitude pulsating high-frequency square wave voltage signal in an estimated d-q axis coordinate system; carries out coordinate transformation of the estimated d-q axis coordinate system into a two-phase static alpha-beta coordinate system, decomposes a high-frequency current response of the static alpha-beta coordinate system into a sum of sinusoidal signals with different frequencies by using Fourier decomposition; multiplies a cosine modulation wave with the same frequency as the injected signal with the high-frequency current signal, and carries out a cascaded second-order generalized integrator, which simplifies signal processing; carries out normalization processing on the modulated high-frequency current response, extracts rotor position and rotating speed information through a quadrature phase-locked loop composed of a PI controller. Compared with other traditional methods, the method does not need to use a filter and differential operation, effectively reduces digital delay in a rotor position identification system and sensitivity to sampling errors.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of permanent magnet synchronous motor control, and particularly relates to a surface-mounted permanent magnet synchronous motor rotor position and rotating speed information detection method. BACKGROUND

[0002] Permanent magnet synchronous motor (PMSM) is widely used in electrical appliances, automobiles, robots and other fields due to its excellent performance, small size, light weight, high power density, good reliability and fast dynamic performance. In order to realize the precise control of PMSM with high efficiency, the rotor position information of PMSM must be accurately obtained. The traditional method is to install sensors such as Hall sensors and optical encoders at the end of the motor shaft, which leads to the problems of increased motor size, increased cost, inconvenient installation and low reliability in special occasions. Therefore, it is more meaningful to study the PMSM position sensorless control.

[0003] Permanent magnet synchronous motor runs in the zero speed and low speed range, due to the rotor back EMF can obtain the signal to noise ratio is very low, based on the back EMF to obtain the rotor position and speed information of the method is difficult to achieve. Therefore, the use of permanent magnet synchronous motor salient pole characteristics, through the motor electronic winding injection high frequency excitation signal, from the feedback of high frequency signal extraction rotor position information, the realization of the initial position detection and low speed range operation of the rotor. The traditional high frequency signal injection method, including the rotating high frequency voltage signal injection and pulse high frequency voltage signal injection method, are in the form of high frequency sinusoidal voltage signal injection, need to use low pass filter and band pass filter in the speed loop signal processing and extraction module, position and speed tracking observation module, and the fundamental current signal is separated by low pass filter in the current loop, the use of these filters will bring different degree of amplitude attenuation and phase lag, has greatly influenced the system bandwidth, and then affect the system dynamic response performance. In addition, the traditional rotating high frequency voltage signal injection method is to inject high frequency sinusoidal wave voltage signal in two phase static axis, although it has good stability, but for the high frequency signal demodulation process is more troublesome; The traditional pulse high frequency voltage signal injection method is to estimate the direct axis injection, to estimate the cross axis extraction effective rotor position information, so the current ripple and torque ripple is smaller, but this method has the risk of convergence failure, and the loss is larger, and the above two kinds of permanent magnet synchronous motor low speed range sensorless control method is only suitable for higher salient ratio of interior permanent magnet synchronous motor. Liu Bing et al. in the paper "SPMSM sensorless control inverter nonlinear compensation strategy based on generalized second order integrator" (Proceedings of the Chinese Institute of Electrical Engineers, 2018, 38(11): 3365-3374) proposed a strategy based on generalized second order integrator to realize high frequency current extraction and harmonic error elimination; Avoid the use of filter in the process of rotor position signal demodulation, but the method uses high frequency sinusoidal signal injection, the injection frequency can not be too high, the phase-locked loop has some limitations. Liu Guohai et al. in the patent "A permanent magnet synchronous motor low speed domain rotor position identification method" (CN110429886B) proposed to inject high frequency square wave signal in two phase static coordinate axis, use generalized second order integrator to separate the current in the β axis and amplitude demodulation, the whole signal demodulation process does not use any filter, effectively improves the dynamic control performance of the system. But the method still contains a lot of harmonic signals in the process of position demodulation, which still has some influence on the position estimation accuracy.

