Position information demodulation method for PMSM position sensorless control

Abstract

The invention discloses a position information demodulation method for PMSM position sensorless control, and relates to the technical field of motor position-less control. An improved phase-locked loop ASFEF-SOGI-QPLL is designed, based on a high-frequency injection method, a high-frequency square wave signal is injected at the d-axis voltage of a motor, current response located on the alpha-axis and the beta-axis of a static coordinate system is extracted, an orthogonal component signal containing position information is extracted through a position observer and sent into the improved phase-locked loop ASFEF-SOGI-QPLL, the orthogonal component signal is processed through a pre-filtering link and SOGI, and the current response of the alpha-axis and the beta-axis of the static coordinate system is obtained. And performing parameter setting on the phase-locked loop, determining a proportionality coefficient and an integral coefficient of a PI controller, and obtaining estimated rotor position and speed information. The invention provides an improved phase-locked loop ASFEF-SOGI-QPLL, which can reduce the harmonic sensitivity in position-free control, significantly improve the reading precision of position and speed information, and ensure the accuracy of demodulation information.

Description

Technical Field

[0001] This invention relates to the field of positionless motor control technology, specifically a position information demodulation method for sensorless control of PMSM (Motor-Mounted Smart Motor). Background Technology

[0002] A phase-locked loop (PLL) consists of a phase detector (PD), a loop filter (LF), and a voltage-controlled oscillator (VCO) forming a closed-loop system. By tracking the phase of the motor's quadrature signal, it estimates the rotor position and speed in real time, replacing mechanical sensors. Compared to the tangent function method, the closed-loop structure of the PLL actively suppresses noise and harmonic interference, avoids the phase jump problem of open-loop calculations, and significantly improves dynamic response speed and robustness, making it highly practical in engineering. A quadrature phase-locked loop (QPLL) generates an error signal by multiplying the quadrature signal by the sine and cosine values ​​of the estimated angle. This error signal is then used to drive the VCO via a PI controller to achieve phase locking. Due to its advantages such as high-precision phase synchronization, smooth output, fast dynamic response, simple structure, and good digitization, it is widely used in various industrial scenarios.

[0003] High-frequency signal injection methods, which extract position information based on the salient polarity of the motor, are commonly used for sensorless control of permanent magnet synchronous motors in the zero-low-speed domain. A typical block diagram of a sensorless driver based on high-frequency injection is provided. Figure 1 As shown, the high-frequency signal injected into the d-axis of the motor will excite a high-frequency current response containing position information in the rotating coordinate system. The signal is then extracted by the observer and demodulated by the phase-locked loop to complete the control closed loop.

[0004] However, the quadrature phase-locked loop (PLL), which is most widely used in positionless control of permanent magnet synchronous motors, still has the following problems:

[0005] 1. The nonlinear characteristics of the inverter and the influence of magnetic flux space harmonics can cause errors between the output voltage and the reference voltage, thereby generating harmonics mainly of the 5th and 7th order in the phase current. Quadrature phase-locked loops lack the ability to suppress harmonic components.

[0006] 2. The PI parameter tuning of the quadrature phase-locked loop depends on the motor parameters, and a trade-off between dynamic response and noise suppression is required, resulting in weak robustness of the positionless control system.

[0007] 3. The operating conditions of motors typically cover both low and high speed ranges. Ideally, the bandwidth design should reduce noise at low speeds and increase it at high speeds to reduce delay. However, the fixed bandwidth of a quadrature phase-locked loop (PLL) will limit the dynamic performance of the system.

[0008] In summary, directly applying the position information of a permanent magnet synchronous motor without position control using a quadrature phase-locked loop (PLL) has limitations in terms of accuracy in current applications. Summary of the Invention

[0009] To address the shortcomings of the prior art, this invention provides a position information demodulation method for sensorless PMSM control. It proposes an improved phase-locked loop (ASFEF-SOGI-QPLL) that can reduce harmonic sensitivity in sensorless control, significantly improve the accuracy of position and speed information reading, and ensure the accuracy of demodulated information.

