Method and system for suppressing current harmonic disturbance of permanent magnet synchronous motor

Abstract

The invention discloses a current harmonic disturbance suppression method and suppression system for a permanent magnet synchronous motor. The method comprises the following steps: acquiring voltage and current signals of an electrical loop phasor test in the operation of the motor; an inductance identification algorithm based on high-frequency response is adopted, high-frequency excitation signals are injected, current response is monitored, and stator inductance parameters are calculated; constructing a multi-synchronous rotating coordinate system based on the parameter, and decomposing the current signal to obtain each harmonic current component; building a fuzzy full-order sliding mode observer, inputting an inductance parameter, voltage data and a harmonic current component, and dynamically adjusting a sliding mode surface parameter to observe the position and the rotating speed of a rotor; independently controlling each harmonic current adjusting channel under a multi-coordinate system in combination with observation information, and generating a compensation control signal; and the synthesized master control signal is output to the driving circuit. The system comprises six units, and the method and the system can accurately obtain parameters, efficiently decompose harmonic waves, accurately observe states, realize multi-frequency harmonic disturbance suppression, and improve motor operation stability and efficiency.

Description

Technical Field

[0001] This invention relates to the field of motor current harmonic disturbance suppression technology, and in particular to a method and system for suppressing current harmonic disturbances in a permanent magnet synchronous motor. Background Technology

[0002] Permanent magnet synchronous motors (PMSMs) are widely used in industrial drives and new energy equipment due to their high efficiency and high power density. However, during motor operation, the voltage and current signals in the electrical circuit are easily affected by factors such as load fluctuations and magnetic circuit saturation, generating multi-frequency current harmonic disturbances. These disturbances can lead to motor torque pulsation, increased losses, and may also interfere with the normal operation of the drive circuit, reducing the overall system's operational stability and lifespan. To address this issue, relevant technical fields need to develop suppression schemes that balance accurate parameter acquisition, effective harmonic decomposition, and dynamic control to achieve efficient suppression of motor current harmonic disturbances and ensure reliable operation of PMSMs under various operating conditions.

[0003] Existing technologies for suppressing harmonic disturbances in permanent magnet synchronous motors (PMSMs) suffer from two significant drawbacks. Firstly, traditional inductance identification methods struggle to adapt to dynamic changes during motor operation, relying heavily on offline measurements under fixed conditions. This fails to capture real-time changes in inductance parameters during high-frequency responses in the electrical circuit, leading to errors in subsequent harmonic decomposition and control due to inaccurate parameters, thus impacting harmonic suppression effectiveness. Secondly, commonly used observer structures lack dynamic adjustment capabilities. When faced with harmonic current component interference, they struggle to accurately observe rotor position and speed information and are susceptible to issues such as sliding mode chattering. This prevents reliable state feedback for harmonic compensation control, resulting in decreased control accuracy of each harmonic current adjustment channel and hindering effective suppression of multi-frequency harmonic disturbances. Summary of the Invention

[0004] In order to overcome the shortcomings and deficiencies of the existing technology, the present invention provides a method and system for suppressing current harmonic disturbances in permanent magnet synchronous motors.

[0005] The technical solution adopted in this invention is a method for suppressing current harmonic disturbances in a permanent magnet synchronous motor, comprising the following steps:

[0006] S1: Voltage and current signals obtained from phasor testing of the electrical circuit during the operation of the permanent magnet synchronous motor are acquired. These signals are continuously captured at a set sampling frequency to obtain multiple sets of time-series voltage and current data. S2: A high-frequency response-based inductance identification algorithm is used to process the voltage and current data acquired in S1. A high-frequency excitation signal of a preset frequency is injected into the electrical circuit, and the current change response under the corresponding high-frequency excitation signal is monitored. The stator inductance parameters of the motor are calculated by combining the voltage and current data. S3: Based on the stator inductance parameters obtained in S2, a coordinate system structure corresponding to the multi-synchronous rotating coordinate system harmonic suppression strategy is constructed. Rotating coordinate systems corresponding to different harmonic orders are set, and the current signals acquired in S1 are mapped to each of these rotating coordinate systems, decomposing them to obtain different harmonics. S4. Construct a fuzzy full-order sliding mode observer, using the stator inductance parameters obtained in S2 as one of the observer's input parameters. Simultaneously, input the voltage data from S1 and the harmonic current components obtained from S3 decomposition. Dynamically adjust the sliding mode parameters using the fuzzy logic rules within the observer to observe the rotor position and speed information of the motor. S5. Based on the rotor position and speed information observed in S4, and combined with the harmonic current components decomposed in S3, independently control the current adjustment channels corresponding to each harmonic in a multi-synchronous rotating coordinate system to generate compensation control signals corresponding to each harmonic. S6. Synthesize the harmonic compensation control signals generated in S5 to obtain a total control signal. Output the total control signal to the drive circuit of the permanent magnet synchronous motor to suppress motor current harmonic disturbances.

[0007] Furthermore, in the S2 inductor identification algorithm based on high-frequency response, the formula for constructing the inductor identification model is as follows: Among them, L s U represents the stator inductance of the motor; hf Indicates the effective value of the high-frequency excitation voltage injected into the electrical circuit; R s I represents the stator resistance of the motor; hf ω represents the effective value of the high-frequency current in the electrical circuit under the action of a high-frequency excitation voltage; hf This represents the angular frequency of the high-frequency excitation signal; simultaneously, in the S3 multi-synchronous rotating coordinate system harmonic suppression strategy, the harmonic current decomposition model formula is constructed as follows: Among them, i n i represents the effective value of the nth harmonic current component; a i b i c These represent the instantaneous values ​​of the three-phase stator currents; n represents the harmonic order; and θ represents the electrical angle of the motor rotor.

[0008] Furthermore, in the process of building the S4 fuzzy full-order sliding mode observer, the formula for the dynamic adjustment model of the observer's sliding surface is as follows: Where s represents the sliding surface parameters; e ω This represents the deviation between the observed rotational speed and the actual rotational speed; k1 and k2 represent the coefficients for dynamic adjustment of the fuzzy logic rules. The derivative represents the speed deviation; simultaneously, in the S5 harmonic current regulation control, the formula for constructing the compensation control signal generation model is as follows: Among them, u n K represents the compensation control signal corresponding to the nth harmonic; pn K represents the proportionality coefficient. in Indicates the integral coefficient; Indicates the command value of the nth harmonic current; i n This represents the actual value of the nth harmonic current.

[0009] Furthermore, in the S2 inductor identification algorithm based on high-frequency response, the formula for constructing the high-frequency response characteristic analysis model is as follows: Among them, G hf (s) represents the high-frequency response transfer function; I hf (s) represents the Laplace transform of the high-frequency current; U hf (s) represents the Laplace transform of the high-frequency voltage; R s L represents the stator resistance. s θ represents the stator inductance; s represents the complex frequency variable; in the construction of the S3 multi-synchronous rotating coordinate system, the formula for calculating the rotation angle of the coordinate system is as follows: n =ω0t+nδ, where θ n ω0 represents the rotation angle of the rotating coordinate system corresponding to the nth harmonic; t represents time; n represents the harmonic order; and δ represents the initial phase angle.

[0010] Furthermore, in the S4 fuzzy full-order sliding mode observer, the formula for constructing the fuzzy inference model is as follows: Where Δk represents the sliding mode gain adjustment amount; e represents the rotor position observation deviation; The derivative of the rotor position observation deviation is represented by f(·); f(·) represents the fuzzy inference function; in the S5 compensation control signal synthesis, the formula for constructing the control signal synthesis model is as follows: Among them, u total Indicates the total control signal; N represents the highest harmonic order to be suppressed; u n θ represents the nth harmonic compensation control signal; n This represents the rotation angle of the rotating coordinate system for the nth harmonic.

[0011] Furthermore, in the S2 inductor identification process, the inductor temperature compensation model formula is constructed as follows: L sT =L s [1+α(T-T0)], where L sT L represents the stator inductance after temperature compensation. s The stator inductance is represented at the reference temperature; α represents the inductance temperature coefficient; T represents the current motor temperature; T0 represents the reference temperature; the filter model formula for S3 harmonic current extraction is as follows: Among them, i nf τ represents the nth harmonic current component after filtering; n i represents the filtering time constant corresponding to the nth harmonic; n The nth harmonic current component before filtering is represented; S represents the complex frequency variable.

[0012] Further, S3 includes the following sub-steps: S31, based on the stator inductance parameters obtained in S2, determine the number of multiple synchronous rotating coordinate systems and the harmonic order corresponding to each coordinate system, wherein the number of coordinate systems corresponds one-to-one with the harmonic order to be suppressed, and the rotation direction of each coordinate system is consistent with the rotation direction of the corresponding harmonic; S32, perform Clarke transform on the three-phase current signal collected in S1, converting the current signal in the three-phase stationary coordinate system into the current signal in the two-phase stationary coordinate system, eliminating the coupling components in the three-phase current signal; S33, calculate the rotation angle of each coordinate system according to the harmonic order corresponding to each rotating coordinate system, wherein the rotation angle is the product of the fundamental rotation angle and the harmonic order, and perform Park transform on the current signal in the two-phase stationary coordinate system to obtain the direct-axis current component and quadrature-axis current component in each rotating coordinate system; S34, perform filtering processing on the direct-axis current component and quadrature-axis current component in each rotating coordinate system to remove the high-frequency interference components introduced during the transformation process, and obtain the pure current component corresponding to each harmonic.

