Permanent magnet synchronous motor rotor position estimation method based on high-frequency signal injection

Abstract

The invention relates to a permanent magnet synchronous motor rotor position estimation method based on high-frequency signal injection, and the method comprises the following steps: carrying out the amplitude modulation of a preset high-frequency voltage carrier signal through an orthogonal rotation modulator, and obtaining an injection signal vector; performing voltage excitation on a stator winding of the permanent magnet synchronous motor based on the injection signal vector to obtain a high-frequency current response signal; performing harmonic separation on the high-frequency current response signal through a comb filter bank to obtain a rotor position modulation current component; orthogonal demodulation is carried out on the rotor position modulation current component through a complex demodulator, and a position-dependent phase difference value is obtained; non-linear mapping processing is carried out on the position-dependent phase difference value through an adaptive state observer, a real-time rotor position angle value is obtained, and the key problem of how to effectively extract information related to the rotor position from a high-frequency current response signal is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of permanent magnet synchronous motors, and particularly to a method for estimating the rotor position of a permanent magnet synchronous motor based on high-frequency signal injection. Background Art

[0002] In the control field of permanent magnet synchronous motors (PMSMs), accurately estimating the rotor position is crucial for achieving an efficient and high-performance drive system. Traditionally, sensors such as Hall effect sensors or resolvers are used to directly measure the rotor position and speed information. However, these sensors increase the cost and complexity of the system and may reduce the system's reliability, especially in harsh working environments. Therefore, sensorless control technology has become a research hotspot, aiming to indirectly infer the rotor position by analyzing the electrical characteristics of the motor, thereby reducing the dependence on physical sensors.

[0003] Although sensorless control strategies have made significant progress in the past few decades, existing methods still face challenges in estimating the rotor position under low-speed and even zero-speed operating conditions. Especially in the high-frequency signal injection method, how to effectively extract the rotor position-related information from the complex high-frequency current response signal has become a key issue. Traditional filtering techniques and demodulation methods often cannot fully distinguish the weak signal components containing rotor position information, resulting in limited accuracy and stability of position estimation. In addition, the computational burden of the algorithm is also an issue that cannot be ignored. An overly complex processing flow will increase the real-time processing pressure of the controller.

[0004] In response to the problems in the above background, a new solution is proposed for the high-frequency signal injection-based method, that is, modulating the high-frequency voltage carrier signal through an orthogonal rotation modulator, combining a comb filter bank and harmonic separation technology to accurately capture the rotor position modulation current component, and then using a complex demodulator and an adaptive state observer to further process to obtain an accurate rotor position angle value. This method not only improves the position estimation accuracy under low-speed conditions but also optimizes the algorithm efficiency and reduces the demand for computing resources, providing strong support for promoting the practical application of sensorless control technology. Summary of the Invention

[0005] The main object of the present invention is to provide a method for estimating the rotor position of a permanent magnet synchronous motor based on high-frequency signal injection, which solves the key problem of how to effectively extract the rotor position-related information from the high-frequency current response signal.

[0006] To achieve the above object, the present invention provides a method for estimating the rotor position of a permanent magnet synchronous motor based on high-frequency signal injection, including the following steps: Amplitude-modulating a preset high-frequency voltage carrier signal through an orthogonal rotation modulator to obtain an injection signal vector; Voltage excitation is performed on the stator windings of the permanent magnet synchronous motor based on the injection signal vector to obtain a high-frequency current response signal; Harmonic separation is performed on the high-frequency current response signal through a comb filter bank to obtain a rotor position modulation current component; Quadrature demodulation is performed on the rotor position modulation current component through a complex demodulator to obtain a position-dependent phase difference; Nonlinear mapping processing is performed on the position-dependent phase difference through an adaptive state observer to obtain a real-time rotor position angle value.

[0007] Further, the obtaining of the injection signal vector by amplitude modulating a preset high-frequency voltage carrier signal through an orthogonal rotation modulator includes: Phase separation processing is performed on the preset high-frequency voltage carrier signal through an orthogonal rotation modulator to obtain orthogonal phase components; wherein, the orthogonal phase components include in-phase components and quadrature components; Coordinate rotation transformation is performed on the orthogonal phase components based on a preset reference coordinate system transformation matrix to obtain a rotation modulation vector, and space vector synthesis is performed on the rotation modulation vector to obtain a three-phase modulation signal; Nonlinear compensation processing is performed on the three-phase modulation signal through a voltage vector decoupler to obtain an injection signal vector.

[0008] Further, the obtaining of the high-frequency current response signal by voltage excitation of the stator windings of the permanent magnet synchronous motor based on the injection signal vector includes: Switching timing allocation is performed on the injection signal vector through a space vector pulse width modulator to obtain a multi-phase excitation voltage waveform, and voltage excitation is applied to the stator windings of the permanent magnet synchronous motor based on the multi-phase excitation voltage waveform to obtain the electromagnetic energy conversion characteristics of the stator windings; Based on the electromagnetic energy conversion characteristics of the stator windings, a magnetic field distribution model is established for the stator windings to obtain rotor salient pole effect parameters, and self-inductance and mutual inductance decoupling processing is performed on the rotor salient pole effect parameters to obtain a predicted value of the stator winding current excitation response; Multi-channel synchronous sampling is performed on the predicted value of the stator winding current excitation response through a phase current sampling circuit to obtain a high-frequency current response signal; wherein, the high-frequency current response signal includes a stator winding fundamental frequency response component and a stator winding high-frequency modulation component.

[0009] Further, the obtaining of the rotor position modulation current component by harmonic separation of the high-frequency current response signal through a comb filter bank includes: The high-frequency current response signal is synchronously demodulated by a lock-in amplifier to obtain a high-frequency carrier modulation signal, and the high-frequency carrier modulation signal is subjected to Hilbert transform to obtain a high-frequency current analytic signal; Based on the high-frequency current analytic signal, spectral zoom analysis is performed on the high-frequency current response signal to obtain a high-frequency current harmonic distribution matrix, and the high-frequency current harmonic distribution matrix is subjected to singular value decomposition to obtain a high-frequency current principal component eigenvector; The high-frequency current principal component eigenvector is subjected to harmonic separation by a comb filter bank to obtain a harmonic separation current vector, and the harmonic separation current vector is subjected to rotational coordinate transformation to obtain a rotor position modulation current vector; The amplitude and phase of the rotor position modulation current vector are extracted to obtain a rotor position modulation current component.

[0010] Further, the orthogonal demodulation of the rotor position modulation current component by the complex demodulator to obtain a position-dependent phase difference includes: The rotor position modulation current component is synchronously sampled by an orthogonal carrier generator to obtain a pair of orthogonal reference signals, and the pair of orthogonal reference signals is subjected to complex transformation processing to obtain a phase reference vector; Based on a mixer, a complex multiplication operation is performed on the rotor position modulation current component and the phase reference vector to obtain a spectrum-shifted component, and the spectrum-shifted component is subjected to low-pass filtering processing by a complex demodulator to obtain a baseband demodulation signal; A four-quadrant arctangent operation is performed on the baseband demodulation signal by a vector phase detector to obtain an initial phase angle measurement, and the initial phase angle measurement is subjected to periodic distortion compensation to obtain a corrected phase value; Based on a harmonic canceller, high-order harmonic suppression is performed on the corrected phase value to obtain a continuous phase trajectory, and the continuous phase trajectory is subjected to digital integration processing to obtain an accumulated phase value; The accumulated phase value is subjected to electrical angle mapping by a space vector analyzer to obtain a position-dependent phase difference; wherein, the position-dependent phase difference includes a magnetic field spatial distribution term and a salient pole effect coupling term.