[0004] Therefore, in order to improve the surface mounted permanent magnet synchronous motor low speed range sensorless control performance, it is urgent to develop a control strategy with good stability, robustness, high dynamic response and high estimation accuracy. SUMMARY

[0005] This invention addresses the above-mentioned problems by proposing a method for detecting rotor position and speed information of surface-mounted permanent magnet synchronous motors based on the injection of pulsating high-frequency square wave signals. Injecting pulsating high-frequency square wave voltage signals into the control method for estimating the direct axis results in smaller current ripple and torque pulsation. Extracting rotor position information from a two-phase stationary shaft system is effective not only for internal permanent magnet synchronous motors but also for surface-mounted permanent magnet synchronous motors. The injected square wave frequency can be increased to half the PWM switching frequency. During current loop processing, because the injected square wave voltage frequency is much higher than the motor operating frequency, the fundamental frequency current remains unchanged between two adjacent current samples within a switching cycle. The fundamental frequency feedback current can be calculated using simple mathematical calculations, effectively improving the current loop bandwidth. In the signal processing and demodulation process for extracting the high-frequency response current and obtaining rotor position information, a cascaded SOGI method is used, reducing the use of multiple filters and effectively avoiding problems such as signal amplitude attenuation and phase lag caused by filter use, thus increasing the accuracy of rotor position estimation.

[0006] The purpose of this invention is to address the problems that traditional high-frequency signal injection methods require a large number of filters during rotor position estimation and demodulation, and that traditional methods are not suitable for surface-mounted permanent magnet synchronous motors with extremely low saliency. This invention proposes a new rotor position estimation and demodulation scheme to improve the sensorless control performance and position estimation accuracy of surface-mounted permanent magnet synchronous motors in the low-speed range.

[0007] The technical solution of this invention is as follows:

[0008] A method for detecting rotor position and speed information of a surface-mounted permanent magnet synchronous motor includes the following steps:

[0009] Step 1: When the actual dq axis differs from the actual dq axis by Δθ e The error angle is estimated by injecting a pulsed high-frequency square wave voltage signal with constant amplitude into the dq-axis coordinate system to obtain a high-frequency current response.

[0010] Step 2: Transform the obtained high-frequency current response from the estimated dq axis coordinate system to the two-phase stationary α-β coordinate system, and use Fourier decomposition to decompose the high-frequency current response in the stationary α-β coordinate system into the sum of sinusoidal signals of different frequencies.

[0011] Step 3: Combine the high-frequency current i in the two-phase stationary coordinate system, which is the sum of sinusoidal signals of different frequencies. αh i βh After passing through a cascaded second-order generalized integrator (SOGI) pre-stage, a high-frequency current signal with the same injection frequency is extracted. Then, a cosine modulation wave with the same frequency as the injection signal is multiplied with the high-frequency current signal and passed through a cascaded SOGI post-stage to suppress the influence of the high-order harmonics contained in the demodulation on the estimated rotor position.

[0012] Step 4: Normalize the modulated high-frequency current response and extract the rotor position and speed information through an orthogonal phase-locked loop composed of a PI controller.

[0013] Furthermore, the pulsed high-frequency square wave voltage signal injected into the estimated dq-axis coordinate system in step 1 specifically refers to:

[0014]

[0015] In the formula, u dh The voltage signal injected into the d-axis, u qh U is the voltage signal injected into the q-axis. inj The value of the injected square wave voltage is denoted by k, which is the control sequence and k = 1, 2, 3...; the frequency of the injected high-frequency signal is half of the PWM carrier frequency.

[0016] Furthermore, the specific process in step 2 is as follows:

[0017] Step 2.1: During motor operation, the current of any two phases of the permanent magnet synchronous motor is collected through the current sampling module, and the resulting three-phase current i is calculated. a i b i c The two-phase stationary coordinate current i is obtained after Clark transformation. α i β This includes the fundamental frequency current i αf i βf High-frequency current i αh i βh And the high-order harmonic current i generated by the inverter power devices αx i βx Three components;

[0018] Among them, the high-frequency response current i αh i βh The differential equation can be expressed as:

[0019]

[0020] In the formula, L dh L qh For the high-frequency inductors of the d and q axes of a permanent magnet synchronous motor, θ e Δθ represents the actual rotor position. e It is the estimation error between the actual rotor position and the estimated rotor position, where P is the differential operator;

[0021] When Δθ e When it is sufficiently small to be close to zero, the above formula can be written as:

[0022]

[0023] Step 2.2: Use Fourier decomposition to decompose the pulsating high-frequency square wave voltage signal in the stationary α-β coordinate system into a sum of sinusoidal signals of different frequencies:

[0024]

[0025] In the formula, ω h The frequency of the injected high-frequency square wave voltage signal is given, and t is the time interval. It represents an odd-order sine wave.