[0013] Further, S4 includes the following sub-steps: S41, based on the mathematical model of the permanent magnet synchronous motor, construct the state equation of the full-order sliding mode observer, and substitute the stator inductance parameters and stator resistance parameters obtained in S2 into the state equation to determine the initial structural parameters of the observer; S42, design the input and output variables of the fuzzy logic controller, with the input variables being the rotor position observation deviation and the speed observation deviation, and the output variable being the sliding mode gain adjustment amount of the sliding mode observer, and establish the fuzzy membership function and fuzzy inference rule base of the input variables; S43, input the voltage data in S1 and the harmonic current components obtained from S3 into the full-order sliding mode observer, and calculate the rotor position observation value and speed observation value through the observer's state equation; S44, input the rotor position observation deviation and speed observation deviation into the fuzzy logic controller, output the sliding mode gain adjustment amount according to the fuzzy inference rule base, dynamically adjust the sliding mode gain of the observer, and reduce the chattering phenomenon of the observer.

[0014] Further, S5 includes the following sub-steps: S51, setting the command values ​​of each harmonic current according to the motor operating conditions, wherein the command values ​​are dynamically adjusted according to the motor load and speed range to ensure that the command values ​​of each harmonic current match the command value of the fundamental current; S52, calculating the current deviation corresponding to each harmonic, wherein the current deviation is the difference between the command value of each harmonic current and the actual value of each harmonic current obtained by decomposition in S3, and performing polarity judgment and amplitude calculation on the current deviation; S53, building a PI controller in each synchronous rotating coordinate system, inputting the current deviation into the PI controller, and adjusting the current deviation through the proportional and integral links of the PI controller to generate the initial compensation control signal corresponding to each harmonic; S54, combining the rotor speed information observed in S4, performing speed feedforward compensation on the initial compensation control signal to eliminate the influence of speed changes on the compensation control signal, and obtaining the final harmonic compensation control signal.

[0015] A permanent magnet synchronous motor current harmonic disturbance suppression system includes: an electrical circuit phasor test signal acquisition unit for continuously acquiring voltage and current signals during the operation of the permanent magnet synchronous motor and outputting multiple sets of time-series voltage and current data; a stator inductance identification unit based on high-frequency response, connected to the electrical circuit phasor test signal acquisition unit, receiving the voltage and current data, calculating and outputting stator inductance parameters by injecting a high-frequency excitation signal and monitoring the current response; a multi-synchronous rotating coordinate system harmonic current decomposition unit, connected to both the electrical circuit phasor test signal acquisition unit and the high-frequency response stator inductance identification unit, receiving the voltage data, current data, and stator inductance parameters, constructing a multi-synchronous rotating coordinate system, and decomposing and outputting each harmonic current component; and a fuzzy full-order sliding mode observation. The measurement unit is connected to the electrical circuit phasor test signal acquisition unit, the stator inductance identification unit based on high-frequency response, and the multi-synchronous rotating coordinate system harmonic current decomposition unit, respectively. It receives the voltage data, stator inductance parameters, and harmonic current components, and observes and outputs rotor position and speed information. The multi-channel harmonic compensation control signal generation unit is connected to the multi-synchronous rotating coordinate system harmonic current decomposition unit and the fuzzy full-order sliding mode observation unit, respectively. It receives the harmonic current components, rotor position and speed information, and generates and outputs harmonic compensation control signals. The total control signal synthesis and drive unit is connected to the multi-channel harmonic compensation control signal generation unit. It receives the harmonic compensation control signals and synthesizes the total control signal, and outputs the total control signal to the drive circuit of the permanent magnet synchronous motor.

[0016] Beneficial effects: This invention proposes a method and system for suppressing current harmonic disturbances in permanent magnet synchronous motors, which can effectively improve the stability and energy conversion efficiency of the motor during operation, reduce torque pulsation caused by current harmonics, reduce internal energy loss of the motor, ensure long-term stable operation of the drive circuit, extend the service life of the entire motor system, and can be adapted to different load fluctuation amplitudes and speed change ranges, achieving precise and efficient suppression of multi-frequency current harmonic disturbances. To address the issue that traditional inductance identification methods cannot obtain accurate parameters dynamically in real time, a high-frequency response-based inductance identification algorithm is adopted. This algorithm injects a specific frequency high-frequency excitation signal into the electrical circuit and monitors the current change response, capturing real-time changes in the motor stator inductance parameters. This provides precise parameter support for subsequent harmonic decomposition and compensation control in a multi-synchronous rotating coordinate system, avoiding harmonic suppression errors caused by parameter deviations. Furthermore, to address the problems of commonly used observers lacking dynamic adjustment capabilities, having low observation accuracy, and being susceptible to chattering, a fuzzy full-order sliding mode observer is constructed. Combined with fuzzy logic rules, the sliding mode surface parameters are dynamically adjusted, effectively reducing chattering and accurately observing rotor position and speed information. This provides reliable state feedback for harmonic compensation control, improves the control accuracy of each harmonic current adjustment channel, and effectively solves the problem of poor harmonic suppression in the background technology. Attached Figure Description

[0017] Figure 1 This is a flowchart of the method steps of the present invention;

[0018] Figure 2 This is a diagram showing the system unit composition of the present invention. Detailed Implementation

[0019] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. The application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0020] like Figure 1 As shown, a method for suppressing current harmonic disturbances in a permanent magnet synchronous motor includes the following steps:

[0021] S1, collects voltage and current signals obtained from phasor testing of electrical circuit during the operation of permanent magnet synchronous motor, continuously captures the voltage and current signals through a set sampling frequency, and obtains multiple sets of time-series voltage and current data;

[0022] Specifically, step S1 is the fundamental data acquisition stage for suppressing current harmonic disturbances in permanent magnet synchronous motors. It provides accurate and continuous raw electrical signals for all subsequent technical stages, ensuring reliable data support for steps such as inductor identification, harmonic decomposition, and observer setup. If the voltage and current signals acquired in this stage are missing or contain errors, it will directly lead to deviations in subsequent parameter calculations and a decrease in control accuracy, thus affecting the harmonic suppression effect. This step must prioritize ensuring the continuity and accuracy of signal acquisition, covering the signal characteristics under different motor operating conditions, including startup, steady-state operation, and load changes, to comprehensively reflect the true state of the electrical circuit during motor operation, laying a data foundation for subsequent targeted harmonic disturbance suppression.

[0023] The specific implementation process is as follows: First, select a signal acquisition module with high-frequency sampling capability, and set the sampling frequency to 10kHz-20kHz to ensure that the voltage and current signal details corresponding to high-frequency harmonics during motor operation can be captured. Second, connect the voltage acquisition channel of the acquisition module to the three-phase voltage output terminal of the motor stator, and connect the current acquisition channel to the three-phase stator circuit through a Hall current sensor in series. The sensor measurement range is set to 0-50A (suitable for permanent magnet synchronous motors with rated current of 10-30A), and the measurement accuracy is controlled within ±0.5%. Then, set the acquisition duration: 5-10 seconds for motor start-up conditions, 15-20 seconds for steady-state operation conditions, and 10-15 seconds for load change conditions (load increases from 20% of rated load to 100% of rated load). Finally, convert the acquired voltage and current signals into digital quantities through the data transmission interface and store them as time-series data groups. Each data group includes a timestamp, instantaneous three-phase voltage value, and instantaneous three-phase current value. The data storage interval is consistent with the sampling frequency to ensure that it can accurately correspond to the motor operation sequence when called later.

[0024] S2, using an inductor identification algorithm based on high-frequency response, processes the voltage and current data obtained in S1, injects a high-frequency excitation signal of a preset frequency into the electrical circuit, monitors the change response of the current under the corresponding high-frequency excitation signal, and calculates the motor stator inductance parameters by combining the voltage and current data.

[0025] Specifically, step S2 is the key parameter acquisition stage for achieving accurate harmonic suppression. Using an inductor identification algorithm based on high-frequency response, the real-time stator inductance parameters of the motor are obtained. As a core parameter of the motor's mathematical model, the stator inductance directly affects the accuracy of harmonic decomposition in the multi-synchronous rotating coordinate system and the observation precision of the fuzzy full-order sliding mode observer. Using a fixed inductance value or an offline measured inductance value cannot adapt to inductance fluctuations caused by magnetic circuit saturation and temperature changes during motor operation, leading to a mismatch between subsequent control strategies and the actual motor state. Therefore, this step needs to track inductance parameter changes in real time to provide dynamic and accurate parameter input for subsequent stages, ensuring the adaptive capability of the entire suppression system.