[0011] Further, the non-linear mapping process of the position-dependent phase difference by the adaptive state observer to obtain a real-time rotor position angle value includes: Based on an elliptic curve mapper, angle non-linear correction is performed on the position-dependent phase difference to obtain an initial position estimate, and the initial position estimate is subjected to phase unwrapping processing to obtain a continuous angle sequence; The continuous angle sequence is subjected to time-domain integration operation through an error compensation integrator to obtain a cumulative phase error, and the cumulative phase error is subjected to polynomial curve fitting through an adaptive state observer to obtain a magnetic field disturbance component; Based on a flux saturation compensator, the magnetic field disturbance component is subjected to non-linear characteristic extraction to obtain a dynamic saturation factor, and the dynamic saturation factor and the continuous angle sequence are weighted and fused to obtain a corrected angle estimation value; The corrected angle estimation value is subjected to closed-loop tracking control through a phase-locked loop to obtain a real-time rotor position angle value, wherein the real-time rotor position angle value includes an electrical angle component and a mechanical angle component.

[0012] Further, the step of obtaining a cumulative phase error by subjecting the continuous angle sequence to time-domain integration operation through an error compensation integrator includes: The continuous angle sequence is subjected to sliding window segmentation processing through a phase differentiator to obtain an angle increment sequence, and the angle increment sequence is subjected to zero drift compensation operation to obtain a compensated angle sequence; Based on a harmonic analyzer, the compensated angle sequence is subjected to frequency-domain decomposition processing to obtain an angle harmonic component matrix, and the angle harmonic component matrix is subjected to phase reconstruction operation to obtain a phase error vector; The phase error vector is subjected to dynamic weight allocation through an adaptive integrator to obtain a weighted error sequence, and the weighted error sequence is subjected to non-linear integration operation through an error compensation integrator to obtain a cumulative error curve; Based on an error compensation network, time-varying characteristics of the cumulative error curve are extracted to obtain an error compensation factor, and the error compensation factor is subjected to adaptive correction processing to obtain a cumulative phase error; wherein the cumulative phase error includes a static deviation component and a dynamic fluctuation component.

[0013] The present invention also provides a permanent magnet synchronous motor rotor position estimation system based on high-frequency signal injection, including: A modulation module for amplitude modulating a preset high-frequency voltage carrier signal through an orthogonal rotation modulator to obtain an injection signal vector; An excitation module for voltage exciting the stator windings of a permanent magnet synchronous motor based on the injection signal vector to obtain a high-frequency current response signal; A separation module for separating harmonics of the high-frequency current response signal through a comb filter bank to obtain a rotor position modulation current component; A demodulation module for quadrature demodulating the rotor position modulation current component through a complex demodulator to obtain a position-dependent phase difference; A mapping module, configured to perform nonlinear mapping processing on the position-dependent phase difference value through an adaptive state observer to obtain a real-time rotor position angle value.

[0014] The present invention also provides a computer device, including a memory and a processor. A computer program is stored in the memory, and when the processor executes the computer program, the steps of the method described in any one of the above are implemented.

[0015] The present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method described in any one of the above are implemented.

[0016] A method for estimating the rotor position of a permanent magnet synchronous motor based on high-frequency signal injection provided by the present invention includes the following steps: amplitude-modulating a preset high-frequency voltage carrier signal through an orthogonal rotation modulator to obtain an injection signal vector; performing voltage excitation on the stator windings of the permanent magnet synchronous motor based on the injection signal vector to obtain a high-frequency current response signal; performing harmonic separation on the high-frequency current response signal through a comb filter bank to obtain a rotor position modulation current component; performing quadrature demodulation on the rotor position modulation current component through a complex demodulator to obtain a position-dependent phase difference value; performing nonlinear mapping processing on the position-dependent phase difference value through an adaptive state observer to obtain a real-time rotor position angle value, solving the key problem of how to effectively extract information related to the rotor position from the high-frequency current response signal, and achieving the technical effect that by adopting an adaptive state observer and nonlinear mapping processing, the influence of noise and nonlinear factors can be effectively suppressed, and high-precision rotor position angle estimation can be realized. Description of the Drawings

[0017] Figure 1 is a schematic diagram of the steps of a method for estimating the rotor position of a permanent magnet synchronous motor based on high-frequency signal injection in an embodiment of the present invention; Figure 2 is a structural block diagram of a device for estimating the rotor position of a permanent magnet synchronous motor based on high-frequency signal injection in an embodiment of the present invention; Figure 3 is a schematic structural block diagram of a computer device in an embodiment of the present invention.

[0018] The implementation, functional features, and advantages of the object of the present invention will be further described in conjunction with the embodiments with reference to the drawings. Detailed Embodiments

[0019] In order to make the object, technical solution, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0020] As Figure 1 shown Figure 1 is a schematic diagram of the steps of a rotor position estimation method for a permanent magnet synchronous motor based on high-frequency signal injection in an embodiment of the present invention; An embodiment of the present invention provides a rotor position estimation method for a permanent magnet synchronous motor based on high-frequency signal injection, including the following steps: Step S1, amplitude-modulate a preset high-frequency voltage carrier signal through an orthogonal rotation modulator to obtain an injection signal vector.

[0021] Specifically, amplitude-modulating a preset high-frequency voltage carrier signal through an orthogonal rotation modulator to obtain an injection signal vector. The core of this process lies in using the orthogonal rotation modulator to precisely control the amplitude characteristics of the high-frequency signal, thereby generating a signal vector suitable for injecting into the stator windings of the permanent magnet synchronous motor. Specifically, the preset high-frequency voltage carrier signal is a signal designed with certain frequency and amplitude characteristics. This signal itself does not contain rotor position information, but its high-frequency characteristics enable it to interact with the magnetic poles of the rotor inside the motor, thereby generating a response signal related to the rotor position. The role of the orthogonal rotation modulator is to amplitude-modulate this high-frequency voltage carrier signal, that is, by adjusting the amplitude size and phase relationship of the signal, to transform the original single high-frequency signal into an injection signal vector with specific directivity and intensity distribution. The direction and amplitude of this injection signal vector will directly affect the excitation effect of the subsequent stator windings, so its generation process must be very precise. For example, in the low-speed operation scenario of a permanent magnet synchronous motor, since traditional sensors cannot provide reliable rotor position information, the system needs to rely on the injected high-frequency signal to estimate the rotor position. Assume that the frequency of the high-frequency voltage carrier signal is 10 kHz and the amplitude is 5 V. The orthogonal rotation modulator modulates this signal and transforms it into an injection signal vector containing two orthogonal components, for example, respectively showing components of 3 V and 4 V in the d-q coordinate system. Such a vector signal, after being injected into the stator windings, can stimulate a high-frequency current response signal closely related to the rotor position, laying a foundation for subsequent harmonic separation and demodulation. Thus, it can be seen that the design and modulation process of the orthogonal rotation modulator is the key link of the entire method. It not only determines the quality of the injection signal but also directly affects the accuracy and reliability of the final rotor position estimation.

[0022] Step S2, perform voltage excitation on the stator windings of the permanent magnet synchronous motor based on the injection signal vector to obtain a high-frequency current response signal.

[0023] Specifically, an orthogonal rotation modulator is used to perform amplitude modulation on a preset high-frequency voltage carrier signal to obtain an injection signal vector. The core of this process lies in using the orthogonal rotation modulator to precisely control the amplitude characteristics of the high-frequency signal, thereby generating a signal vector suitable for injection into the stator windings of a permanent magnet synchronous motor. Specifically, the preset high-frequency voltage carrier signal is a signal designed with certain frequency and amplitude characteristics. This signal itself does not contain rotor position information, but its high-frequency characteristics enable it to interact with the magnetic poles of the rotor inside the motor, thereby generating a response signal related to the rotor position. The role of the orthogonal rotation modulator is to perform amplitude modulation on this high-frequency voltage carrier signal, that is, by adjusting the amplitude size and phase relationship of the signal, to convert the original single high-frequency signal into an injection signal vector with specific directivity and intensity distribution. The direction and amplitude of this injection signal vector will directly affect the excitation effect of the subsequent stator windings, so its generation process must be very precise. For example, in the low-speed operation scenario of a permanent magnet synchronous motor, since traditional sensors cannot provide reliable rotor position information, the system needs to rely on the injected high-frequency signal to estimate the rotor position. Assuming that the frequency of the high-frequency voltage carrier signal is 10 kHz and the amplitude is 5 V, the orthogonal rotation modulator modulates this signal and converts it into an injection signal vector containing two orthogonal components, for example, components of 3 V and 4 V in the d-q coordinate system respectively. Such a vector signal, after being injected into the stator windings, can stimulate a high-frequency current response signal closely related to the rotor position, laying a foundation for subsequent harmonic separation and demodulation. It can be seen that the design and modulation process of the orthogonal rotation modulator is the key link of the entire method. It not only determines the quality of the injection signal but also directly affects the accuracy and reliability of the final rotor position estimation.