[0026] Substituting the sum of the sinusoidal signals obtained from the Fourier decomposition of the square wave voltage signal into the high-frequency response current equation in the two-phase stationary α-β coordinate system...

[0027]

[0028] The high-frequency response current equation in the stationary axis system is obtained by integral solution.

[0029] Furthermore, the specific process of step 3 is as follows:

[0030] Step 3.1: Set the center frequency of the preamplifier of the cascaded SOGI to ω. h1 The current i in the two-phase stationary coordinate system α i β After passing through the preamplifier of the cascaded SOGI, a high-frequency current signal i with the same injection frequency is extracted. αh i βh It suppresses interference from other subharmonics and achieves the function of a bandpass filter.

[0031] Furthermore, the pre-stage transfer function of the cascaded SOGI is:

[0032]

[0033] Wherein, the input signal x is the high-frequency current signal i α i β The output signal y1 is the extracted center frequency ω h high-frequency response current i αh i βh k1 is the damping coefficient.

[0034] Step 3.2: Utilize a cosine modulation wave with the same frequency as the injected signal, cosω h Multiplying t by the high-frequency current signal described in step 3.1 yields

[0035]

[0036] The above formula contains a signal with the same injection frequency as the signal and its higher frequency signals. Therefore, as long as the signal with the same injection frequency can be obtained, the estimated rotor position information can be calculated.

[0037] Step 3.3: Multiply by the demodulated signal cosω h After t, as can be seen from the above formula, the second term in the demodulated signal is the higher harmonic component, and the first term is the direct component containing position information. Therefore, it is considered to use the subsequent stage of cascaded SOGI to filter out the second harmonic with higher component content, so as to achieve the same effect as low-pass filtering in rotor position observation and suppress the influence of the higher harmonics contained in the demodulation on the estimated rotor position.

[0038] Furthermore, the transfer function of the subsequent stage of the cascaded SOGI is:

[0039]

[0040] Wherein, the input signal x is the signal after passing through cosω h The high-frequency response current signal i after demodulation αh1 i βh1 The output signal y2 is the extracted through component i containing location information. αh2 i βh2 k2 is the damping coefficient.

[0041] Furthermore, the specific process of step 4 is as follows:

[0042] Step 4.1 The amplitude of the high-frequency signal after cascading SOGI is related to the frequency and amplitude of the injected voltage, as well as the inductance parameters of the motor itself. Therefore, per-unit processing is used to improve the robustness to the amplitude and inductance value of the injected signal.

[0043]

[0044] In the formula, i α-pu i β-pu This represents the current along the stationary axis α-β after standardization.

[0045] Step 4.2 According to the heterodyne method, multiply the high-frequency current response by... and Subtracting these two values ​​gives the position tracking error signal:

[0046]

[0047] Step 4.3 The obtained position tracking error signal is used in the form of a quadrature phase-locked loop. The PI controller is adjusted to control ε to converge to zero, thereby obtaining the speed and rotor position information.

[0048] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0049] 1) The present invention adopts a scheme of injecting a high-frequency voltage signal into the estimated direct axis and extracting the high-frequency current response in the two-phase stationary axis system. Compared with the traditional pulsating high-frequency voltage signal injection method, which extracts the high-frequency current response in the estimated quadrature axis, this method can reduce the sensitivity of obtaining rotor position information to motor inductance parameters, avoid the risk of position estimation error convergence, and is more suitable for surface-mounted permanent magnet synchronous motors.