[0026] The specific implementation process is as follows: First, determine the high-frequency excitation signal parameters. The excitation signal frequency is set to 500Hz-1kHz (higher than the motor fundamental frequency of 50Hz / 60Hz to avoid superposition and interference with the fundamental signal), and the excitation signal amplitude is set to 5%-10% of the motor's rated voltage (e.g., for a motor with a rated voltage of 380V, the excitation voltage amplitude is 19V-38V) to ensure that an observable current response is generated without affecting the normal operation of the motor. Second, inject the high-frequency excitation signal into any one phase of the three-phase electrical circuit of the motor stator. Through the signal acquisition module built in step S1, synchronously acquire the voltage and current data before and after the injection of the excitation signal. The acquisition time is set to 2-3 seconds to ensure that the complete excitation response process is obtained. Subsequently, perform trend analysis on the acquired current signal after excitation to extract the amplitude and phase change data of the current response. Finally, combine the acquired excitation voltage data and current response data to calculate the real-time stator inductance parameters through an algorithm. The calculation result update cycle is set to 0.5 seconds to ensure that the inductance parameters can track the changes in the motor's operating state in real time, and the parameter error is controlled within ±2%.

[0027] S3. Based on the stator inductance parameters obtained in S2, construct the coordinate system structure corresponding to the multi-synchronous rotating coordinate system harmonic suppression strategy, set rotating coordinate systems corresponding to different harmonic orders, map the current signals collected in S1 to each of the rotating coordinate systems respectively, and decompose the current components corresponding to different harmonic orders.

[0028] Specifically, step S3 achieves precise harmonic separation by decomposing the mixed current signal into harmonic current components of different orders, providing a prerequisite for subsequent independent control of each harmonic. Since motor current harmonics include multiple frequencies such as the 3rd, 5th, and 7th, and different harmonic frequencies have different effects on motor operation, if each harmonic cannot be effectively separated, only an overall control strategy can be adopted, making targeted suppression difficult and resulting in poor harmonic suppression. This step constructs a multi-synchronous rotating coordinate system based on the real-time stator inductance parameters obtained in step S2. The rotational characteristics of the coordinate system can be used to convert the harmonic current of a specific frequency into a DC quantity, thereby achieving precise extraction and laying the foundation for subsequent independent compensation control.

[0029] The specific implementation process is as follows: First, based on the common harmonic frequencies of the motor, determine the number of multiple synchronous rotating coordinate systems. Typically, this includes a fundamental coordinate system, a 3rd harmonic coordinate system, a 5th harmonic coordinate system, and a 7th harmonic coordinate system, for a total of four coordinate systems. If higher frequency harmonics exist during motor operation, additional corresponding coordinate systems can be added. Second, combining the stator inductance parameters obtained in step S2, determine the parameter settings for each coordinate system, including the rotational angular velocity (the angular velocity of the fundamental coordinate system is consistent with the electric angular velocity of the motor rotor, the angular velocity of the 3rd harmonic coordinate system is 3 times that of the fundamental, and so on), and the initial phase of the coordinate system (consistent with the initial phase of the motor rotor). The deviation is controlled within ±5°. Then, the three-phase current signal collected in step S1 is converted into a current signal in a two-phase stationary coordinate system. The standard coordinate transformation rule is adopted during the conversion process to ensure the signal conversion accuracy. Finally, the current signal in the two-phase stationary coordinate system is mapped to each synchronous rotating coordinate system. Through coordinate transformation operation, each harmonic current is converted into the direct axis current component and quadrature axis current component in the corresponding coordinate system to complete the decomposition of each harmonic current component. The update cycle of the decomposed current component data is consistent with the update cycle of the inductance parameter in step S2, which is 0.5 seconds. The component extraction error is controlled within ±3%.

[0030] S4. Construct a fuzzy full-order sliding mode observer. Use the stator inductance parameters obtained in S2 as one of the input parameters of the observer. At the same time, input the voltage data in S1 and the harmonic current components obtained by decomposition in S3. Dynamically adjust the sliding mode parameters through the fuzzy logic rules inside the observer to observe the rotor position and speed information of the motor.

[0031] Specifically, step S4 involves acquiring key operating status information of the motor. This is achieved by observing the rotor position and speed information using a fuzzy full-order sliding mode observer, providing necessary status feedback for subsequent harmonic compensation control. Rotor position and speed are critical variables for motor control, directly affecting the phase and amplitude calculations of the harmonic compensation control signal. Using traditional sensors to acquire this information would increase system cost and installation complexity, and is prone to failure under harsh operating conditions. Furthermore, ordinary sliding mode observers suffer from chattering issues, affecting observation accuracy. Therefore, this step employs a fuzzy full-order sliding mode observer, combining the stator inductance parameters from step S2 with the harmonic current components from step S3. This enables high-precision observation without sensors, while simultaneously mitigating chattering through fuzzy logic.

[0032] The specific implementation process is as follows: First, the basic structure of the full-order sliding mode observer is built. Based on the mathematical model of the permanent magnet synchronous motor, the state equation of the observer is set. The stator inductance parameters obtained in step S2 and the pre-measured stator resistance parameters (resistance error controlled within ±1%) are substituted into the state equation to determine the initial parameters of the observer. Second, a fuzzy logic control module is designed. The input variables are set as rotor position observation deviation (range -π to π radians) and speed observation deviation (range -50 to 50 rad / s), and the output variable is set as sliding mode gain adjustment (range 0.1 to 1.0). Five fuzzy subsets are established (negative large, negative small, zero, positive small, positive large). Twenty-five fuzzy inference rules were established to ensure dynamic adjustment of gain based on deviation. Subsequently, the voltage data (instantaneous values ​​of three-phase voltage) collected in step S1 and the harmonic current components (direct-axis and quadrature-axis components) decomposed in step S3 were input into the observer. The preliminary rotor position observation value and speed observation value were obtained by calculating through the state equation. Finally, the position observation deviation and speed observation deviation were input into the fuzzy logic module, and the sliding mode gain adjustment amount was output to dynamically adjust the sliding mode gain of the observer, reduce chattering phenomenon, and set the observation result update cycle to 0.1 seconds. The position observation error was controlled within ±0.05π radians, and the speed observation error was controlled within ±2 rad / s.

[0033] S5. Based on the rotor position and speed information obtained from S4, and combined with the harmonic current components decomposed in S3, the current regulation channels corresponding to each harmonic are independently controlled in the multi-synchronous rotating coordinate system to generate the compensation control signal corresponding to each harmonic.

[0034] Specifically, step S5 generates targeted harmonic suppression signals. Based on the rotor state information observed in step S4 and the harmonic current components decomposed in step S3, independent compensation control channels are designed for each harmonic to achieve precise suppression. Since different harmonics have different frequencies and amplitudes, their impact on the motor also varies. Using uniform compensation control cannot simultaneously address the suppression requirements of each harmonic, easily leading to over-suppression of some harmonics and under-suppression of others. Therefore, this step independently controls each harmonic in a multi-synchronous rotating coordinate system. Control parameters can be adjusted according to the actual situation of each harmonic to ensure targeted suppression of each harmonic, thereby improving the overall harmonic suppression effect.

[0035] The specific implementation process is as follows: First, independent current adjustment channels are set for the synchronous rotating coordinate system corresponding to each harmonic, including the 3rd, 5th, and 7th harmonic channels, each configured with independent control parameters; second, the command values ​​of each harmonic current are set according to the motor operating conditions. The fundamental current command value is set according to the motor load requirements (e.g., set to the motor's rated current under rated load), and the command values ​​of the 3rd, 5th, and 7th harmonic currents are set to 0A to ensure that the goal is to eliminate these harmonic components; then, the current deviation of each harmonic is calculated, that is, the difference between the current command value of each channel and the corresponding harmonic current component obtained from step S3, and the deviation changes are tracked in real time; next, a PI controller is configured for each adjustment channel, and the PI parameters are set according to the harmonic frequency (the proportional coefficient of the 3rd harmonic channel is set to 0.8-). 1.2, with the integral coefficient set to 0.05-0.1; the proportional coefficient for the 5th harmonic channel set to 1.0-1.4, and the integral coefficient set to 0.03-0.08; and the proportional coefficient for the 7th harmonic channel set to 1.2-1.6, and the integral coefficient set to 0.02-0.06), preliminary compensation control signals for each harmonic are generated through PI regulation. Finally, combined with the rotor speed information observed in step S4, the preliminary compensation control signals are corrected by speed feedforward. The correction coefficient is set according to the speed (0.9-1.0 for speeds below 1000 rpm, 1.0-1.1 for speeds between 1000-3000 rpm, and 1.1-1.2 for speeds above 3000 rpm), resulting in the final harmonic compensation control signals. The signal update period is 0.1 seconds.