[0024] Step S3: Use a comb filter bank to perform harmonic separation on the high-frequency current response signal to obtain the rotor position modulation current component.

[0025] Specifically, the harmonic separation of the high-frequency current response signal is performed by a comb filter bank to obtain the rotor position modulation current component. The key to this process lies in utilizing the frequency selection characteristics of the comb filter bank to extract the modulation current component directly related to the rotor position from the complex high-frequency current response signal. Specifically, the high-frequency current response signal is generated after the stator winding is excited by the injection signal vector, which not only contains information related to the rotor position but also is mixed with various harmonic components and noise interference. These harmonic components may originate from the nonlinear effects inside the motor, the parasitic parameters of the stator winding, and external electromagnetic interference, etc. They will obscure the characteristics of the rotor position modulation current component. Therefore, it is necessary to separate it through a comb filter bank. The design feature of the comb filter bank is that it can select signals of specific frequencies in a periodic manner in the frequency domain while suppressing other frequency components. This characteristic makes it very suitable for processing high-frequency current response signals containing various harmonic components. For example, in the application scenario of a permanent magnet synchronous motor running at low speed, assume that the high-frequency current response signal contains a fundamental component of 10 kHz and its harmonic components of integer multiples. The rotor position modulation current component usually appears as sideband frequencies near the fundamental wave. The comb filter bank can separate these sideband frequencies from the complex harmonic components by setting specific center frequencies and bandwidths. For example, if the frequency of the rotor position modulation current component is distributed between 9.8 kHz and 10.2 kHz, the comb filter bank can specifically extract the signals within this frequency range while suppressing other irrelevant harmonic components. In this way, the rotor position modulation current component after harmonic separation can more clearly reflect the actual position information of the rotor, providing a high-quality data basis for subsequent quadrature demodulation and phase difference calculation. It can be seen that the role of the comb filter bank is not only signal separation but also an important link to ensure that the entire estimation method can accurately capture the rotor position information. Its design and application directly determine the performance of the system.

[0026] Step S4: Perform quadrature demodulation on the rotor position modulation current component through a complex demodulator to obtain a position-dependent phase difference.

[0027] Specifically, the orthogonal demodulation of the rotor position modulation current component is performed by a complex demodulator to obtain a position-dependent phase difference. The core of this process lies in utilizing the orthogonal characteristics of the complex demodulator to extract the phase information directly related to the rotor position from the rotor position modulation current component. Specifically, after being separated by the comb filter bank, the rotor position modulation current component has removed most of the irrelevant harmonic interferences, but still contains complex amplitude and phase modulation information. The key features of the rotor position are hidden in this information, and the role of the complex demodulator is to explicitly express these features as a position-dependent phase difference through orthogonal demodulation. During the demodulation process, the complex demodulator generates two orthogonal signal components simultaneously, namely the in-phase component (I) and the quadrature component (Q), and these two components correspond to the real part and the imaginary part of the modulation signal respectively. By further processing the I and Q components, the change in their phase angles can be calculated, and this change in the phase angle actually reflects the change in the rotor position. For example, in the application scenario of a permanent magnet synchronous motor operating at low speed, assume that the rotor position modulation current component after harmonic separation contains sideband signals with a frequency of 10 kHz, and the amplitudes and phases of these signals fluctuate periodically with the change in the rotor position. The complex demodulator can decompose this signal into an in-phase component and a quadrature component through orthogonal demodulation. For example, if the I component is 3 mA and the Q component is 4 mA at a certain moment, the corresponding phase angle can be calculated as 53.1° through the arctangent function. When the rotor position changes, the values of the I and Q components also change accordingly, resulting in a change in the phase angle. This change in the phase angle is the position-dependent phase difference, which directly reflects the actual position information of the rotor. Thus, it can be seen that the role of the complex demodulator is not only to extract phase information, but more importantly, to convert the complex modulation signal into an easily processed phase difference, providing key data support for subsequent non-linear mapping and position estimation. The accuracy and real-time performance of this process directly affect the performance of the entire rotor position estimation method. Therefore, it is necessary to ensure that the design and implementation of the demodulator can meet the high-precision requirements of the system.

[0028] Step S5: Perform non-linear mapping processing on the position-dependent phase difference through an adaptive state observer to obtain a real-time rotor position angle value.

[0029] Specifically, the position-dependent phase difference is processed by an adaptive state observer through nonlinear mapping to obtain the real-time rotor position angle value. The core of this process lies in leveraging the dynamic characteristics of the adaptive state observer to convert the position-dependent phase difference into accurate rotor position angle information. Specifically, the position-dependent phase difference is obtained by quadrature demodulating the rotor position modulation current component. Although it can reflect the rotor position change, it is not directly the rotor angle value itself but contains a certain nonlinear mapping relationship. This nonlinear mapping relationship may be affected by internal parameters of the motor (such as inductance, resistance, etc.) and external operating conditions (such as load changes, temperature fluctuations, etc.). Therefore, an adaptive state observer is required to dynamically adjust and compensate for these effects. The design feature of the adaptive state observer is that it can continuously update its internal parameters according to the actual state of the system during operation, thereby ensuring the accuracy of the nonlinear mapping. For example, in the application scenario of a permanent magnet synchronous motor operating at low speed, assume that the position-dependent phase difference exhibits periodic fluctuations with the change of the rotor position. However, due to the existence of internal nonlinear effects and external disturbances in the motor, there may be a certain deviation between this fluctuation and the actual rotor angle. For example, at a certain moment, the position-dependent phase difference is 53.1°, but in fact, the corresponding rotor position angle may be 60°. At this time, the adaptive state observer will dynamically adjust its internal parameters according to the system model and real-time collected data to eliminate this deviation. For example, if the observer detects that the error between the current phase difference and the actual angle is large, it will reduce the error by adjusting the gain matrix or compensation factor, and finally output the accurate rotor position angle value. In this way, even when the motor parameters change or the external conditions are complex, the system can still accurately estimate the real-time rotor position angle value. Thus, it can be seen that the role of the adaptive state observer is not only to achieve nonlinear mapping, but more importantly, to ensure the high precision and robustness of the estimation result through dynamic adjustment, thereby providing reliable technical support for the sensorless control of permanent magnet synchronous motors.

[0030] In a specific embodiment, the process of obtaining the injection signal vector by amplitude modulating a preset high-frequency voltage carrier signal through an orthogonal rotation modulator includes: Performing phase separation processing on the preset high-frequency voltage carrier signal through an orthogonal rotation modulator to obtain orthogonal phase components; wherein, the orthogonal phase components include in-phase components and quadrature components; Performing coordinate rotation transformation on the orthogonal phase components based on a preset reference coordinate system transformation matrix to obtain a rotation modulation vector, and performing space vector synthesis on the rotation modulation vector to obtain a three-phase modulation signal; Performing nonlinear compensation processing on the three-phase modulation signal through a voltage vector decoupler to obtain the injection signal vector.