[0050] 2) This invention uses square wave voltage signal injection. Compared with the sinusoidal voltage signal injection of the traditional pulse high-frequency voltage signal injection method, the injection frequency can be increased to half of the PWM switching frequency, or even to the PWM switching frequency. In the current loop processing, since the injected square wave voltage frequency is much greater than the motor operating frequency, the base frequency current of two adjacent current samplings remains unchanged within one switching cycle. The base frequency feedback current can be calculated using simple mathematical calculations, which effectively improves the current loop bandwidth.

[0051] 3) This invention decomposes the discrete high-frequency square wave Fourier into the sum of sine waves of different frequencies, thereby extracting the rotor position signal using a continuous signal demodulation method. There is no need to perform discrete differential operations on the current. Compared with traditional methods, it can effectively reduce processor computation and resource consumption.

[0052] 4) In the signal processing and demodulation process of extracting high-frequency response current and obtaining rotor position information, this invention adopts a cascaded SOGI method, which reduces the use of multiple filters and effectively avoids problems such as signal amplitude attenuation and phase lag caused by the use of filters, thereby increasing the accuracy of rotor position estimation.

[0053] 5) The amplitude of the high-frequency signal after cascading SOGI is related to the frequency and amplitude of the injected voltage and the inductance parameters of the motor itself. This invention improves the robustness to the amplitude and inductance value of the injected signal through per-unit processing, and the parameter tuning of the quadrature phase-locked loop PI regulator is simpler. Attached Figure Description

[0054] Figure 1 A schematic diagram of a method for detecting rotor position and speed information of a surface-mounted permanent magnet synchronous motor.

[0055] Figure 2 A diagram showing the relationship between the various coordinate systems;

[0056] Figure 3 A diagram showing the relationship between the injected high-frequency square wave signal and the PWM carrier signal;

[0057] Figure 4 This is a block diagram of a second-order generalized integrator (SOGI).

[0058] Figure 5 A block diagram illustrating the principle of the method for estimating rotational speed and extracting rotor position;

[0059] Figure 6 Simulation waveforms of the actual rotor position, estimated rotor position, and position estimation error obtained using the control method of this invention when the motor is running unloaded at 50 rpm.

[0060] Figure 7 Simulation waveforms of the actual rotor position, estimated rotor position, and position estimation error obtained by the traditional pulsed sinusoidal voltage signal injection control method when the motor is running unloaded at 50 rpm.

[0061] Figure 8 Simulation waveforms of the actual rotor position, estimated rotor position, and estimated position error obtained by the control method of this invention when the motor speed jumps from 50 rpm to 100 rpm under no-load operation.

[0062] Figure 9 The simulation waveforms of the actual rotor position, estimated rotor position, and estimated position error obtained by using the control method of this invention are shown for a motor running under no-load conditions when its speed suddenly changes from 50 rpm to -50 rpm. Detailed Implementation

[0063] The present invention will now be described in detail with reference to the accompanying drawings.

[0064] like Figure 1 As shown, this invention provides a method for detecting the rotor position and speed information of a surface-mounted permanent magnet synchronous motor. Figure 1 middle For a given angular velocity value, The angular velocity value estimated by the rotor position observer. The given value for the d-axis current. The q-axis current setpoint output by the speed regulator, u d u q These are the voltage setpoints output from the d-axis and q-axis current regulators, respectively, i d i q These are the d-axis and q-axis feedback currents, respectively, i α i β These are the α-axis and β-axis feedback currents, u α u β The α and β axis voltage setpoints are respectively, and SVPWM is a space vector modulation module. a i b The stator currents sampled from phases a and b are U, respectively. dc Given the DC bus voltage of the inverter and PMSM as a permanent magnet synchronous motor, we obtain i. α i βThe high-frequency current response and square wave signal processing are implemented as described in step 2; the signal demodulation is implemented as described in step 3; and the speed and rotor position observation are implemented as described in step 3. The specific implementation steps are as follows:

[0065] Step 1, with a phase difference of Δθ from the actual dq axis e To estimate the error angle, a high-frequency square wave voltage signal with constant amplitude is injected into the dq-axis coordinate system to obtain a high-frequency current response.