[0036] S6. The harmonic compensation control signals generated in S5 are combined to obtain the total control signal. The total control signal is then output to the drive circuit of the permanent magnet synchronous motor to suppress the harmonic disturbance of the motor current.

[0037] Specifically, step S6 is the final execution stage for harmonic disturbance suppression. It synthesizes the harmonic compensation control signals generated in step S5 to form a total control signal that can directly drive the motor. This signal is then applied to the motor through the drive circuit to complete the entire suppression process. If the compensation control signal is not properly synthesized and the harmonic signals are directly output, mutual interference between the signals will occur, failing to form an effective control effect and potentially even introducing new disturbances. Therefore, this step requires a scientific synthesis method to ensure that the effects of the harmonic compensation signals are superimposed to form a targeted total control signal that accurately cancels the harmonic components in the motor current, achieving the suppression target.

[0038] The specific implementation process is as follows: First, the synthesis rules for the compensation control signals are determined. A time-domain superposition method is adopted, superimposing the 3rd, 5th, and 7th harmonic compensation control signals generated in step S5 with the fundamental wave control signal (generated according to motor operating requirements) on the same time dimension. Second, a time synchronization mechanism for signal synthesis is set to ensure that the timestamps of each harmonic compensation control signal and the fundamental wave control signal are completely consistent, avoiding distortion of the synthesized signal due to time deviation, with synchronization accuracy controlled within ±0.01 seconds. Subsequently, the amplitude of the synthesized signal is limited, restricting the amplitude of the total control signal to within 0-110% of the motor's rated voltage (e.g., for a motor with a rated voltage of 380V, the total control signal amplitude is limited to 0-110% of the motor's rated voltage). The signal amplitude is limited to 0-418V to prevent overload of the drive circuit or damage to the motor due to excessive signal amplitude. Next, the synthesized digital control signal is converted into an analog signal through a D / A conversion module with a conversion accuracy set to 12 bits to ensure that the analog signal can accurately reflect the changes in the digital signal. Finally, the converted analog total control signal is output to the three-phase bridge drive circuit of the motor. The drive circuit adjusts the voltage output to the motor stator according to the control signal, thereby changing the stator current, canceling the original harmonic components, and realizing the suppression of current harmonic disturbances. The signal output frequency is consistent with the update cycle of the compensation control signal in step S5, which is 0.1 seconds, to ensure real-time response to changes in the motor's operating status.

[0039] Preferably, in the S2 inductor identification algorithm based on high-frequency response, the formula for constructing the inductor identification model is as follows: Among them, L s U represents the stator inductance of the motor; hf Indicates the effective value of the high-frequency excitation voltage injected into the electrical circuit; R s I represents the stator resistance of the motor; hf ω represents the effective value of the high-frequency current in the electrical circuit under the action of a high-frequency excitation voltage; hf This represents the angular frequency of the high-frequency excitation signal; simultaneously, in the S3 multi-synchronous rotating coordinate system harmonic suppression strategy, the harmonic current decomposition model formula is constructed as follows: Among them, i n i represents the effective value of the nth harmonic current component; a i b i c These represent the instantaneous values ​​of the three-phase stator currents; n represents the harmonic order; and θ represents the electrical angle of the motor rotor.

[0040] Specifically, models are constructed for the high-frequency response-based inductor identification algorithm in S2 and the multi-synchronous rotating coordinate system harmonic suppression strategy in S3. These models provide quantitative basis for inductance calculation and harmonic decomposition, improving parameter accuracy and decomposition accuracy. In the S2 inductor identification, the parameters involved in the model need to be set according to actual operating conditions. For example, the effective value of the high-frequency excitation voltage is usually taken as 5%-10% of the rated voltage of the motor (taking a 380V rated voltage motor as an example, the value range is 19V-38V). The angular frequency of the high-frequency excitation signal is set to the angular frequency value corresponding to 500Hz-1kHz. The stator resistance is obtained through offline measurement, with the error controlled within ±1%. The effective value of the high-frequency current is acquired in real time by the signal acquisition module, with an accuracy of ±0.5%. The stator inductance parameters calculated by this model have an update cycle of 0.5 seconds and an error of ±2%, providing dynamic parameter support for subsequent stages. In the S3 harmonic current decomposition... In this process, the model parameters need to match the motor's operating state. The instantaneous values ​​of the three-phase stator current are acquired by the S1 acquisition module at a frequency of 10kHz-20kHz. The harmonic order is set according to the common harmonics of the motor (usually the 3rd, 5th, and 7th). The rotor electrical angle is provided by the subsequent observer with an error of ±0.05π radians. This model can decompose the mixed current signal into harmonic current components with an extraction error of ±3%. The update cycle of the decomposition results is consistent with the inductance parameters, laying the foundation for targeted suppression of each harmonic. During implementation, it is necessary to ensure the synchronization of the acquired data and the timing matching of the model calculation and data update to avoid errors caused by parameter delays or data asynchrony.

[0041] Preferably, in the process of building the S4 fuzzy full-order sliding mode observer, the formula for constructing the dynamic adjustment model of the observer sliding surface is as follows: Where s represents the sliding surface parameters; e ω This represents the deviation between the observed rotational speed and the actual rotational speed; k1 and k2 represent the coefficients for dynamic adjustment of the fuzzy logic rules. The derivative represents the speed deviation; simultaneously, in the S5 harmonic current regulation control, the formula for constructing the compensation control signal generation model is as follows: Among them, u n K represents the compensation control signal corresponding to the nth harmonic; pn K represents the proportionality coefficient. in Indicates the integral coefficient; Indicates the command value of the nth harmonic current; i n This represents the actual value of the nth harmonic current.

[0042] Specifically, the construction of the S4 fuzzy full-order sliding mode observer and the S5 harmonic current regulation control, through model building to improve observation accuracy and control signal generation accuracy, are key to ensuring harmonic suppression effectiveness. In the S4 sliding surface dynamic adjustment model, parameter settings need to balance observation stability and response speed. The rotational speed deviation ranges from -50 rad / s to 50 rad / s. The coefficients k1 and k2 for dynamic adjustment of the fuzzy logic rules are set to 0.1-0.5 and 0.05-0.2, respectively. The derivative of the rotational speed deviation is calculated through numerical differentiation, with an error of ±1 rad / s. 2 This model, by integrating the rotational speed deviation and its real-time changes, dynamically adjusts the sliding surface parameters to reduce chattering, controlling the rotor position observation error within ±0.05π radians, the rotational speed observation error within ±2 rad / s, and the observation result update cycle within 0.1 seconds; in the S5 compensation control signal generation model, the proportional coefficient K pn With integral coefficient K in It needs to be adapted according to the harmonic frequency, the 3rd harmonic K pn Take 0.8-1.2, K in Taking values ​​of 0.05-0.1, the 5th harmonic K pn Take values ​​of 1.0-1.4, K in Taking values ​​between 0.03 and 0.08, the 7th harmonic K... pn The values ​​are set to 1.2-1.6 for 1.2 and 0.02-0.06 for Kin. The harmonic current command value is set according to the suppression target (usually 0A). The actual value is obtained from the S3 decomposition with an error of ±3%. The model generates a compensation control signal through PI regulation calculation. The signal amplitude is limited to 5%-15% of the rated voltage, and the update cycle is 0.1 seconds. During implementation, the offline calibration of the fuzzy rules and the debugging of the PI parameters must be completed first to ensure that the model can output stably under different operating conditions. At the same time, it is necessary to keep the timing synchronized with the S4 observation results and the S3 decomposition data to avoid parameter mismatch affecting the control effect.

[0043] Preferably, in the S2 inductor identification algorithm based on high-frequency response, the formula for constructing the high-frequency response characteristic analysis model is as follows: Among them, G hf (s) represents the high-frequency response transfer function; I hf (s) represents the Laplace transform of the high-frequency current; U hf (s) represents the Laplace transform of the high-frequency voltage; R s L represents the stator resistance. s θ represents the stator inductance; s represents the complex frequency variable; in the construction of the S3 multi-synchronous rotating coordinate system, the formula for calculating the rotation angle of the coordinate system is as follows: n =ω0t+nδ, where θ nω0 represents the rotation angle of the rotating coordinate system corresponding to the nth harmonic; t represents time; n represents the harmonic order; and δ represents the initial phase angle.