[0031] Specifically, an orthogonal rotation modulator is used to perform amplitude modulation on a preset high-frequency voltage carrier signal to obtain an injection signal vector. This process includes several key steps aimed at ensuring that the generated injection signal vector can accurately reflect the internal state of the motor and provide a high-quality data basis for subsequent processing. First, the orthogonal rotation modulator performs phase separation processing on the preset high-frequency voltage carrier signal to obtain orthogonal phase components; these orthogonal phase components include in-phase components and quadrature components, which together constitute a complete representation system. This phase separation processing is crucial because it allows us to decompose the original single-channel signal into two mutually independent but closely related signal channels, which respectively correspond to electromagnetic field components in different directions, thereby enabling us to more precisely control the injection direction and intensity of the signal. Based on a preset reference coordinate system transformation matrix, coordinate rotation transformation is performed on the orthogonal phase components to obtain a rotation modulation vector, and space vector synthesis is performed on the rotation modulation vector to obtain three-phase modulation signals. The work in this stage is to transform the previously obtained orthogonal phase components from the original coordinate system to another reference coordinate system that is more suitable for motor control. For example, in the control of a permanent magnet synchronous motor, the d-q coordinate system is usually used to simplify the analysis process, where the d-axis represents the magnetic field direction and the q-axis is perpendicular to the d-axis and orthogonal to the rotor magnetic field. By applying the preset reference coordinate system transformation matrix, the in-phase components and quadrature components can be transformed into the expression form in this coordinate system, that is, the rotation modulation vector. Subsequently, by performing space vector synthesis on these rotation modulation vectors, three-phase modulation signals suitable for the actual motor drive circuit can be generated. This process not only takes into account the complex electromagnetic relationships inside the motor but also improves the flexibility and adaptability of the control strategy through reasonable mathematical transformations. The last step is to perform non-linear compensation processing on the three-phase modulation signals through a voltage vector decoupler to obtain an injection signal vector. This step is crucial for eliminating non-linear effects caused by factors such as changes in internal parameters of the motor and external load disturbances. Specifically, in actual operation, the three-phase modulation signals may deviate from the ideal theoretical model due to various reasons. For example, parameters such as resistance and inductance change with temperature or frequency, or speed fluctuations are caused by mechanical loads. The role of the voltage vector decoupler is to address these issues by performing precise non-linear compensation processing on the three-phase modulation signals through a series of advanced algorithms and technical means. For example, in some application scenarios, to improve the response speed and stability of the system, feed-forward compensation technology can be used to pre-cancel possible disturbances; or a feedback control mechanism can be used to continuously monitor the system output and compare it with the expected value, and dynamically adjust the input signal to achieve the best performance. Finally, the signal after the above processing is the required injection signal vector, which can be directly used to excite the stator windings of the permanent magnet synchronous motor to generate the necessary high-frequency current response signal, providing a basis for further position estimation.For example, in a specific application scenario of a permanent magnet synchronous motor running at low speed, it is assumed that a high-frequency voltage carrier signal with a frequency of 10kHz and an initial phase angle of 30° needs to be generated as input. First, the signal is phase-separated by an orthogonal rotary modulator to obtain an in-phase component (I) and an orthogonal component (Q). Next, the two components are converted to the dq coordinate system by applying the dq coordinate system conversion matrix to form a rotating modulation vector. Then, these vectors are synthesized into a three-phase modulation signal by the space vector synthesis method. In this process, considering the special working conditions of the motor at low speed and the influence of the external environment, a voltage vector decoupler is used to perform nonlinear compensation processing on the generated three-phase modulation signal to ensure that the required injection signal vector can be accurately generated even under complex conditions. In this way, the generated injection signal vector can not only effectively stimulate the electromagnetic response inside the motor, but also lay a solid foundation for accurately extracting the rotor position information in the subsequent steps. The entire process embodies a series of precise designs and technical implementations from signal generation to processing to application, fully demonstrating the advancement and complexity of modern motor control technology.

[0032] In a specific embodiment, the step of performing voltage excitation on the stator winding of the permanent magnet synchronous motor based on the injected signal vector to obtain a high-frequency current response signal includes: The injection signal vector is allocated with a switching timing by a space vector pulse width modulator to obtain a multi-phase excitation voltage waveform, and voltage excitation is applied to the stator winding of the permanent magnet synchronous motor based on the multi-phase excitation voltage waveform to obtain the electromagnetic energy conversion characteristics of the stator winding; Based on the electromagnetic energy conversion characteristics of the stator winding, the magnetic field distribution of the stator winding is modeled to obtain rotor salient pole effect parameters, and the rotor salient pole effect parameters are decoupled from self-inductance and mutual inductance to obtain a predicted amount of stator winding current excitation response; The current excitation response prediction quantity of the stator winding is sampled synchronously through multi-channel through the phase current sampling circuit to obtain a high-frequency current response signal; wherein the high-frequency current response signal includes a stator winding fundamental frequency response component and a stator winding high-frequency modulation component.

[0033] Specifically, the process of applying voltage excitation to the stator windings of a permanent magnet synchronous motor based on the injected signal vector to obtain a high-frequency current response signal is a complex and precise technical process. It not only involves a series of steps from signal generation to actual application but also needs to consider the complex electromagnetic interactions within the motor. First, the space vector pulse width modulator (SVPWM) is used to allocate the switching timings for the injected signal vector. This is to convert the abstract injected signal vector into a specific and executable multi-phase excitation voltage waveform. The space vector pulse width modulation technology is widely used in motor drive systems due to its high efficiency and flexibility. In this process, SVPWM calculates the corresponding switching timings according to the preset injected signal vector to ensure that each power device is turned on or off at the correct time point, thereby generating the required three-phase or multi-phase voltage waveform. For example, in a specific case, if the injected signal vector points to a certain specific direction in the d-q coordinate system, SVPWM will adjust the action sequence and time ratio of each switching element in the inverter according to this direction and magnitude, so that the finally output voltage waveform can accurately reflect the requirements of the original injected signal vector. Next, based on the multi-phase excitation voltage waveform, voltage excitation is applied to the stator windings of the permanent magnet synchronous motor to obtain the electromagnetic energy conversion characteristics of the stator windings. When these carefully designed voltage waveforms are applied to the motor stator windings, they will cause changes in the electromagnetic field in the windings and further lead to current flow. Due to the complex electromagnetic relationships within the motor, including the influence of parameters such as resistance and inductance, this electromagnetic energy conversion process is not a simple linear transformation. At this stage, the physical and electrical characteristics of the motor begin to emerge, especially the electromagnetic interaction between the rotor and the stator becomes a key factor. By analyzing the electromagnetic energy conversion characteristics in this process, the working state of the motor and its response behavior under different operating conditions can be deeply understood. Subsequently, based on the electromagnetic energy conversion characteristics of the stator windings, a magnetic field distribution model of the stator windings is established to obtain the rotor salient pole effect parameters, and self-inductance and mutual-inductance decoupling processing is performed on the rotor salient pole effect parameters to obtain the predicted value of the stator winding current excitation response. The rotor salient pole effect mentioned here refers to the magnetic field asymmetry phenomenon caused by reasons such as irregular rotor shape or uneven material distribution. This asymmetry will affect the operating performance of the motor. Through a detailed analysis of the electromagnetic energy conversion characteristics of the stator windings, a model reflecting the internal magnetic field distribution of the motor can be constructed, and the parameters related to the rotor salient pole effect can be extracted from it. Then, mathematical methods are used to further process these parameters, especially to decouple the self-inductance and mutual-inductance, in order to accurately predict the current response of the stator windings under given excitation conditions. For example, in some cases, due to the rotor salient pole effect, even if the same voltage excitation is input, the current response in the stator windings will vary depending on the rotor position. Through decoupling processing, the self-inductance and mutual-inductance components can be more clearly separated, providing reliable data support for subsequent current response prediction.Finally, the current excitation response prediction of the stator winding is synchronously sampled in multiple channels through a phase current sampling circuit to obtain a high-frequency current response signal; wherein, the high-frequency current response signal includes a fundamental frequency response component of the stator winding and a high-frequency modulation component of the stator winding. The design of the phase current sampling circuit aims to capture the transient current information generated during the operation of the motor, which is crucial for accurately estimating the rotor position. At this stage, the synchronous sampling of multiple channels ensures the integrity and accuracy of data acquisition, avoiding errors caused by asynchronous sampling. For example, in actual operation, assuming the motor operating frequency is 50 Hz and the injected high-frequency voltage carrier signal frequency is 10 kHz, then through the phase current sampling circuit, not only can the fundamental frequency response component reflecting the basic operating state of the motor be obtained, but also the modulation component caused by the high-frequency voltage excitation can be captured. These high-frequency modulation components carry rich information about the rotor position, and after appropriate signal processing, they can be used for subsequent harmonic separation and position estimation steps. In summary, the entire process starts from the generation of the injected signal vector, goes through a series of precise control and measurement means, and finally obtains a high-frequency current response signal containing rich information, providing a solid foundation for the sensorless control strategy. This series of operations demonstrates the high integration and intelligence of modern motor control technology, and at the same time emphasizes the technical challenges that need to be overcome in achieving high-precision position estimation.