[0010] To achieve the above objectives, the present invention adopts the following technical solution: a method for demodulating position information for sensorless control of PMSM, comprising the following steps:

[0011] S1. Based on the quadrature phase-locked loop (QPLL), two pre-filtering stages composed of adaptive synchronous frequency filters (ASFEF) are added to its front end. A second-order generalized integrator (SOGI) is added to the feedforward channel of the PI controller to design an improved phase-locked loop (ASFEF-SOGI-QPLL). In the positionless control of the PMSM based on the high-frequency injection method, a high-frequency square wave signal is injected at the d-axis voltage of the motor. The current response located on the αβ axis of the stationary coordinate system is extracted. The quadrature component signal containing position information is extracted by the position observer and sent to the improved phase-locked loop (ASFEF-SOGI-QPLL).

[0012] S2. The quadrature component signals are fed into the pre-filtering stage for processing. ASFEF acts as a bandpass filter, and the output is as follows:

[0013]

[0014] In the formula, To estimate the rotor position, σ is the sharpness coefficient used to adjust the filtering strength of ASFEF, e1 and e h These represent the fundamental and high-frequency harmonic components of the input signal, respectively, A0, ω0, and... 0 represents the amplitude, angular frequency, and phase of the fundamental frequency current component, respectively;

[0015] S3. The orthogonal component signals after pre-filtering are multiplied by the trigonometric function values ​​of the estimated rotor position to calculate the rotor position error ε. The rotor position error ε is adjusted by the PI controller to obtain the estimated rotor speed. After integration, the estimated rotor position is obtained. The estimated rotor position is then used as a feedback signal to participate in the calculation of the rotor position error ε, thus completing the closed loop of the phase-locked loop. SOGI acts as a band-stop filter to filter out the sixth harmonic.

[0016] S4. Perform parameter tuning on the improved phase-locked loop ASFEF-SOGI-QPLL to determine the proportional and integral coefficients of the PI controller and obtain estimated rotor position and speed information.

[0017] Furthermore, in step S1, the current response of the stationary coordinate system αβ axis is expressed as:

[0018]

[0019] In the formula, U m Let p be the amplitude of the injected signal, and L be the differential operator. d and L q These are the d-axis and q-axis inductances, respectively, L avg L is the average value of the dq axis inductance. dif The difference between half the dq-axis inductance is θ, where θ is the actual rotor position.

[0020] Furthermore, in step S2, the orthogonal component signals entering the pre-filtering stage are represented as follows:

[0021]

[0022] In the formula, A h ω h and h These represent the amplitude, angular frequency, and phase of the high-frequency current component, respectively.

[0023] Furthermore, in step S2, the transfer function of the adaptive synchronization frequency filter ASFEF is as follows:

[0024]

[0025] In the formula, ω is the angular frequency.

[0026] Furthermore, in step S3, the transfer function of the second-order generalized integrator SOGI is expressed as follows:

[0027]

[0028] In the formula, ω r The value is six times the estimated rotor speed, where is the target frequency to be filtered out, and the value of k is used to adjust the blocking strength of SOGI for high-frequency signals.

[0029] Furthermore, in step S4, during the parameter tuning process, the bandwidth of the improved phase-locked loop ASFEF-SOGI-QPLL is fixed, expressed as:

[0030]

[0031] The main transfer function is expressed as:

[0032]

[0033] In the formula, and These are the proportional and integral coefficients of the PI controller, respectively, ξ is the damping coefficient, and ω is the integral coefficient. n It is the natural frequency.

[0034] Furthermore, the damping coefficient is taken as 0.8~1.2.

[0035] Compared with existing technologies, the beneficial effects of this invention are as follows: Based on the concept of adaptive filtering, this invention introduces two ASFEFs and one SOGI into the traditional quadrature phase-locked loop structure to form an improved phase-locked loop ASFEF-SOGI-QPLL. According to the transfer function, ASFEFs act as bandpass filters and SOGIs act as bandstop filters. The combined application of the two extracts the fundamental wave used for rotor position estimation, which can adaptively eliminate the 6th harmonic component of the position estimation error in the feedforward channel. The extracted position and speed information is fed back to the motor closed-loop control system. Compared with traditional quadrature phase-locked loops, it reduces the harmonic sensitivity in positionless control, significantly improves the reading accuracy of position and speed information, ensures the accuracy of demodulated information, and is conducive to improving the overall performance and accuracy of sensorless control systems. Attached Figure Description

[0036] Figure 1 This is a block diagram of positionless control for a traditional permanent magnet synchronous motor based on the high-frequency injection method;

[0037] Figure 2 This is a structural block diagram of the improved phase-locked loop ASFEF-SOGI-QPLL in this invention;

[0038] Figure 3 This is a structural block diagram of ASFEF in this invention;