[0044] Specifically, the design model for the S2 high-frequency response characteristics analysis and the S3 multi-synchronous rotating coordinate system construction aims to optimize the dynamic characteristic analysis of the inductor identification process and the accurate calculation of the coordinate system rotation angle, thereby improving the adaptability of the overall technical solution. In the S2 high-frequency response characteristic analysis model, parameters must closely match the high-frequency excitation and motor electrical characteristics. The stator resistance is an offline measured value (error ±1%), and the stator inductance is a real-time calculated value (error ±2%). The complex frequency variable is set according to the characteristics of the high-frequency excitation signal. The Laplace transform of the high-frequency voltage and current is indirectly obtained through the Fourier transform of the signal acquisition data, with a transformation accuracy of ±2%. This model quantifies the relationship between voltage and current at high frequencies through the transfer function form, which can assist in analyzing the dynamic error of inductor identification and provide a basis for adjusting the excitation signal parameters (e.g., if the excitation frequency is too high and causes instability in the transfer function, it needs to be lowered to 500Hz-800Hz). The S3 coordinate system rotation angle calculation model... In this model, the fundamental angular frequency is matched with the rated speed of the motor (e.g., a 1500 rpm motor corresponds to a fundamental angular frequency of 157 rad / s). The time parameter is provided by the system clock (accuracy ±0.001 seconds). The harmonic orders cover the 3rd, 5th, and 7th harmonics. The initial phase angle is determined by the position detection when the motor starts (error ±5°). The rotation angle update cycle calculated by the model is consistent with the inductance parameter (0.5 seconds) to ensure that the rotation of each coordinate system is synchronized with the corresponding harmonic. During implementation, the high-frequency response transfer function needs to be verified offline to ensure that the model converges under the set excitation parameters. At the same time, the initial phase angle needs to be calibrated to avoid the current mapping decomposition error from increasing due to phase deviation, which would affect the accuracy of subsequent harmonic control.

[0045] Preferably, in the S4 fuzzy full-order sliding mode observer, the formula for constructing the fuzzy inference model is as follows: Where Δk represents the sliding mode gain adjustment amount; e represents the rotor position observation deviation; The derivative of the rotor position observation deviation is represented by f(·); f(·) represents the fuzzy inference function; in the S5 compensation control signal synthesis, the formula for constructing the control signal synthesis model is as follows: Among them, u total Indicates the total control signal; N represents the highest harmonic order to be suppressed; u n θ represents the nth harmonic compensation control signal; n This represents the rotation angle of the rotating coordinate system for the nth harmonic.

[0046] Specifically, a model is introduced into the synthesis of the S4 fuzzy full-order sliding mode observer and the S5 compensated control signal to optimize the gain adjustment logic of the observer and the synthesis rules of the control signal, thereby improving the dynamic response and suppression accuracy of the system. In the S4 fuzzy inference model, the rotor position observation deviation ranges from -π to π radians, the deviation derivative ranges from -10 rad / s to 10 rad / s, the fuzzy inference function is constructed based on 25 fuzzy rules (input variables are divided into 5 fuzzy subsets), and the sliding mode gain adjustment output range is 0.1-1.0. This model achieves dynamic gain adaptation by associating the position deviation and its real-time changes with the sliding mode gain. When the position deviation is large (e.g., exceeding π / 2 radians), the gain is adjusted to 0.8-1.0 to accelerate convergence; when the deviation is small (e.g., less than π / 10 radians), it is adjusted to 0.1-0.3 to reduce chattering, stabilizing the observation accuracy within ±0.05π radians. In the S5 control signal synthesis model, it is necessary to suppress... The highest harmonic order is set to 7th. The amplitude of each harmonic compensation control signal is the PI adjustment output value (error ±3%). The rotation angle of the rotating coordinate system is the calculated value of S4 (error ±0.02π radians). During synthesis, it is necessary to ensure that the timestamps of each signal are synchronized (accuracy ±0.01 seconds). The amplitude of the total control signal is limited to 0-110% of the rated voltage. This model avoids mutual interference between harmonic control signals by cosine weighted superposition of multiple signals. During implementation, the fuzzy inference rules need to be simulated and verified first (such as building a model in MATLAB to test the rationality of gain adjustment under different deviations). At the same time, it is necessary to calibrate the phase relationship between the rotation angle of each coordinate system and the compensation signal to prevent phase shift of the synthesized signal, which would lead to the failure of harmonic suppression.

[0047] Preferably, in the S2 inductor identification process, the formula for constructing the inductor temperature compensation model is as follows: L sT =L s [1+α(T-T0)], where L sT L represents the stator inductance after temperature compensation. s The stator inductance is represented at the reference temperature; α represents the inductance temperature coefficient; T represents the current motor temperature; T0 represents the reference temperature; the filter model formula for S3 harmonic current extraction is as follows: Among them, i nf τ represents the nth harmonic current component after filtering; n i represents the filtering time constant corresponding to the nth harmonic; n represents the nth harmonic current component before filtering; s represents the complex frequency variable.

[0048] Specifically, the design model for S2 inductor temperature compensation and S3 harmonic current filtering aims to eliminate the influence of temperature changes on inductor parameters and high-frequency interference during the decomposition process, thereby improving the system's stability under complex operating conditions. In the S2 inductor temperature compensation model, the reference temperature is set to 25℃ (industrial standard ambient temperature). The current motor temperature is collected by a temperature sensor built into the stator winding (accuracy ±1℃). The inductance temperature coefficient is determined based on the motor core material (e.g., silicon steel sheet material typically takes 0.001 / ℃-0.003 / ℃). The stator inductance at the reference temperature is the offline measurement value at room temperature (error ±1%). The inductance update cycle calculated by this model after temperature compensation is consistent with the temperature acquisition (1 second). When the motor temperature rises to 80℃, the inductance value can be corrected through the model (e.g., reference inductance 10mH, temperature coefficient 0.002 / ℃, corrected to 10×[1+0.002×(80-25)]=11.1mH), avoiding underestimation of the inductance due to temperature rise, which would affect harmonic decomposition. Accuracy: In the S3 harmonic current filtering model, the filtering time constant is set according to the harmonic frequency (0.01 seconds for the 3rd harmonic, 0.005 seconds for the 5th harmonic, and 0.003 seconds for the 7th harmonic). The complex frequency variable is adapted to the filtering characteristics. The harmonic current component before filtering is the result of S3 decomposition (error ±3%). The model weakens the high-frequency interference introduced during the decomposition process through the characteristics of first-order low-pass filtering (such as attenuating interference signals above 1kHz by more than 80%). The error of the current component after filtering is controlled within ±2%. During implementation, it is necessary to first calibrate the correspondence between the temperature sensor and the inductance parameters (such as testing the inductance value at different temperatures through a high and low temperature chamber and correcting the temperature coefficient). At the same time, it is necessary to adjust the filtering time constant to ensure that the dynamic response speed of the harmonic current component is not affected while filtering out interference, and to avoid control lag.

[0049] Preferably, S3 includes the following sub-steps: S31, based on the stator inductance parameters obtained in S2, determine the number of multiple synchronous rotating coordinate systems and the harmonic order corresponding to each coordinate system, wherein the number of coordinate systems corresponds one-to-one with the harmonic order to be suppressed, and the rotation direction of each coordinate system is consistent with the rotation direction of the corresponding harmonic; S32, perform Clarke transform on the three-phase current signal collected in S1, converting the current signal in the three-phase stationary coordinate system into the current signal in the two-phase stationary coordinate system, eliminating the coupling component in the three-phase current signal; S33, calculate the rotation angle of each coordinate system according to the harmonic order corresponding to each rotating coordinate system, wherein the rotation angle is the product of the fundamental rotation angle and the harmonic order, and perform Park transform on the current signal in the two-phase stationary coordinate system to obtain the direct-axis current component and quadrature-axis current component in each rotating coordinate system; S34, perform filtering processing on the direct-axis current component and quadrature-axis current component in each rotating coordinate system to remove the high-frequency interference components introduced during the transformation process, and obtain the pure current component corresponding to each harmonic.

[0050] Specifically, the S3 multi-synchronous rotating coordinate system harmonic current decomposition clarifies the specific implementation of coordinate system construction, signal transformation, coordinate mapping and filtering through four steps, ensuring the accuracy of the decomposition of each harmonic current component and providing a clear harmonic signal basis for subsequent independent control. In S31, the number of coordinate systems must correspond one-to-one with the harmonic orders to be suppressed. Typically, four are set (fundamental + 3rd, 5th, and 7th harmonics). The rotation direction of each coordinate system is consistent with the rotation direction of the corresponding harmonic (e.g., the 5th harmonic rotates in negative sequence, so the coordinate system rotates synchronously in negative sequence). Furthermore, the coordinate system parameter reference must be determined in conjunction with the stator inductance parameters obtained in S2 (error ±2%) to avoid decomposition deviations due to parameter mismatches. In S32, the standard Clarke transform is used to convert the three-phase current signal (acquired by S1 at a frequency of 10kHz-20kHz, accuracy ±0.5%) into a two-phase stationary coordinate system signal. During the transformation, the synchronization of the three-phase current sampling must be ensured (time deviation ≤0.001 seconds) to eliminate three-phase coupling components. After conversion, the error of the two-phase current signal should be controlled within ±1%. In S33, the rotation angle calculation of each coordinate system is based on the fundamental rotation angle (the fundamental rotation angle is consistent with the electric angular velocity of the motor rotor, such as 15°). The 00rpm motor corresponds to 157rad / s. The rotation angles of the 3rd, 5th, and 7th harmonic coordinate systems are 3 times, 5 times, and 7 times that of the fundamental wave, respectively. Combined with the two-phase static current signal, Park transformation is performed. During the transformation, the phase deviation must be controlled within ±5° to obtain the direct axis and quadrature axis current components in each coordinate system. The component calculation update cycle is consistent with the inductance parameter (0.5 seconds). In S34, a first-order low-pass filter is used to filter the decomposed current components. The filter cutoff frequency is set to 100Hz-200Hz (higher than the harmonic frequency and lower than the high-frequency interference frequency). After filtering, the error of each harmonic current component is ≤±3%, effectively removing high-frequency interference introduced by the transformation (such as interference above 1kHz attenuation ≥80%). During implementation, the matrix parameters of Clark and Park transformations need to be calibrated first to ensure the transformation accuracy. At the same time, the filter cutoff frequency is adjusted to balance the filtering effect and the dynamic response speed of the signal, avoiding the filter lag from affecting subsequent control.