[0034] In a specific embodiment, the harmonic separation of the high-frequency current response signal by a comb filter bank to obtain a rotor position modulation current component includes: Synchronously demodulating the high-frequency current response signal through a lock-in amplifier to obtain a high-frequency carrier modulation signal, and performing a Hilbert transform on the high-frequency carrier modulation signal to obtain a high-frequency current analytic signal; Based on the high-frequency current analytic signal, performing a spectrum zoom analysis on the high-frequency current response signal to obtain a high-frequency current harmonic distribution matrix, and performing a singular value decomposition on the high-frequency current harmonic distribution matrix to obtain a high-frequency current principal component eigenvector; Performing harmonic separation on the high-frequency current principal component eigenvector through a comb filter bank to obtain a harmonic separation current vector, and performing a rotational coordinate transformation on the harmonic separation current vector to obtain a rotor position modulation current vector; Extracting the amplitude and phase of the rotor position modulation current vector to obtain a rotor position modulation current component.

[0035] Specifically, the process of separating harmonics from the high-frequency current response signal through a comb filter bank to obtain the rotor position modulation current component is a complex and delicate technical process. It not only involves the acquisition of the original signal to the extraction of the final target information but also requires the application of a series of advanced signal processing techniques. First, the high-frequency current response signal is synchronously demodulated by a lock-in amplifier to obtain a high-frequency carrier modulation signal, and the Hilbert transform is performed on the high-frequency carrier modulation signal to obtain a high-frequency current analytic signal. The function of the lock-in amplifier is to accurately track the frequency and phase changes of the input signal and effectively extract useful information even in the presence of a large amount of noise. In this process, the lock-in amplifier uses a reference signal (usually synchronized with the injected high-frequency voltage carrier signal) to achieve synchronous demodulation of the high-frequency current response signal. For example, in a permanent magnet synchronous motor control scenario, assuming that the high-frequency current response signal contains complex frequency components, including the fundamental frequency, harmonics, and sideband frequencies caused by the rotor salient pole effect, the lock-in amplifier can effectively extract the high-frequency carrier modulation signal related to the rotor position. Subsequently, through the Hilbert transform, these real signals are converted into analytic signals in complex form. This step greatly simplifies the subsequent signal analysis process because it allows direct operation on the amplitude and phase information of the signal. Based on the high-frequency current analytic signal, a spectral zoom analysis is performed on the high-frequency current response signal to obtain a high-frequency current harmonic distribution matrix, and the singular value decomposition is performed on the high-frequency current harmonic distribution matrix to obtain a high-frequency current principal component eigenvector. The work in this stage mainly focuses on the in-depth analysis of the high-frequency current response signal. By performing spectral zoom analysis on the analytic signal, the various frequency components in the signal and their corresponding intensity distribution can be detailedly depicted, forming the so-called high-frequency current harmonic distribution matrix. This matrix essentially describes the characteristics of the signal in the frequency domain and provides a basis for further data processing. Next, the singular value decomposition (SVD) technique is applied to process this matrix. SVD is a powerful mathematical tool that can decompose a complex matrix into several simpler matrices, facilitating the extraction of key information. In this case, through SVD, the principal components that can best represent the signal characteristics, namely the high-frequency current principal component eigenvectors, can be identified from numerous frequency components. These eigenvectors not only concentrate the main information of the original signal but also provide a basis for subsequent harmonic separation. Then, the harmonic separation is performed on the high-frequency current principal component eigenvector through a comb filter bank to obtain a harmonic separation current vector, and the rotational coordinate transformation is performed on the harmonic separation current vector to obtain a rotor position modulation current vector. The design purpose of the comb filter bank is to select specific frequency components from complex signals. Considering that the high-frequency current principal component eigenvector has concentratedly reflected the main characteristics of the signal, the comb filter bank can further separate the harmonic components directly related to the rotor position.For example, in the above application scenario, if the goal is to extract rotor position information, then the sideband frequency components caused by the rotor salient pole effect need to be concerned about. The comb filter bank can accurately select these frequency components to form a harmonic separation current vector. Then, through a rotational coordinate transformation, these vectors are transformed into a space that is more suitable for representing rotor position information, obtaining a rotor position modulation current vector. This coordinate transformation helps to eliminate unnecessary interference factors, making the signal more intuitively reflect the position change of the rotor. Finally, the amplitude and phase of the rotor position modulation current vector are extracted to obtain the rotor position modulation current components. In this step, the key is to accurately extract the amplitude and phase information reflecting the rotor position from the processed vector. Through appropriate mathematical methods, the changes in its amplitude and phase can be directly calculated from the rotor position modulation current vector, and these changes directly correspond to the actual position of the rotor. For example, at a certain moment, if the amplitude increases and the phase shifts by a certain angle, it indicates that the position of the rotor relative to the stator has changed accordingly. This precise amplitude and phase information is crucial for subsequent rotor position estimation. It not only improves the accuracy of the estimation but also guarantees the effective implementation of the sensorless control strategy. The entire process demonstrates the complete chain from the original high-frequency current response signal to the final rotor position modulation current components, emphasizing the close connection between each step and the importance of technical details. Through such a multi-level and multi-step processing method, it is ensured that even in a complex operating environment, the rotor position information can be accurately captured.

[0036] In a specific embodiment, the orthogonal demodulation of the rotor position modulation current component by the complex demodulator to obtain a position-dependent phase difference includes: Synchronously sampling the rotor position modulation current component through an orthogonal carrier generator to obtain a pair of orthogonal reference signals, and performing complex transformation processing on the pair of orthogonal reference signals to obtain a phase reference vector; Based on a mixer, performing complex multiplication operation on the rotor position modulation current component and the phase reference vector to obtain a spectrum-shifted component, and performing low-pass filtering processing on the spectrum-shifted component through a complex demodulator to obtain a baseband demodulation signal; Performing a four-quadrant arctangent operation on the baseband demodulation signal through a vector phase detector to obtain an initial phase angle measurement, and performing periodic distortion compensation on the initial phase angle measurement to obtain a corrected phase value; Based on a harmonic canceller, performing high-order harmonic suppression on the corrected phase value to obtain a continuous phase trajectory, and performing digital integration processing on the continuous phase trajectory to obtain an accumulated phase value; The cumulative phase value is subjected to electrical angle mapping by a spatial vector analyzer to obtain a position-dependent phase difference; wherein the position-dependent phase difference includes a magnetic field spatial distribution term and a salient pole effect coupling term.