[0066] In step 1, the pulsed high-frequency square wave voltage signal injected into the estimated dq-axis coordinate system is specifically as follows:

[0067]

[0068] In the formula, u dh The voltage signal injected into the d-axis, u qh U is the voltage signal injected into the q-axis. inj Let be the amplitude of the injected square wave voltage, k be the control sequence, and k = 1, 2, 3...

[0069] Figure 3 To illustrate the relationship between the injected high-frequency square wave signal and the carrier signal, the frequency of the injected high-frequency signal is half the PWM carrier frequency.

[0070] Using i d Vector control strategy for permanent magnet synchronous motors with dual closed-loop control of current loop and speed loop at 0, such as... Figure 2 The coordinate transformation is as follows: the two-phase stationary coordinate system is obtained by Clark transformation of the natural coordinate system ABC axis, where the α axis coincides with the A axis of the natural coordinate system, and the β axis is perpendicular to the α axis and coincides with the α axis rotated 90° counterclockwise. The rotating coordinate system is obtained by Park rotor position angle transformation of the two-phase stationary coordinate system αβ axis, where the d axis differs from the α axis by θ along the rotor rotation direction. e Angle, the q-axis is perpendicular to the d-axis and coincides with the d-axis rotated 90° counterclockwise.

[0071] u d To maintain the normal operation of the PMSM, the voltage is the sum of the injected high-frequency voltage; then, space vector pulse width modulation (SVPWM) is used to obtain the six switching signals of the three-phase inverter, thereby driving the SPMSM.

[0072] Step 2, according to Figure 2 The coordinate system relationship diagram is used to transform the obtained high-frequency current response from the estimated dq axis coordinate system to the two-phase stationary α-β coordinate system. Fourier decomposition is then used to decompose the high-frequency square wave current response in the stationary α-β coordinate system into the sum of sinusoidal signals of different frequencies.

[0073] The specific process in step 2 is as follows:

[0074] Step 2.1: During motor operation, the current of any two phases of the permanent magnet synchronous motor is collected through the current sampling module, and the resulting three-phase current i is... a i b i c The two-phase stationary coordinate current i is obtained after Clark transformation. α i β This includes the fundamental frequency current i αf i βf High-frequency current i αh i βh And the high-order harmonic current i generated by the inverter power devices αx i βx Three components;

[0075] Among them, the high-frequency response current i αh i βh The differential equation can be expressed as:

[0076]

[0077] In the formula, L dh L qh For the high-frequency inductors of the d and q axes of a permanent magnet synchronous motor, θ e Δθ represents the actual rotor position. e It is the estimation error between the actual rotor position and the estimated rotor position, where P is the differential operator;

[0078] When Δθ e When it is sufficiently small to be close to zero, the above formula can be written as:

[0079]

[0080] Step 2.2: Use Fourier decomposition to decompose the pulsating high-frequency square wave voltage signal in the stationary α-β coordinate system into a sum of sinusoidal signals of different frequencies:

[0081]

[0082] In the formula, ω h The frequency of the injected high-frequency square wave voltage signal is given, and t is the time interval. It represents an odd-order sine wave.

[0083] Substituting the sum of the sinusoidal signals obtained from the Fourier decomposition of the square wave voltage signal into the high-frequency response current equation in the two-phase stationary α-β coordinate system:

[0084]

[0085] Integrating both sides of the equation, we get:

[0086]

[0087] Figure 4 This is a block diagram of a second-order generalized integrator (SOGI), where x is the input signal, y1 and y2 are the output signals, and ω is the center frequency.

[0088] Figure 5 The diagram shows the structure of the cascaded SOGI for high-frequency signal extraction and demodulation along the α-axis. The structure of the β-axis is the same as that of the α-axis.

[0089] Step 3: Combine the high-frequency current i in the two-phase stationary coordinate system, which is the sum of sinusoidal signals of different frequencies. aβh After passing through a cascaded second-order generalized integrator (SOGI) preamplifier, a high-frequency current signal with the same injection frequency is extracted. Then, a cosine modulated wave with the same frequency as the injection signal is multiplied with the high-frequency current signal and passed through a cascaded SOGI postamplifier to suppress the influence of high-order harmonics contained in the demodulation on the estimated rotor position.