[0051] Preferably, S4 includes the following sub-steps: S41, based on the mathematical model of the permanent magnet synchronous motor, construct the state equation of the full-order sliding mode observer, and substitute the stator inductance parameters and stator resistance parameters obtained in S2 into the state equation to determine the initial structural parameters of the observer; S42, design the input and output variables of the fuzzy logic controller, the input variables are the rotor position observation deviation and the speed observation deviation, and the output variable is the sliding mode gain adjustment amount of the sliding mode observer, and establish the fuzzy membership function and fuzzy inference rule base of the input variables; S43, input the voltage data in S1 and the harmonic current components obtained by decomposition in S3 into the full-order sliding mode observer, and calculate the rotor position observation value and speed observation value through the observer state equation; S44, input the rotor position observation deviation and speed observation deviation into the fuzzy logic controller, output the sliding mode gain adjustment amount according to the fuzzy inference rule base, dynamically adjust the sliding mode gain of the observer, and reduce the chattering phenomenon of the observer.

[0052] Specifically, the S4 fuzzy full-order sliding mode observer construction clarifies the implementation details of observer modeling, fuzzy controller design, state observation and gain adjustment through four sub-steps, improving the accuracy and stability of rotor position and speed observation, reducing sliding mode chattering, and providing reliable state feedback for harmonic compensation control. In S41, the state equation of a full-order sliding mode observer is constructed based on the mathematical model of the permanent magnet synchronous motor. The stator inductance parameters (error ±2%) obtained in S2 and the stator resistance (error ±1%) measured offline are substituted to set the initial structural parameters of the observer (e.g., the state matrix dimension matches the number of motor phases; a 3-phase motor corresponds to a 6th-order state matrix). This ensures that the state equation is consistent with the actual characteristics of the motor. Initial parameter debugging needs to be verified through offline simulation (e.g., observation error ≤ ±0.03π radians in MATLAB simulation). In S42, the input variables of the fuzzy logic controller are set as rotor position observation deviation (range -π to π radians) and speed observation deviation (range -50 rad / s to 50 rad / s). The output variable is the sliding mode gain adjustment (range 0.1-1.0). Five fuzzy subsets (negative large, negative small, zero, positive small, positive large) and 25 fuzzy inference rules are constructed (e.g., when both the position deviation and speed deviation are negative large, the gain adjustment is 0.9-1.0). The membership function adopts a triangular distribution to ensure the smoothness of the inference logic. In S43, the voltage data (instantaneous three-phase voltage values, accuracy ±0.5%) collected in S1 and the harmonic current components (error ±3%) decomposed in S3 are input into the observer. The rotor position and speed observation values ​​are calculated iteratively through the state equation. The iteration step size is set to 0.001 seconds, and the observation result update cycle is 0.1 seconds. In S44, the position and speed observation deviations are input into the fuzzy controller, and the sliding mode gain adjustment amount is output to dynamically adjust the observer gain. When the deviation is large (e.g., position deviation > π / 2 radians), the gain is increased (0.8-1.0) to speed up convergence. When the deviation is small (e.g., position deviation < π / 10 radians), the gain is decreased (0.1-0.3) to reduce chattering. Finally, the position observation error is controlled within ±0.05π radians, and the speed observation error is ±2 rad / s. During implementation, the fuzzy rules need to be calibrated offline first, and then the gain adjustment range needs to be optimized through online debugging to ensure the stability of the observation.

[0053] Preferably, S5 includes the following sub-steps: S51, setting the command values ​​of each harmonic current according to the motor operating conditions, wherein the command values ​​are dynamically adjusted according to the motor load and speed range to ensure that the command values ​​of each harmonic current match the command value of the fundamental current; S52, calculating the current deviation corresponding to each harmonic, wherein the current deviation is the difference between the command value of each harmonic current and the actual value of each harmonic current obtained by decomposition in S3, and performing polarity judgment and amplitude calculation on the current deviation; S53, building a PI controller in each synchronous rotating coordinate system, inputting the current deviation into the PI controller, and adjusting the current deviation through the proportional and integral links of the PI controller to generate the initial compensation control signal corresponding to each harmonic; S54, combining the rotor speed information observed in S4, performing speed feedforward compensation on the initial compensation control signal to eliminate the influence of speed changes on the compensation control signal, and obtaining the final compensation control signal for each harmonic.

[0054] Specifically, the S5 multi-channel harmonic compensation control signal generation process clearly defines the implementation flow through four steps: instruction setting, deviation calculation, PI adjustment, and feedforward compensation. This generates precise compensation control signals for each harmonic, ensuring targeted suppression of harmonic disturbances at each frequency. In S51, the instruction values ​​for each harmonic current are set according to the suppression target. The fundamental current instruction value matches the motor load requirements (e.g., the motor's rated current is used under rated load, with an error of ±2%). The instruction values ​​for the 3rd, 5th, and 7th harmonic currents are set to 0A, and these instruction values ​​need to be dynamically adjusted with load changes (e.g., when the load increases from 20% to 100% of the rated load, the fundamental instruction value increases synchronously and linearly). The adjustment response time is ≤0.01 seconds to ensure compatibility with the motor's operating conditions. In S52, the deviation of each harmonic current is calculated (the instruction value is decomposed from S3). The deviation calculation (actual harmonic current component difference) must ensure that the command value and the actual value are synchronized in time (time deviation ≤ 0.001 seconds). Simultaneously, the polarity of the deviation (positive deviation / negative deviation) and amplitude calculation (accuracy ±0.1A) must be performed. The deviation data update cycle must be consistent with the observer (0.1 seconds) to provide an accurate deviation signal for subsequent adjustment. In S53, an independent PI controller is configured for each harmonic channel. The proportional coefficient (Kpn) for the 3rd harmonic channel is set to 0.8-1.2, and the integral coefficient (Kin) is set to 0.05-0. 1. The 5th harmonic Kpn is set to 1.0-1.4, and Kin is set to 0.03-0.08; the 7th harmonic Kpn is set to 1.2-1.6, and Kin is set to 0.02-0.06. The PI regulation operation adopts a discretization algorithm (sampling period of 0.001 seconds) to generate the initial compensation control signal. The signal amplitude is limited to 5%-15% of the rated voltage (e.g., 19V-57V for a 380V motor) to avoid overloading the drive circuit due to excessive signal. In S54, the rotor speed information observed in S4 (error ±2r) is combined. (ad / s) speed feedforward compensation is performed, and the compensation coefficient is set according to the speed range (0.9-1.0 below 1000rpm, 1.0-1.1 between 1000-3000rpm, and 1.1-1.2 above 3000rpm) to eliminate the influence of speed changes on the compensation signal. The final compensation control signal error is ≤±3%. During implementation, it is necessary to optimize the PI parameters and feedforward compensation coefficient through a combination of offline simulation and online debugging to ensure that the compensation signal can accurately cancel harmonic components under different speed and load conditions.

[0055] The multi-synchronous rotating coordinate system harmonic suppression strategy in this invention is a precise separation and independent control technology for multiple harmonics in the mixed current signal of a permanent magnet synchronous motor. The core of this strategy is to construct rotating coordinate systems that match the frequencies of each harmonic, converting different frequency harmonic currents into DC quantities for targeted suppression. The implementation process requires combining previously acquired stator inductance parameters to first determine the harmonic orders to be suppressed (typically 3rd, 5th, and 7th), and then constructing the same number of rotating coordinate systems accordingly. The rotational angular velocity of each coordinate system is consistent with the angular velocity of the corresponding harmonic (e.g., the angular velocity of the 3rd harmonic coordinate system is 3 times that of the fundamental wave). Next, the acquired three-phase current signal is converted into a two-phase stationary signal using a Clarke transform, then mapped to each rotating coordinate system and subjected to a Park transform to decompose the direct-axis and quadrature-axis current components of each harmonic. Finally, an independent adjustment channel is configured for each coordinate system, generating a compensation control signal and synthesizing the overall control signal. This strategy overcomes the limitations of traditional overall control in accurately adapting to different harmonics, achieving independent suppression of each harmonic and avoiding insufficient or excessive suppression of some harmonics. Its significance lies in significantly improving the accuracy of harmonic suppression, reducing motor torque ripple and energy loss, ensuring stable operation of the motor under different operating conditions, and providing reliable technical support for high-requirement industrial drive scenarios.