[0037] Specifically, the process of quadrature demodulating the rotor position modulation current component through a plurality of demodulators to obtain the position-dependent phase difference is a complex and precise signal processing flow, aiming to extract accurate information related to the motor rotor position from the original current signal. First, the rotor position modulation current component is synchronously sampled through an orthogonal carrier generator to obtain a pair of orthogonal reference signals, and the pair of orthogonal reference signals is subjected to complex transformation processing to obtain a phase reference vector. The core of this process lies in ensuring that the generated orthogonal reference signals can accurately reflect the phase characteristics of the original signal. For example, in an application scenario of a permanent magnet synchronous motor, if the goal is to estimate the rotor position, reference signals that change synchronously with the rotor magnetic field need to be generated. The in-phase (I) and quadrature (Q) reference signals generated by the orthogonal carrier generator not only provide the amplitude information of the signal but also contain its phase information, which is crucial for subsequent phase detection. Then, these orthogonal reference signals are converted into a phase reference vector through complex transformation processing, which lays the foundation for further complex multiplication operations. Next, based on a mixer, complex multiplication operations are performed on the rotor position modulation current component and the phase reference vector to obtain a spectrum-shifted component, and the spectrum-shifted component is subjected to low-pass filtering processing through a complex demodulator to obtain a baseband demodulated signal. In this process, the role of the mixer is to "shift" the high-frequency rotor position modulation current component to the baseband, enabling the originally complex high-frequency signal to be processed by a simple low-pass filter. For example, assume that the original signal contains multiple frequency components, some of which are sideband frequencies caused by rotor position changes. Through complex multiplication with the phase reference vector, these sideband frequencies can be effectively shifted to the vicinity of the baseband for subsequent processing. Then, the unnecessary high-frequency components are removed through the low-pass filter in the complex demodulator, and only the baseband demodulated signal directly related to the rotor position is retained. This step significantly simplifies the difficulty of signal analysis and improves the processing efficiency. Subsequently, through a vector phase detector, a four-quadrant arctangent operation is performed on the baseband demodulated signal to obtain an initial phase angle measurement, and the initial phase angle measurement is compensated for periodic distortion to obtain a corrected phase value. The key to this step lies in accurately calculating the phase information of the baseband demodulated signal. The four-quadrant arctangent operation is one of the standard methods for determining the signal phase, which can process signals in all four quadrants to ensure the comprehensiveness and accuracy of phase calculation. However, due to the complexity of the motor operating environment, the actual phase information obtained may be affected by periodic distortion, such as small offsets caused by temperature changes or mechanical stress. Therefore, it is necessary to further compensate the initial phase angle measurement to eliminate these distortions and ensure the high precision of the phase information. For example, in some cases, since the resistance of the stator winding increases with temperature rise, it may cause phase measurement deviation, and this deviation can be adjusted and corrected in real time through an appropriate algorithm.Subsequently, based on the harmonic canceler, high-order harmonic suppression is performed on the corrected phase value to obtain a continuous phase trajectory, and digital integration processing is performed on the continuous phase trajectory to obtain an accumulated phase value. In this step, the key lies in eliminating unnecessary harmonic components in the phase data to ensure the smoothness and continuity of the phase trajectory. For example, due to the existence of internal nonlinear effects or external electromagnetic interference in the motor, the corrected phase value may contain high-order harmonics, which will interfere with the final rotor position estimation. Through the harmonic canceler, these harmonics can be effectively identified and suppressed, thereby obtaining a more pure and smooth phase trajectory. Then, through digital integration processing, this continuous phase trajectory is converted into an accumulated phase value. This step not only considers the instantaneous change of the phase but also comprehensively considers the phase accumulation situation within the entire time period, providing a basis for subsequent space vector analysis. Finally, through the space vector analyzer, electrical angle mapping is performed on the accumulated phase value to obtain a position-dependent phase difference; wherein, the position-dependent phase difference includes a magnetic field spatial distribution term and a salient pole effect coupling term. The goal of this step is to extract characteristic information directly related to the motor rotor position from the accumulated phase value. The space vector analyzer uses the spatial distribution characteristics of the internal magnetic field of the motor and the influence of the rotor salient pole effect on the phase to convert the accumulated phase value into an electrical angle with physical significance. For example, in a specific example, if the accumulated phase value shows fluctuations within a certain range, it can be mapped by the space vector analyzer into the actual rotation angle of the motor rotor. In this process, not only the basic distribution pattern of the magnetic field needs to be considered, but also the salient pole effect caused by irregular rotor shape or uneven material distribution needs to be considered. These factors act together to form the final position-dependent phase difference, which is not only an important input for the motor control strategy but also provides reliable data support for realizing sensorless position estimation. The entire process demonstrates the complete chain from the original current signal to the extraction of the final position information, reflecting the high integration and intelligence of modern motor control technology.

[0038] In a specific embodiment, the non-linear mapping process of the position-dependent phase difference by the adaptive state observer to obtain the real-time rotor position angle value includes: Based on the elliptic curve mapper, angle non-linear correction is performed on the position-dependent phase difference to obtain an initial position estimate, and phase unwrapping processing is performed on the initial position estimate to obtain a continuous angle sequence; Through the error compensation integrator, time-domain integration operation is performed on the continuous angle sequence to obtain an accumulated phase error, and through the adaptive state observer, polynomial curve fitting is performed on the accumulated phase error to obtain a magnetic field disturbance component; Nonlinear characteristic extraction is performed on the magnetic field disturbance component based on a flux saturation compensator to obtain a dynamic saturation factor, and weighted fusion is performed on the dynamic saturation factor and the continuous angle sequence to obtain a corrected angle estimation value; Closed-loop tracking control is performed on the corrected angle estimation value through a phase-locked loop to obtain a real-time rotor position angle value, where the real-time rotor position angle value includes an electrical angle component and a mechanical angle component.

[0039] Specifically, the process of non-linearly mapping the position-dependent phase difference through an adaptive state observer to obtain the real-time rotor position angle value is a highly complex and precise signal processing flow, aiming to extract the angle value that accurately reflects the actual position of the motor rotor from the original phase information. First, based on an elliptic curve mapper, non-linear angle correction is performed on the position-dependent phase difference to obtain an initial position estimate, and phase unwrapping processing is performed on the initial position estimate to obtain a continuous angle sequence. In this process, the role of the elliptic curve mapper is to compensate for the phase error caused by the non-linear characteristics inside the motor, such as the phase shift caused by magnetic saturation effects or temperature changes. By applying a specific mathematical model (such as an elliptic curve) to the position-dependent phase difference, these non-linear errors can be effectively corrected to ensure that the initial position estimate reflects the true rotor position as accurately as possible. Then, in order to eliminate the discontinuity problem caused by phase wrapping, phase unwrapping processing needs to be performed on the initial position estimate to smooth the originally possible jump points and generate a continuous angle sequence. This step is crucial for subsequent integration operations because it ensures the continuity and consistency of the data. Next, through an error compensation integrator, time-domain integration operation is performed on the continuous angle sequence to obtain the cumulative phase error, and through an adaptive state observer, polynomial curve fitting is performed on the cumulative phase error to obtain the magnetic field perturbation component. In this step, the main function of the error compensation integrator is to accumulate and correct the tiny errors in the continuous angle sequence. During the operation of the motor, it will inevitably be affected by various external factors, such as load fluctuations, temperature changes, etc. These factors will cause slight deviations in angle measurement. Through time-domain integration operation, these cumulative errors can be quantified and provide a basis for subsequent compensation. Subsequently, the adaptive state observer is used to analyze the cumulative phase error, and the polynomial curve fitting method is adopted to extract the component related to the magnetic field perturbation. This fitting method can effectively identify and separate the error components caused by the magnetic field perturbation and provide a basis for further compensation processing. Then, based on a flux saturation compensator, non-linear characteristic extraction is performed on the magnetic field perturbation component to obtain a dynamic saturation factor, and weighted fusion is performed on the dynamic saturation factor and the continuous angle sequence to obtain a corrected angle estimate value. In this step, the role of the flux saturation compensator is to compensate for the non-linear behavior of the motor flux linkage characteristics. Under high-load operating conditions of the motor, the flux linkage may saturate, resulting in its electrical characteristics deviating from the ideal state. By analyzing the magnetic field perturbation component, the dynamic saturation factor related to flux saturation can be extracted. Then, this factor is combined with the previously obtained continuous angle sequence, and through an appropriate weighted fusion algorithm, the angle estimate value is adjusted to make it closer to the true value. This method not only takes into account the dynamic changes during the operation of the motor but also improves the accuracy of angle estimation through reasonable weight allocation.Finally, a closed-loop tracking control is performed on the corrected angle estimation value through a phase-locked loop to obtain a real-time rotor position angle value, where the real-time rotor position angle value includes an electrical angle component and a mechanical angle component. The phase-locked loop is a classic feedback control system that can maintain accurate tracking of the target signal in a dynamic environment. In this application scenario, the phase-locked loop is used to monitor and adjust the corrected angle estimation value in real time to ensure that it always accurately reflects the actual position of the motor rotor. The electrical angle component reflects the rotation angle of the electromagnetic field inside the motor, while the mechanical angle component directly corresponds to the physical position of the rotor relative to the stator. By combining this information from both aspects, the phase-locked loop can provide a comprehensive and accurate rotor position angle value. For example, in a specific application scenario of a permanent magnet synchronous motor operating at low speed, assuming the motor is under heavy load conditions, due to the presence of flux saturation and external interference, simple angle estimation methods are difficult to meet the accuracy requirements. Through the above series of complex signal processing steps, including elliptic curve mapping, phase unwrapping, error compensation integration, adaptive state observation, and flux saturation compensation, etc., a high-precision real-time rotor position angle value can finally be obtained, which not only improves the performance of the motor control system but also provides reliable technical support for realizing sensorless position detection. The entire process demonstrates the complete technical chain from the original phase information to the extraction of the final rotor position angle value, reflecting the high level of integration and intelligence of modern motor control technology.