[0039] Figure 4 These are Bode plots of ASFEF under different sharpness coefficients in this invention;

[0040] Figure 5 This is a structural block diagram of SOGI in this invention;

[0041] Figure 6 These are Bode plots of SOGI under different k values ​​according to the present invention;

[0042] Figure 7 This is a comparison chart of the estimated position signals obtained by SOGI-QPLL and ASFEF-SOGI-QPLL processing in the embodiment;

[0043] Figure 8 This is a comparison chart of position error and velocity error obtained by different phase-locked loop processing in the embodiments. Detailed Implementation

[0044] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the invention, not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0045] like Figures 2-6 As shown, a method for demodulating position information for sensorless control of a PMSM includes the following steps:

[0046] S1. Design an improved phase-locked loop ASFEF-SOGI-QPLL, whose structure combines... Figure 2 As shown, based on the quadrature phase-locked loop (QPLL), two pre-filtering stages consisting of adaptive synchronous frequency filters (ASFEF) are added to its front end to optimize signal quality and ensure the dynamic performance of the system. A second-order generalized integrator (SOGI) is added to the feedforward channel of the PI controller to filter out the sixth harmonic and improve the steady-state characteristics of the system. The reference signals of both are from the input of the phase-locked loop, which can adaptively realize signal extraction at different rotor speeds.

[0047] In positionless control of a PMSM based on high-frequency injection, a high-frequency square wave signal is injected at the d-axis voltage of the motor. The current response along the αβ axis in the stationary coordinate system is extracted, and the orthogonal component signal containing position information is extracted by a position observer and fed into an improved phase-locked loop (PLL) ASFEF-SOGI-QPLL. The current response along the αβ axis can be expressed as:

[0048]

[0049] In the formula, U m Let p be the amplitude of the injected signal, and L be the differential operator. d and L q These are the d-axis and q-axis inductances, respectively, L avg L is the average value of the dq axis inductance. dif The difference between half the dq-axis inductance is θ, where θ is the actual rotor position. To estimate the rotor position.

[0050] S2. After the quadrature component signals are fed into the improved phase-locked loop ASFEF-SOGI-QPLL, position information is extracted. A pair of quadrature component signals contains a large number of high-order harmonics, and the two are 90° out of phase. The quadrature component signals entering the pre-filtering stage can be expressed as:

[0051]

[0052] In the formula, A0 and A hLet ω0 and ω be the amplitudes of the fundamental frequency and high-frequency current components, respectively. h These are the angular frequencies of the fundamental frequency and the high-frequency current component, respectively. 0 and h These are the phases of the fundamental frequency and the high-frequency current component, respectively.

[0053] The orthogonal component signals are first pre-filtered by an adaptive synchronous frequency filter (ASFEF), whose structure combines... Figure 3 As shown, the transfer function is expressed as:

[0054]

[0055] In the formula, ω is the angular frequency, and σ is the sharpness coefficient, used to adjust the filtering strength of ASFEF. The Bode plots corresponding to different sharpness coefficients are combined with... Figure 4 As shown, the output of ASFEF can be represented as follows:

[0056]

[0057] In the formula, e1 and e h These are the fundamental frequency component and the high-frequency harmonic component of the input signal, respectively.

[0058] The ASEFE function as a bandpass filter, ensuring that the fundamental component passes through without attenuation, while the higher the frequency of the harmonic components, the greater the attenuation of the signal amplitude. The reference signal for the ASEFE is obtained using an estimated rotor position derived from a phase-locked loop, thus allowing it to adaptively adjust to the motor system. Since neither the poles nor the zeros of the ASEFE have positive real parts, it remains stable across the entire speed range of the motor.

[0059] S3. The orthogonal component signals after pre-filtering are multiplied by the trigonometric function values ​​of the estimated rotor position, and the rotor position error is calculated as follows:

[0060]

[0061] The rotor position error ε is adjusted by a PI controller to obtain the estimated rotor speed. After integration, the estimated rotor position is obtained. The estimated rotor position is then used as a feedback signal to calculate the rotor position error ε, completing the phase-locked loop (PLL) closure. By selecting appropriate PI parameters, the improved PLL ASFEF-SOGI-QPLL can accurately estimate rotor position and speed. The transfer function corresponding to the quadrature PLL is:

[0062]

[0063] In the formula, and These are the proportional and integral coefficients of the PI controller, respectively.