[0090] Step 3 is as follows:

[0091] Step 3.1, set the center frequency of the preamplifier of the cascaded SOGI to ω. h1 The current i in the two-phase stationary coordinate system α i β After passing through the cascaded SOGI preamplifier, a high-frequency current signal with the same injection frequency is extracted, suppressing interference from other subharmonics and realizing the function of a bandpass filter.

[0092] The pre-stage transfer function of the cascaded SOGI is:

[0093]

[0094] Wherein, the input signal x1 is the high-frequency current signal i αh1 i βh1 The output signal y1 is the extracted center frequency ω h1 The high-frequency response current, where k1 is the damping coefficient.

[0095] Step 3.2, using a cosine modulation wave with the same frequency as the injected signal, cosω h Multiplying t by the high-frequency current signal described in step 3.1 yields

[0096]

[0097] The above formula contains a signal with the same injection frequency as the signal and its higher frequency signals. Therefore, as long as the signal with the same injection frequency can be obtained, the estimated rotor position information can be calculated.

[0098] Step 3.3, multiply by the demodulated signal cosωh After t, as can be seen from the above formula, the second term in the demodulated signal is the higher harmonic component, and the first term is the direct component containing position information. Therefore, it is considered to use the subsequent stage of cascaded SOGI to filter out the second harmonic with higher component content, so as to achieve the same effect as low-pass filtering in rotor position observation and suppress the influence of the higher harmonics contained in the demodulation on the estimated rotor position.

[0099] The transfer function of the subsequent stage of the cascaded SOGI is:

[0100]

[0101] Wherein, the input signal x2 is the signal after passing through cosω h The high-frequency response current signal i after demodulation αh1 i βh1 The output signal y2 is the extracted through component i containing location information. αh2 i βh2 k2 is the damping coefficient.

[0102] Figure 6 This is a structural block diagram of the high-frequency signal extraction, demodulation, and rotor position information acquisition method proposed in this invention.

[0103] Step 4: The modulated high-frequency current responses of the α and β axes are normalized, and the rotor position and speed information are extracted by an orthogonal phase-locked loop composed of a PI controller for sensorless control of the SPMSM.

[0104] Step 4 is as follows:

[0105] Step 4.1: The amplitude of the high-frequency signal after cascading SOGI is related to the frequency and amplitude of the injected voltage, as well as the inductance parameters of the motor itself. Therefore, per-unit processing is used to improve the robustness to the amplitude and inductance value of the injected signal.

[0106]

[0107] Step 4.2, according to the heterodyne method, multiply the high-frequency current response by... and Subtracting these two values ​​gives the position tracking error signal:

[0108]

[0109] Step 4.3: The obtained position tracking error signal is used in the form of a quadrature phase-locked loop. By adjusting the PI controller to control ε to converge to zero, the rotor position estimate converges to the actual value, thereby obtaining the speed and rotor position information for dual closed-loop control of speed and current.

[0110] To verify the correctness and effectiveness of the proposed scheme, a sensorless SPMSM control model based on high-frequency square wave signal injection was built on the Matlab / Simulink simulation platform. The motor used in the simulation was a 400W low-voltage servo motor, and the motor parameters are shown in Table 1. The simulation conditions were set as follows: PWM switching frequency of 10kHz, injected square wave frequency of 5kHz, and amplitude of 6V.

[0111]

[0112] Table 1 - SPMSM Parameter Table

[0113] Figure 6 , Figure 8 and Figure 9 The simulation results using the rotor position observation method proposed in this patent show that the estimated rotor position can accurately follow the actual rotor position angle without zero-point drift. Figure 6 As can be seen from this, the maximum electrical angle error is 0.025 rad.