[0040] In a specific embodiment, the performing a time-domain integration operation on the continuous angle sequence through an error compensation integrator to obtain a cumulative phase error includes: Performing a sliding window segmentation process on the continuous angle sequence through a phase differentiator to obtain an angle increment sequence, and performing a zero-drift compensation operation on the angle increment sequence to obtain a compensated angle sequence; Performing a frequency-domain decomposition process on the compensated angle sequence based on a harmonic analyzer to obtain an angle harmonic component matrix, and performing a phase reconstruction operation on the angle harmonic component matrix to obtain a phase error vector; Performing a dynamic weight assignment on the phase error vector through an adaptive integrator to obtain a weighted error sequence, and performing a non-linear integration operation on the weighted error sequence through an error compensation integrator to obtain a cumulative error curve; Performing a time-varying feature extraction on the cumulative error curve based on an error compensation network to obtain an error compensation factor, and performing an adaptive correction process on the error compensation factor to obtain a cumulative phase error; where the cumulative phase error includes a static deviation component and a dynamic fluctuation component.

[0041] Specifically, the process of performing time-domain integration on the continuous angle sequence through the error compensation integrator to obtain the cumulative phase error is a meticulous and complex signal processing procedure, aiming to accurately extract the cumulative phase error caused by various error factors from the continuous angle data. First, the continuous angle sequence is processed by a sliding window segmentation through a phase differentiator to obtain an angle increment sequence, and a zero-drift compensation operation is performed on the angle increment sequence to obtain a compensated angle sequence. In this process, the phase differentiator divides the continuous angle sequence into multiple small segments, each segment representing the angle change amount within a specific time period, that is, the angle increment. This sliding window method helps to capture the subtle differences in angle changes. Then, by performing a zero-drift compensation operation on these angle increments, the offset error caused by the long-term operation of the system or environmental temperature changes can be eliminated, ensuring that each angle increment can accurately reflect the actual angle change and generating a compensated angle sequence. Next, based on a harmonic analyzer, frequency-domain decomposition processing is performed on the compensated angle sequence to obtain an angle harmonic component matrix, and a phase reconstruction operation is performed on the angle harmonic component matrix to obtain a phase error vector. In this step, the role of the harmonic analyzer is to convert the compensated angle sequence in the time domain into a frequency-domain representation to identify and separate various frequency components therein. For example, during the operation of a motor, due to factors such as electromagnetic interference and mechanical vibration, the angle sequence may contain multiple harmonic components with different frequencies. Through frequency-domain decomposition, these harmonic components and their corresponding intensities and phase information can be clearly identified, forming an angle harmonic component matrix. Subsequently, through the phase reconstruction operation, a vector directly related to the phase error can be extracted from this matrix. This step not only simplifies the data structure for subsequent processing but also improves the accuracy of phase error calculation. Then, through an adaptive integrator, dynamic weight allocation is performed on the phase error vector to obtain a weighted error sequence, and a nonlinear integration operation is performed on the weighted error sequence through the error compensation integrator to obtain a cumulative error curve. In this step, the adaptive integrator assigns different weights to each phase error according to its importance to highlight the influence of the main error sources. This dynamic weight allocation strategy can effectively improve the pertinence and efficiency of error compensation. Then, by using the error compensation integrator to perform a nonlinear integration operation on the weighted error sequence, the small errors in the short term can be accumulated to form a cumulative error curve that describes the error change trend over the entire time period. This curve not only reflects the law of error change over time but also provides basic data for subsequent error compensation. Finally, based on an error compensation network, time-varying feature extraction is performed on the cumulative error curve to obtain an error compensation factor, and an adaptive correction process is performed on the error compensation factor to obtain the cumulative phase error; wherein, the cumulative phase error includes a static deviation component and a dynamic fluctuation component.In this step, the task of the error compensation network is to extract the time-varying features from the cumulative error curve, that is, the error change pattern that occurs over time. These features may include periodic fluctuations, long-term trends, etc. By analyzing these features, appropriate error compensation factors can be determined to correct the errors in the original angle measurement values. In addition, in order to adapt to the changing working conditions, the error compensation factors need to be adaptively corrected to ensure that they can always effectively handle the latest error situations. The finally obtained cumulative phase error not only includes the static deviation component caused by the inherent deviation of the system, but also includes the dynamic fluctuation component caused by external interference or internal non-linear effects. This series of steps demonstrates the complete technical path of how to start from a continuous angle sequence, through multiple transformations and processes, and finally accurately calculate the cumulative phase error, providing a solid foundation for achieving high-precision motor control. In the practical application of permanent magnet synchronous motors, this method can significantly improve the accuracy of position estimation, especially in complex and changing operating environments, and still maintain high reliability.

[0042] The rotor position estimation method of the permanent magnet synchronous motor based on high-frequency signal injection in the embodiments of the present invention is described above. Next, the rotor position estimation system of the permanent magnet synchronous motor based on high-frequency signal injection in the embodiments of the present invention will be described. Please refer to Figure 2 , an embodiment of the rotor position estimation system of the permanent magnet synchronous motor based on high-frequency signal injection in the embodiments of the present invention includes: A modulation module 21, configured to perform amplitude modulation on a preset high-frequency voltage carrier signal through an orthogonal rotation modulator to obtain an injection signal vector; An excitation module 22, configured to perform voltage excitation on the stator winding of the permanent magnet synchronous motor based on the injection signal vector to obtain a high-frequency current response signal; A separation module 23, configured to perform harmonic separation on the high-frequency current response signal through a comb filter bank to obtain a rotor position modulation current component; A demodulation module 24, configured to perform quadrature demodulation on the rotor position modulation current component through a complex demodulator to obtain a position-dependent phase difference; A mapping module 25, configured to perform non-linear mapping processing on the position-dependent phase difference through an adaptive state observer to obtain a real-time rotor position angle value.

[0043] In this embodiment, for the specific implementation of each unit in the above system embodiment, please refer to that described in the above method embodiment, and details will not be repeated here.

[0044] Refer to Figure 3 , an embodiment of the present invention also provides a computer device, and its internal structure can be as Figure 3As shown in the figure. The computer device includes a processor, a memory, a display screen, an input device, a network interface, and a database connected through a system bus. Among them, the processor of the computer design is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the corresponding data in this embodiment. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the above method is implemented.

[0045] Those skilled in the art can understand that Figure 3 the structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied.

[0046] An embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the above method is implemented. It can be understood that the computer-readable storage medium in this embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0047] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium provided by the present invention and used in the embodiments can include non-volatile and / or volatile memories. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or an external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM, etc.

[0048] It should be noted that in this text, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, such that a process, apparatus, article or method comprising a series of elements not only includes those elements but also includes other elements not expressly listed, or further includes elements inherent to such process, apparatus, article or method. Without further limitation, an element defined by the statement "comprising one..." does not exclude the presence of additional identical elements in the process, apparatus, article or method comprising such element.

[0049] The above are only the preferred embodiments of the present invention, and do not limit the patent scope of the present invention accordingly. Any equivalent structure or equivalent process transformation made by using the content of the specification and drawings of the present invention, or directly or indirectly applied in other related technical fields, shall equally be included in the patent protection scope of the present invention.