[0056] The fuzzy full-order sliding mode observer of this invention is an observation device that integrates fuzzy logic and full-order sliding mode technology for sensorless acquisition of rotor position and speed of permanent magnet synchronous motors. Its core function is to reduce chattering and improve observation accuracy by dynamically adjusting the sliding mode gain. Its implementation requires first constructing the state equation of the full-order sliding mode observer based on the motor's mathematical model, and substituting parameters such as stator inductance and resistance to determine the initial structure. Then, a fuzzy logic module is designed, using rotor position observation deviation and speed observation deviation as inputs and sliding mode gain adjustment as output, establishing 25 fuzzy inference rules (input divided into 5 fuzzy subsets). Subsequently, voltage data and harmonic current components are input, and preliminary observation values ​​are calculated through the state equation. The sliding mode parameters are then dynamically optimized by combining the gain adjustment output from the fuzzy module. The observer replaces traditional sensors, accurately acquiring rotor state without additional hardware costs, providing real-time state feedback for harmonic compensation control, and simultaneously reducing the chattering problem of traditional sliding mode observers. Its significance lies in reducing system complexity and cost, improving reliability under harsh operating conditions, ensuring accurate phase and amplitude of harmonic control signals, further optimizing harmonic suppression effects, and promoting the practical application of sensorless motor control technology.

[0057] The high-frequency response-based inductance identification algorithm in this invention is a technology for real-time acquisition of the dynamic changes in stator inductance of a permanent magnet synchronous motor. Its core is to calculate inductance parameters by injecting a high-frequency excitation signal and utilizing the high-frequency response characteristics of the electrical circuit. Its implementation requires first setting the high-frequency excitation signal parameters (frequency 500Hz-1kHz, amplitude 5%-10% of rated voltage) to avoid interference with the fundamental signal; then injecting the excitation signal into the stator circuit, simultaneously acquiring voltage and current data before and after injection; subsequently analyzing the amplitude and phase changes of the current response, combining it with voltage data to calculate the real-time stator inductance, and updating the parameters every 0.5 seconds. Simultaneously, a temperature compensation model is used to correct the influence of temperature on the inductance. The algorithm addresses the problem that traditional offline inductance measurements cannot adapt to magnetic circuit saturation and temperature changes during motor operation, providing accurate dynamic parameters for harmonic decomposition in multi-synchronous rotating coordinate systems and the construction of fuzzy full-order sliding mode observers. Its significance lies in eliminating control errors caused by parameter mismatch, ensuring the stability and reliability of subsequent technical aspects, improving the adaptive capability of the entire harmonic suppression system, ensuring the motor maintains high efficiency under complex operating conditions, and extending the system's service life.

[0058] like Figure 2 As shown, a permanent magnet synchronous motor current harmonic disturbance suppression system includes: an electrical circuit phasor test signal acquisition unit, used to continuously acquire voltage and current signals during the operation of the permanent magnet synchronous motor, and output multiple sets of time-series voltage and current data; a stator inductance identification unit based on high-frequency response, connected to the electrical circuit phasor test signal acquisition unit, receiving the voltage and current data, and calculating and outputting stator inductance parameters by injecting high-frequency excitation signals and monitoring current response; a multi-synchronous rotating coordinate system harmonic current decomposition unit, connected to the electrical circuit phasor test signal acquisition unit and the high-frequency response stator inductance identification unit respectively, receiving the voltage data, current data and stator inductance parameters, constructing a multi-synchronous rotating coordinate system and decomposing and outputting each harmonic current component; and a fuzzy full-order sliding... The mode observation unit is connected to the electrical circuit phasor test signal acquisition unit, the stator inductance identification unit based on high-frequency response, and the multi-synchronous rotating coordinate system harmonic current decomposition unit, respectively. It receives the voltage data, stator inductance parameters, and harmonic current components, and observes and outputs rotor position and speed information. The multi-channel harmonic compensation control signal generation unit is connected to the multi-synchronous rotating coordinate system harmonic current decomposition unit and the fuzzy full-order sliding mode observation unit, respectively. It receives the harmonic current components, rotor position and speed information, and generates and outputs harmonic compensation control signals. The total control signal synthesis and drive unit is connected to the multi-channel harmonic compensation control signal generation unit. It receives the harmonic compensation control signals and synthesizes the total control signal, and outputs the total control signal to the drive circuit of the permanent magnet synchronous motor.

[0059] A method and system for suppressing current harmonic disturbances in a permanent magnet synchronous motor (PMSM) significantly improves the stability and energy conversion efficiency of the motor. It effectively reduces torque pulsation caused by current harmonics, avoids vibration and noise during motor operation, reduces core and copper losses, and minimizes unnecessary energy waste. Simultaneously, the method and system ensure long-term stable operation of the drive circuit, preventing harmonic impacts and damage to circuit components, extending the lifespan of the entire motor system. Furthermore, it is flexibly adaptable to various load fluctuations and speed variations, achieving precise and efficient suppression of multi-frequency current harmonic disturbances in light-load, heavy-load, low-speed, and high-speed operating scenarios, demonstrating strong applicability.

[0060] In the parameter acquisition and harmonic decomposition stages, addressing the issue that traditional inductance identification methods cannot acquire accurate parameters dynamically in real time, this approach employs an inductance identification algorithm based on high-frequency response. This algorithm injects a high-frequency excitation signal of a specific frequency into the electrical circuit, monitors the current change response in real time, and calculates the motor stator inductance parameters by combining the collected voltage and current data. This method dynamically captures changes in inductance parameters during motor operation, providing accurate parameter support for subsequent stages and avoiding harmonic suppression errors caused by fixed or lagging parameters. Simultaneously, a multi-synchronous rotating coordinate system harmonic suppression strategy built based on these inductance parameters can accurately map the collected current signal to rotating coordinate systems of different harmonic orders, decomposing each harmonic current component and laying the foundation for targeted suppression of harmonics at various frequencies.

[0061] At the observation and control level, addressing the shortcomings of commonly used observers—such as lack of dynamic adjustment capability, low observation accuracy, and susceptibility to chattering—a fuzzy full-order sliding mode observer is constructed. Stator inductance parameters, voltage data, and harmonic current components are used as inputs, and the sliding surface parameters are dynamically adjusted through fuzzy logic rules. This design effectively reduces chattering in the sliding mode observer, improves the observation accuracy of rotor position and speed information, and provides reliable state feedback for harmonic compensation control. Based on the observed information, each harmonic current regulation channel is independently controlled in a multi-synchronous rotating coordinate system, generating precise compensation control signals and synthesizing a total control signal. This significantly improves the control accuracy of each harmonic current regulation channel, effectively solving the problem of poor harmonic suppression in the prior art.

[0062] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "set," "install," "connect," "link," and "fix" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal communication between two components. Those skilled in the art will understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0063] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various equivalent changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for suppressing current harmonic disturbances in a permanent magnet synchronous motor, characterized in that, Includes the following steps: S1: Voltage and current signals obtained from phasor testing of the electrical circuit during the operation of the permanent magnet synchronous motor are acquired. These signals are continuously captured at a set sampling frequency to obtain multiple sets of time-series voltage and current data. S2: A high-frequency response-based inductance identification algorithm is used to process the voltage and current data acquired in S1. A high-frequency excitation signal of a preset frequency is injected into the electrical circuit, and the current change response under the corresponding high-frequency excitation signal is monitored. The stator inductance parameters of the motor are calculated by combining the voltage and current data. S3: Based on the stator inductance parameters obtained in S2, a coordinate system structure corresponding to the multi-synchronous rotating coordinate system harmonic suppression strategy is constructed. Rotating coordinate systems corresponding to different harmonic orders are set, and the current signals acquired in S1 are mapped to each of these rotating coordinate systems, decomposing them to obtain different harmonics. S4. Construct a fuzzy full-order sliding mode observer, using the stator inductance parameters obtained in S2 as one of the observer's input parameters. Simultaneously, input the voltage data from S1 and the harmonic current components obtained from S3 decomposition. Dynamically adjust the sliding mode parameters using the fuzzy logic rules within the observer to observe the rotor position and speed information of the motor. S5. Based on the rotor position and speed information observed in S4, and combined with the harmonic current components decomposed in S3, independently control the current adjustment channels corresponding to each harmonic in a multi-synchronous rotating coordinate system to generate compensation control signals corresponding to each harmonic. S6. Synthesize the harmonic compensation control signals generated in S5 to obtain a total control signal. Output the total control signal to the drive circuit of the permanent magnet synchronous motor to suppress motor current harmonic disturbances.