[0114] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for detecting rotor position and speed information of a surface-mounted permanent magnet synchronous motor, characterized in that, Includes the following steps: Step 1: Inject a pulsed high-frequency square wave voltage signal with constant amplitude into the estimated dq-axis coordinate system to obtain the high-frequency current response; The pulsed high-frequency square wave voltage signal injected into the estimated dq-axis coordinate system is specifically as follows: In the formula, u dh The voltage signal injected into the d-axis, u qh U is the voltage signal injected into the q-axis. inj The amplitude of the injected square wave voltage is given by k, which is the control sequence. The frequency of the injected high-frequency signal is half of the PWM carrier frequency. Step 2 involves transforming the obtained high-frequency current response from the estimated dq-axis coordinate system to the two-phase stationary α-β coordinate system. Fourier decomposition is then used to decompose the high-frequency square wave current component in the stationary α-β coordinate system into a sum of sinusoidal signals of different frequencies. The specific process includes: Step 2.1: During motor operation, the current of any two phases of the permanent magnet synchronous motor is collected through the current sampling module, and the resulting three-phase current i is... a i b i c The two-phase stationary coordinate current i is obtained after Clark transformation. α i β This includes the fundamental frequency current i αf i βf High-frequency response current i αh i βh And the high-order harmonic current i generated by the inverter power devices αx i βx Three components; Among them, the high-frequency response current i αh i βh The differential equation is expressed as: In the formula, L dh L qh For a permanent magnet synchronous motor, θe represents the high-frequency inductance of the d and q axes, θe represents the actual rotor position, and Δθ represents the high-frequency inductance of the rotor. e It is the estimation error between the actual rotor position and the estimated rotor position, where P is the differential operator; When Δθ e When it is sufficiently small to be close to zero, the above formula can be written as: Step 2.2: Use Fourier decomposition to decompose the pulsating high-frequency square wave voltage signal into a sum of sinusoidal signals of different frequencies: In the formula, ω h The frequency of the injected high-frequency square wave voltage signal is given, and t is the time interval. Represents odd-order sine waves; Substituting the sum of the sinusoidal signals obtained from the Fourier decomposition of the square wave voltage signal into the high-frequency response current differential equation in the two-phase stationary α-β coordinate system: The high-frequency response current i in the stationary axis system is obtained by integral calculation. αh i βh equation; Step 3: Combine the high-frequency current i in the two-phase stationary coordinate system, which is the sum of sinusoidal signals of different frequencies. αh i βh After passing through the cascaded second-order generalized integrator SOGI pre-stage, the high-frequency current signal with the same injection frequency is extracted. Then, the high-frequency current signal is multiplied by the cosine modulation wave with the same frequency as the injection signal and passed through the cascaded second-order generalized integrator SOGI post-stage to suppress the influence of the high-order harmonics contained in the demodulation on the estimated rotor position. Step 4: Normalize the modulated high-frequency current response and extract the rotor position and speed information through an orthogonal phase-locked loop composed of a PI controller.

2. The method for detecting rotor position and speed information of a surface-mounted permanent magnet synchronous motor according to claim 1, characterized in that, The specific process of step 3 includes: Step 3.1: Set the center frequency of the pre-stage of the cascaded second-order generalized integrator SOGI to ω. h1 The current i in the two-phase stationary coordinate system α i β After passing through the pre-stage of the cascaded second-order generalized integrator SOGI, the high-frequency current signal with the same injection frequency is extracted, suppressing the interference of other subharmonics and realizing the function of a bandpass filter. Step 3.2, using a cosine modulation wave with the same frequency as the injected signal, cosω h Multiplying t by the high-frequency current signal described in step 3.1 yields: Step 3.3: The second harmonic is filtered out by the stage of the cascaded second-order generalized integrator SOGI to suppress the influence of the higher harmonics contained in the demodulation on the estimated rotor position.

3. The method for detecting rotor position and speed information of a surface-mounted permanent magnet synchronous motor according to claim 1, characterized in that, The specific process of step 4 includes: Step 4.1: Improve robustness to injected signal amplitude and inductance value through per-unit processing. Step 4.2, according to the heterodyne method, multiply the high-frequency current response by... and Subtracting these two values ​​gives the position tracking error signal ε: In the formula, i α-pu i β-pu θ represents the current along the α-β axis of the stationary axis system after standardization. e This represents the actual rotor position; Step 4.3, the obtained position tracking error signal i αh i βh An orthogonal phase-locked loop is used, and the PI controller is adjusted to control ε to converge to zero, thereby obtaining the speed and rotor position information.

Images