Claims

1. A method for estimating the rotor position of a permanent magnet synchronous motor based on high-frequency signal injection, characterized in that: The following steps are involved: A preset high-frequency voltage carrier signal is amplitude modulated by an orthogonal rotary modulator to obtain an injection signal vector; Based on the injection signal vector, voltage excitation is performed on the stator winding of the permanent magnet synchronous motor to obtain a high-frequency current response signal; Performing harmonic separation on the high-frequency current response signal through a comb filter group to obtain a rotor position modulated current component; Performing quadrature demodulation on the rotor position modulated current component by a complex demodulator to obtain a position-dependent phase difference value; The position-dependent phase difference value is subjected to nonlinear mapping processing by an adaptive state observer to obtain a real-time rotor position angle value.

2. The method for estimating the rotor position of a permanent magnet synchronous motor based on high frequency signal injection according to claim 1, characterized in that: The method of performing amplitude modulation on a preset high-frequency voltage carrier signal by an orthogonal rotary modulator to obtain an injection signal vector comprises: The preset high-frequency voltage carrier signal is subjected to phase separation processing by an orthogonal rotary modulator to obtain an orthogonal phase component; wherein the orthogonal phase component includes an in-phase component and an orthogonal component; Performing coordinate rotation transformation on the orthogonal phase component based on a preset reference coordinate system conversion matrix to obtain a rotation modulation vector, and performing space vector synthesis on the rotation modulation vector to obtain a three-phase modulation signal; The three-phase modulated signal is subjected to nonlinear compensation processing by a voltage vector decoupler to obtain an injected signal vector.

3. The method for estimating the rotor position of a permanent magnet synchronous motor based on high frequency signal injection according to claim 1, characterized in that: The step of performing voltage excitation on the stator winding of the permanent magnet synchronous motor based on the injected signal vector to obtain a high-frequency current response signal includes: The injection signal vector is allocated with a switching timing by a space vector pulse width modulator to obtain a multi-phase excitation voltage waveform, and voltage excitation is applied to the stator winding of the permanent magnet synchronous motor based on the multi-phase excitation voltage waveform to obtain the electromagnetic energy conversion characteristics of the stator winding; Based on the electromagnetic energy conversion characteristics of the stator winding, the magnetic field distribution of the stator winding is modeled to obtain rotor salient pole effect parameters, and the rotor salient pole effect parameters are decoupled from self-inductance and mutual inductance to obtain a predicted amount of current excitation response of the stator winding; The current excitation response prediction quantity of the stator winding is sampled synchronously through multi-channel through the phase current sampling circuit to obtain a high-frequency current response signal; wherein the high-frequency current response signal includes a stator winding fundamental frequency response component and a stator winding high-frequency modulation component.

4. The method for estimating the rotor position of a permanent magnet synchronous motor based on high frequency signal injection according to claim 1, characterized in that: The step of performing harmonic separation on the high frequency current response signal by using a comb filter group to obtain a rotor position modulated current component comprises: Synchronously demodulating the high-frequency current response signal through a phase-locked amplifier to obtain a high-frequency carrier modulation signal, and performing Hilbert transform on the high-frequency carrier modulation signal to obtain a high-frequency current analysis signal; Based on the high-frequency current analytical signal, a spectrum refinement analysis is performed on the high-frequency current response signal to obtain a high-frequency current harmonic distribution matrix, and a singular value decomposition is performed on the high-frequency current harmonic distribution matrix to obtain a high-frequency current principal component eigenvector; Performing harmonic separation on the high-frequency current principal component eigenvector through a comb filter group to obtain a harmonic separation current vector, and performing a rotation coordinate transformation on the harmonic separation current vector to obtain a rotor position modulation current vector; The amplitude and phase of the rotor position modulated current vector are extracted to obtain a rotor position modulated current component.

5. The method for estimating the rotor position of a permanent magnet synchronous motor based on high frequency signal injection according to claim 1, characterized in that: The method of performing quadrature demodulation on the rotor position modulated current component by a complex demodulator to obtain a position-dependent phase difference value includes: Synchronously sampling the rotor position modulated current component through an orthogonal carrier generator to obtain an orthogonal reference signal pair, and performing complex transformation processing on the orthogonal reference signal pair to obtain a phase reference vector; Performing a complex multiplication operation on the rotor position modulated current component and the phase reference vector based on a mixer to obtain a spectrum shift component, and performing a low-pass filtering process on the spectrum shift component through a complex demodulator to obtain a baseband demodulated signal; Performing a four-quadrant inverse tangent operation on the baseband demodulated signal through a vector phase detector to obtain an initial phase angle, and performing periodic distortion compensation on the initial phase angle to obtain a corrected phase value; Performing high-order harmonic suppression on the modified phase value based on a harmonic canceller to obtain a continuous phase trajectory, and performing digital integration processing on the continuous phase trajectory to obtain a cumulative phase value; The accumulated phase value is subjected to electrical angle mapping by a space vector analyzer to obtain a position-dependent phase difference value; wherein the position-dependent phase difference value includes a magnetic field spatial distribution term and a salient pole effect coupling term.

6. The method for estimating the rotor position of a permanent magnet synchronous motor based on high frequency signal injection according to claim 1, characterized in that: The step of performing nonlinear mapping processing on the position-dependent phase difference value by an adaptive state observer to obtain a real-time rotor position angle value includes: Based on the elliptic curve mapper, the position-dependent phase difference value is subjected to angle nonlinear correction to obtain an initial position estimate, and the initial position estimate is subjected to phase unwrapping processing to obtain a continuous angle sequence; Performing a time domain integration operation on the continuous angle sequence through an error compensation integrator to obtain a cumulative phase error, and performing polynomial curve fitting on the cumulative phase error through an adaptive state observer to obtain a magnetic field disturbance component; Extracting nonlinear characteristics of the magnetic field disturbance component based on a flux saturation compensator to obtain a dynamic saturation factor, and weightedly fusing the dynamic saturation factor with the continuous angle sequence to obtain a corrected angle estimate; The corrected angle estimation value is subjected to closed-loop tracking control through a phase-locked loop to obtain a real-time rotor position angle value, wherein the real-time rotor position angle value includes an electrical angle component and a mechanical angle component.

7. The method for estimating the rotor position of a permanent magnet synchronous motor based on high frequency signal injection according to claim 6, characterized in that: The step of performing a time domain integration operation on the continuous angle sequence by an error compensation integrator to obtain a cumulative phase error comprises: Performing sliding window segmentation processing on the continuous angle sequence through a phase differentiator to obtain an angle increment sequence, and performing zero drift compensation operation on the angle increment sequence to obtain a compensated angle sequence; Performing frequency domain decomposition processing on the compensation angle sequence based on a harmonic analyzer to obtain an angle harmonic component matrix, and performing phase reconstruction operation on the angle harmonic component matrix to obtain a phase error vector; Dynamically weighting the phase error vector using an adaptive integrator to obtain a weighted error sequence, and performing a nonlinear integration operation on the weighted error sequence using an error compensation integrator to obtain a cumulative error curve; Based on the error compensation network, the time-varying features of the cumulative error curve are extracted to obtain an error compensation factor, and the error compensation factor is adaptively corrected to obtain a cumulative phase error; wherein the cumulative phase error includes a static deviation component and a dynamic fluctuation component.

8. A permanent magnet synchronous motor rotor position estimation system based on high frequency signal injection, characterized in that: include: A modulation module, used for performing amplitude modulation on a preset high-frequency voltage carrier signal through an orthogonal rotary modulator to obtain an injection signal vector; An excitation module, used for performing voltage excitation on the stator winding of the permanent magnet synchronous motor based on the injection signal vector to obtain a high-frequency current response signal; A separation module, used for performing harmonic separation on the high-frequency current response signal through a comb filter group to obtain a rotor position modulation current component; A demodulation module, used for performing orthogonal demodulation on the rotor position modulated current component through a complex demodulator to obtain a position-dependent phase difference value; The mapping module is used to perform nonlinear mapping processing on the position-dependent phase difference value through an adaptive state observer to obtain a real-time rotor position angle value.

9. A computer device comprising a memory and a processor, wherein a computer program is stored in the memory, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images