2. The method for suppressing current harmonic disturbances in a permanent magnet synchronous motor according to claim 1, characterized in that, In the S2 inductor identification algorithm based on high-frequency response, the formula for constructing the inductor identification model is as follows: Among them, L s U represents the stator inductance of the motor; hf Indicates the effective value of the high-frequency excitation voltage injected into the electrical circuit; R s I represents the stator resistance of the motor; hf ω represents the effective value of the high-frequency current in the electrical circuit under the action of a high-frequency excitation voltage; hf This represents the angular frequency of the high-frequency excitation signal; simultaneously, in the S3 multi-synchronous rotating coordinate system harmonic suppression strategy, the harmonic current decomposition model formula is constructed as follows: Among them, i n i represents the effective value of the nth harmonic current component; a i b i c These represent the instantaneous values ​​of the three-phase stator currents; n represents the harmonic order; and θ represents the electrical angle of the motor rotor.

3. The method for suppressing current harmonic disturbances in a permanent magnet synchronous motor according to claim 1, characterized in that, In the process of building the S4 fuzzy full-order sliding mode observer, the formula for constructing the dynamic adjustment model of the observer's sliding surface is as follows: Where s represents the sliding surface parameters; e ω This represents the deviation between the observed rotational speed and the actual rotational speed; k1 and k2 represent the coefficients for dynamic adjustment of the fuzzy logic rules. The derivative represents the speed deviation; simultaneously, in the S5 harmonic current regulation control, the formula for constructing the compensation control signal generation model is as follows: Among them, u n K represents the compensation control signal corresponding to the nth harmonic; pn K represents the proportionality coefficient. in Indicates the integral coefficient; Indicates the command value of the nth harmonic current; i n This represents the actual value of the nth harmonic current.

4. The method for suppressing current harmonic disturbances in a permanent magnet synchronous motor according to claim 1, characterized in that, In the S2 inductor identification algorithm based on high-frequency response, the formula for constructing the high-frequency response characteristic analysis model is as follows: Among them, G hf (s) represents the high-frequency response transfer function; I hf (s) represents the Laplace transform of the high-frequency current; U hf (s) represents the Laplace transform of the high-frequency voltage; R s L represents the stator resistance. s θ represents the stator inductance; s represents the complex frequency variable; in the construction of the S3 multi-synchronous rotating coordinate system, the formula for calculating the rotation angle of the coordinate system is as follows: n =ω0t+nδ, where θ n ω0 represents the rotation angle of the rotating coordinate system corresponding to the nth harmonic; t represents time; n represents the harmonic order; and δ represents the initial phase angle.

5. The method for suppressing current harmonic disturbances in a permanent magnet synchronous motor according to claim 1, characterized in that, In the S4 fuzzy full-order sliding mode observer, the formula for constructing the fuzzy inference model is as follows: Where Δk represents the sliding mode gain adjustment amount; e represents the rotor position observation deviation; The derivative of the rotor position observation deviation is represented by f(·); f(·) represents the fuzzy inference function; in the S5 compensation control signal synthesis, the formula for constructing the control signal synthesis model is as follows: Among them, u total Indicates the total control signal; N represents the highest harmonic order to be suppressed; u n θ represents the nth harmonic compensation control signal; n This represents the rotation angle of the rotating coordinate system for the nth harmonic.

6. The method for suppressing current harmonic disturbances in a permanent magnet synchronous motor according to claim 1, characterized in that, In the S2 inductor identification process, the inductor temperature compensation model formula is constructed as follows: L sT =L s [1+α(T-T0)], where L sT L represents the stator inductance after temperature compensation. s The stator inductance is represented at the reference temperature; α represents the inductance temperature coefficient; T represents the current motor temperature; T0 represents the reference temperature; the filter model formula for S3 harmonic current extraction is as follows: Among them, i nf τ represents the nth harmonic current component after filtering; n i represents the filtering time constant corresponding to the nth harmonic; n represents the nth harmonic current component before filtering; s represents the complex frequency variable.

7. The method for suppressing current harmonic disturbances in a permanent magnet synchronous motor according to claim 1, characterized in that, S3 includes the following sub-steps: S31, based on the stator inductance parameters obtained in S2, determine the number of multi-synchronous rotating coordinate systems and the harmonic order corresponding to each coordinate system, wherein the number of coordinate systems corresponds one-to-one with the harmonic order to be suppressed, and the rotation direction of each coordinate system is consistent with the rotation direction of the corresponding harmonic; S32, perform Clarke transform on the three-phase current signal collected in S1, converting the current signal in the three-phase stationary coordinate system into the current signal in the two-phase stationary coordinate system, eliminating the coupling components in the three-phase current signal; S33, calculate the rotation angle of each coordinate system according to the harmonic order corresponding to each rotating coordinate system, wherein the rotation angle is the product of the fundamental rotation angle and the harmonic order, and perform Park transform on the current signal in the two-phase stationary coordinate system to obtain the direct-axis current component and quadrature-axis current component in each rotating coordinate system; S34, perform filtering processing on the direct-axis current component and quadrature-axis current component in each rotating coordinate system to remove the high-frequency interference components introduced during the transformation process, and obtain the pure current component corresponding to each harmonic.

8. The method for suppressing current harmonic disturbances in a permanent magnet synchronous motor according to claim 1, characterized in that, S4 includes the following sub-steps: S41, based on the mathematical model of the permanent magnet synchronous motor, construct the state equation of the full-order sliding mode observer, substitute the stator inductance parameters and stator resistance parameters obtained in S2 into the state equation, and determine the initial structural parameters of the observer; S42, Design the input and output variables of the fuzzy logic controller. The input variables are the rotor position observation deviation and the speed observation deviation, and the output variable is the sliding mode gain adjustment of the sliding mode observer. Establish the fuzzy membership function and fuzzy inference rule base of the input variables. S43, Input the voltage data in S1 and the harmonic current components obtained by decomposition in S3 into the full-order sliding mode observer, and calculate the rotor position observation value and speed observation value through the observer state equation. S44 inputs the rotor position observation deviation and speed observation deviation into the fuzzy logic controller, outputs the sliding mode gain adjustment amount according to the fuzzy inference rule base, dynamically adjusts the sliding mode gain of the observer, and reduces the chattering phenomenon of the observer.

9. The method for suppressing current harmonic disturbances in a permanent magnet synchronous motor according to claim 1, characterized in that, S5 includes the following sub-steps: S51, setting the command values ​​of each harmonic current according to the motor operating conditions. The command values ​​are dynamically adjusted according to the motor load and speed range to ensure that the command values ​​of each harmonic current match the command value of the fundamental current; S52, calculating the current deviation corresponding to each harmonic. The current deviation is the difference between the command value of each harmonic current and the actual value of each harmonic current obtained from S3. The polarity of the current deviation is determined and the amplitude is calculated; S53, building a PI controller in each synchronous rotating coordinate system, inputting the current deviation into the PI controller, and adjusting the current deviation through the proportional and integral links of the PI controller to generate the initial compensation control signal corresponding to each harmonic; S54, combining the rotor speed information observed in S4, performing speed feedforward compensation on the initial compensation control signal to eliminate the influence of speed changes on the compensation control signal, and obtaining the final harmonic compensation control signal.

10. A current harmonic disturbance suppression system for a permanent magnet synchronous motor, characterized in that, include: The electrical circuit phasor test signal acquisition unit is used to continuously acquire voltage and current signals during the operation of the permanent magnet synchronous motor and output multiple sets of time-series voltage and current data. A stator inductance identification unit based on high-frequency response is connected to the electrical circuit phasor test signal acquisition unit, receives the voltage and current data, and calculates and outputs stator inductance parameters by injecting a high-frequency excitation signal and monitoring the current response. A multi-synchronous rotating coordinate system harmonic current decomposition unit is connected to the electrical circuit phasor test signal acquisition unit and the stator inductance identification unit based on high-frequency response, respectively, receives the voltage data, current data and stator inductance parameters, constructs a multi-synchronous rotating coordinate system and decomposes and outputs each harmonic current component. A fuzzy full-order sliding mode observation unit is connected to the electrical circuit phasor test signal acquisition unit, the stator inductance identification unit based on high-frequency response, and the multi-synchronous rotating coordinate system harmonic current decomposition unit, respectively. It receives the voltage data, stator inductance parameters, and harmonic current components, and observes and outputs rotor position and speed information. A multi-channel harmonic compensation control signal generation unit is connected to the multi-synchronous rotating coordinate system harmonic current decomposition unit and the fuzzy full-order sliding mode observation unit, respectively. It receives the harmonic current components, rotor position and speed information, and generates and outputs harmonic compensation control signals. A total control signal synthesis and driving unit is connected to the multi-channel harmonic compensation control signal generation unit. It receives the harmonic compensation control signals and synthesizes the total control signal, and outputs the total control signal to the drive circuit of the permanent magnet synchronous motor.

Images