[0064] A second-order generalized integrator SOGI is added to the feedforward channel of the PI controller to filter out the sixth harmonic. Its structure combines... Figure 5 As shown, the transfer function is expressed as:

[0065]

[0066] In the formula, ω r The value is six times the estimated rotor speed. The k value is used to adjust the blocking strength of SOGI for high-frequency signals. The Bode plots corresponding to different k values ​​are combined. Figure 6 As shown.

[0067] SOGI behaves as a band-stop filter, ω r To filter out the target frequency, the SOGI reference signal is obtained using the estimated rotor speed from the phase-locked loop (PLL), thus enabling adaptive adjustments to follow the motor system. If the input signal contains high-order harmonics, these harmonic components reduce the bandwidth of the PLL's loop filter, affecting the system's transient response. This degradation is more pronounced at low speeds, and the introduced pre-filter effectively addresses this issue. The combined system components yield highly accurate estimated rotor position and speed information.

[0068] S4. Harmonic components of the input signal reduce the bandwidth of the PI controller. The added ASEFE pre-filter effectively extracts the fundamental component of the input signal, essentially solving this problem. Therefore, the key to parameter tuning lies in selecting the overall bandwidth of the phase-locked loop. From the above transfer function, it can be seen that the bandwidth of the improved phase-locked loop ASFEF-SOGI-QPLL is fixed, expressed as:

[0069]

[0070] The main transfer function can be expressed as:

[0071]

[0072] In the formula, ξ is the damping coefficient, ω n It is the natural frequency.

[0073] This is equivalent to adding a closed-loop zero to the standard second-order transfer function. The presence of the closed-loop zero will increase the system response speed, but also increase the overshoot. For the standard second-order transfer function, the system is usually designed as underdamped, with a damping coefficient of 0.4 to 0.8. Considering the influence of the closed-loop zero, a damping coefficient of 0.8 to 1.2 is used.

[0074] This invention is applied to a dual closed-loop control system for permanent magnet synchronous motors. By combining the bandwidth of the current controller and the bandwidth of the speed controller, and selecting the specific phase-locked loop bandwidth, the PI controller's bandwidth can be determined. and Specific values ​​are used to complete the tuning of system parameters, in order to Figure 6 Taking the experimental object shown as an example, the bandwidth of its current controller and the bandwidth of its speed controller are 10kHz and 1kHz, respectively, and the phase-locked loop bandwidth is selected as 1.24kHz. Since the overall system has strong harmonic removal capability, a larger phase-locked loop bandwidth can be selected without considering the influence of higher-order harmonic noise.

[0075] Example

[0076] Figure 7 The invention demonstrates the motor rotor position information obtained by processing high-frequency current signals in a permanent magnet synchronous motor at a bus voltage of 80V and a given speed of 120rpm, using both SOGI-QPLL and the ASFEF-SOGI-QPLL proposed in this invention.

[0077] Figure 8 The paper demonstrates the position and speed errors of a permanent magnet synchronous motor (PMSM) under different phase-locked loop (PLL) processing of high-frequency current signals at a bus voltage of 80V and a given speed of 120rpm. The high-frequency injection method uses a square wave with an injection frequency half the switching frequency. Figure a) shows the processing results of QPLL, b) shows the processing results of SOGI-QPLL, and c) shows the processing results of ASFEF-SOGI-QPLL.

[0078] It can be seen that when using QPLL, the peak-to-peak values ​​of the motor's speed and position errors are 30.18 r / min and 14.18°, respectively, with an average position error of 4.63°. When using SOGI-QPLL, the peak-to-peak values ​​of the motor's speed and position errors are 27.63 r / min and 10.69°, respectively, with an average position error of 3.71°. This indicates that the introduction of SOGI improves the steady-state performance of sensorless control, reducing the peak position error by 24.61%, the average position error by 19.8%, and the maximum speed error by 8.44%. When using ASFEF-SOGI-QPLL, the peak-to-peak values ​​of the motor's speed and position errors are 10.18 r / min and 8.29°, respectively, with an average position error of 1.96°. It can be seen that the application of ASFEF-SOGI-QPLL reduces the peak position error by 41.54%, the average position error by 57.67%, and the maximum speed error by 66.27%.

[0079] Experimental results show that, among the three methods, the ASFEF-SOGI-QPLL of this invention performs best in eliminating harmonic fluctuations and can achieve high-precision positionless control of permanent magnet synchronous motors.

[0080] In summary, this invention extracts high-frequency current containing position information from the motor's rotating coordinate system, processes it using a filterless carrier separation strategy to obtain orthogonal signals, and then transmits them to a designed improved phase-locked loop (PLL) ASFEF-SOGI-QPLL. Through the addition of a pre-filtering stage and a second-order generalized integrator in the feedforward channel, the signal frequency is dynamically tracked, accurately filtering out high-order harmonics in the target signal, thus obtaining high-precision position and velocity information and ensuring the stability and reliability of the control system. This method can be implemented in standard digital control chips (such as DSPs or FPGAs) without additional hardware modifications to improve the control accuracy and reliability of motors. It is applicable to the widely used positionless control of permanent magnet synchronous motors and has extremely high application value in the field of industrial control.

[0081] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered illustrative and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of the equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0082] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A method for demodulating position information for sensorless control of PMSM, characterized in that: Includes the following steps: S1. Based on the quadrature phase-locked loop (QPLL), two pre-filtering stages composed of adaptive synchronous frequency filters (ASFEF) are added to its front end. A second-order generalized integrator (SOGI) is added to the feedforward channel of the PI controller to design an improved phase-locked loop (ASFEF-SOGI-QPLL). In the positionless control of the PMSM based on the high-frequency injection method, a high-frequency square wave signal is injected at the d-axis voltage of the motor. The current response located on the αβ axis of the stationary coordinate system is extracted. The quadrature component signal containing position information is extracted by the position observer and sent to the improved phase-locked loop (ASFEF-SOGI-QPLL). S2. The quadrature component signals are fed into the pre-filtering stage for processing. ASFEF acts as a bandpass filter, and the output is as follows: In the formula, To estimate the rotor position, σ is the sharpness coefficient used to adjust the filtering strength of ASFEF, e1 and e h These represent the fundamental and high-frequency harmonic components of the input signal, respectively, A0, ω0, and... These represent the amplitude, angular frequency, and phase of the fundamental frequency current component, respectively. S3. The orthogonal component signals after pre-filtering are multiplied by the trigonometric function values ​​of the estimated rotor position to calculate the rotor position error ε. The rotor position error ε is adjusted by the PI controller to obtain the estimated rotor speed. After integration, the estimated rotor position is obtained. The estimated rotor position is then used as a feedback signal to participate in the calculation of the rotor position error ε, thus completing the closed loop of the phase-locked loop. SOGI acts as a band-stop filter to filter out the sixth harmonic. S4. Perform parameter tuning on the improved phase-locked loop ASFEF-SOGI-QPLL to determine the proportional and integral coefficients of the PI controller and obtain estimated rotor position and speed information.

2. The method for demodulating position information for sensorless control of PMSM according to claim 1, characterized in that: In step S1, the current response of the stationary coordinate system αβ axis is expressed as: In the formula, U m Let p be the amplitude of the injected signal, and L be the differential operator. d and L q These are the d-axis and q-axis inductances, respectively, L avg L is the average value of the dq axis inductance. dif The difference between half the dq-axis inductance is θ, where θ is the actual rotor position.

3. The method for demodulating position information for sensorless control of PMSM according to claim 1, characterized in that: In step S2, the quadrature component signals entering the pre-filtering stage are represented as follows: In the formula, A h ω h and h These represent the amplitude, angular frequency, and phase of the high-frequency current component, respectively.

4. The method for demodulating position information for sensorless control of PMSM according to claim 1, characterized in that: In step S2, the transfer function of the adaptive synchronization frequency filter ASFEF is as follows: In the formula, ω is the angular frequency.

5. A method for demodulating position information for sensorless control of a PMSM according to claim 1, characterized in that: In step S3, the transfer function of the second-order generalized integrator SOGI is expressed as follows: In the formula, ω r The value is six times the estimated rotor speed, where is the target frequency to be filtered out, and the value of k is used to adjust the blocking strength of SOGI for high-frequency signals.

6. The method for demodulating position information for sensorless control of PMSM according to claim 1, characterized in that: In step S4, during the parameter tuning process, the bandwidth of the improved phase-locked loop ASFEF-SOGI-QPLL is fixed, as shown below: The main transfer function is expressed as: In the formula, and These are the proportional and integral coefficients of the PI controller, respectively, ξ is the damping coefficient, and ω is the integral coefficient. n It is the natural frequency.

7. A method for demodulating position information for sensorless control of PMSM according to claim 6, characterized in that: The damping coefficient is taken as 0.8~1.2.

Images