A Harmonic Current Injection Active Disturbance Rejection Control Method for Motors

Abstract

This invention discloses a harmonic current injection active disturbance rejection control method for motors. It is a harmonic current suppression and tracking control scheme for permanent magnet synchronous motors based on an improved QRADRC. Strong disturbance suppression is achieved by embedding quasi-resonance in the ESO (Electronic Stability Oscillator), and high-precision tracking is achieved by embedding quasi-resonance in the control law. Combined with a dual-quasi-resonance coordinated tuning rule, harmonic current injection and control are completed directly within the dq synchronous rotating coordinate system without additional coordinate transformation. This invention is applicable to motor vector control systems based on the dq0 synchronous rotating coordinate system, and is particularly suitable for high-precision current control and active harmonic injection applications of surface-mounted permanent magnet synchronous motors under inverter nonlinearity, dead-zone effects, and parameter perturbations.

Description

Technical Field

[0001] This invention belongs to the field of motor control technology, specifically relating to a harmonic current injection active disturbance rejection control method for motors. Background Technology

[0002] Permanent magnet synchronous motors (PMSMs) offer advantages such as high efficiency, high power density, and high dynamic response, and are widely used in servo drives, new energy vehicles, industrial transmissions, and high-end equipment. However, in actual operation, nonlinear factors such as inverter dead-zone effect, power device conduction voltage drop, and motor magnetic circuit saturation can cause significant 6k-1 and 6k+1 harmonics in the stator three-phase current. In the dq synchronous rotating coordinate system, these manifest as 6k-fold harmonic currents, leading to torque pulsation, speed fluctuations, electromagnetic noise, and vibration, severely reducing system control accuracy and operational stability.

[0003] Traditional vector control uses PI (proportional-integral) regulators, which have limited ability to suppress periodic harmonic disturbances and struggle to balance dynamic response and harmonic suppression performance. Active disturbance rejection control (ADRC), on the other hand, does not rely on a precise mathematical model of the controlled object. It uses an extended state observer (ESO) to treat changes in internal system parameters, external disturbances, and unmodeled dynamics as a unified total disturbance for real-time observation and feedforward compensation, thus exhibiting strong robustness and excellent dynamic performance.

[0004] To enhance the suppression capability of harmonics at specific frequencies, existing technologies combine quasi-resonant controllers (QR) with active disturbance rejection control (ADRC) to form quasi-resonant active disturbance rejection control (QRADRC): by embedding a quasi-resonant element in the ESO, the ability to observe and compensate for 6 kHz harmonic disturbances is enhanced. Current academic research has applied QRADRC to suppress motor harmonic currents, mainly suppressing 6 kHz harmonic currents caused by inverter dead zones and motor nonlinearity, in order to improve the sinusoidal nature of phase currents, and has achieved good suppression effects, such as in the literature [Chen Zhe, Zhang Xuxuan, Liu Chunqiang. Research on current decoupling and harmonic suppression strategy of permanent magnet synchronous motor based on proportional resonant active disturbance rejection control [J]. Proceedings of the CSEE, 2022, 42(24): 9062-9072] and the literature [Xu Jiaqun, Wang Tianqi, Jia Pufan. Harmonic suppression of quasi-resonant active disturbance rejection current of permanent magnet synchronous motor [J]. Proceedings of the CSEE, 2023, 43(06): 2450-2460].

[0005] However, the existing QRADRC and related improved structures have obvious technical defects:

[0006] 1. Traditional QRADRC only has the ability to suppress harmonic disturbances and cannot accurately track harmonic currents of specified amplitude and phase, making it difficult to meet the application requirements of active harmonic injection to suppress torque pulsation and vibration noise.

[0007] 2. When using a parallel structure of QR and ADRC, harmonic tracking can be achieved, but other higher harmonics will be introduced, and the disturbance suppression capability will be significantly reduced.

[0008] 3. When QR and QRADRC are directly connected in parallel, ESO will treat the QR output command as a complete disturbance compensation, resulting in the failure of harmonic tracking capability.

[0009] It is evident that existing technologies cannot simultaneously achieve strong harmonic disturbance suppression and high-precision harmonic current tracking in the same controller, thus limiting their practical application scenarios. Summary of the Invention

[0010] To address the technical bottleneck of existing QRADRC structures in simultaneously achieving harmonic suppression and harmonic tracking, this invention provides a harmonic current injection active disturbance rejection control method for motors (denoted as QR). 2 ADRC has the ability to suppress the sixth harmonic disturbance caused by inverter dead zone and nonlinearity, as well as the ability to actively inject and track the sixth harmonic current with specified amplitude and phase. It also retains the inherent strong robustness, fast dynamic response and low model dependence of ADRC.

[0011] A method for harmonic current injection active disturbance rejection control for electric motors includes the following steps:

[0012] (1) Establish the first-order ADRC model of the dq-axis current of the permanent magnet synchronous motor and its state equations;

[0013] (2) Design a quasi-resonant linear ESO based on the state equation, that is, introduce a quasi-resonant element into the observation gain of the traditional linear ESO to generate the observation values ​​of current and total disturbance.

[0014] (3) The traditional ADRC control law is improved into a control law with quasi-resonant gain by adopting the quasi-resonant parameter coordinated tuning rule. The observed values ​​of current and total disturbance are input into the control law to generate voltage reference values ​​and apply them to the motor control system to realize harmonic current injection and active disturbance rejection control.

[0015] Furthermore, the first-order ADRC model expression for the dq-axis current of the permanent magnet synchronous motor in step (1) is as follows:

[0016] ;

[0017] in: and These are the d-axis current and q-axis current of the motor, respectively. and These are the d-axis voltage and q-axis voltage of the motor, respectively. and These are the d-axis inductance and q-axis inductance of the motor, respectively. and These are the d-axis fundamental flux linkage and the q-axis fundamental flux linkage of the motor, respectively. and These represent the total disturbance along the d-axis and the total disturbance along the q-axis, respectively. and These represent unmodeled perturbations along the d-axis and q-axis, respectively. This is the stator resistance of the motor. The electric angular velocity of the motor. t Indicates time.

[0018] Furthermore, the expression for the state equation in step (1) is as follows:

[0019]

[0020] in: and For state variables, u To control variables, This is the critical gain. G For the perturbation derivative, y The output of the state equation. and They are respectively and The first derivative; for d-axis state observation, , , u , They correspond to as , , , For q-axis state observation, , , u , They correspond to as , , , .

[0021] Furthermore, the equation expression for the quasi-resonant linear ESO in step (2) is as follows:

[0022]

[0023] in: and State variables and The observed values, For output quantity y The observed values, and They are respectively and The first derivative, and For observer gain, The gain transfer function, For QR transfer function, For QR resonant gain, For QR bandwidth, ω g The harmonic frequency to be suppressed s For the Laplace operator.

[0024] Furthermore, the control law with quasi-resonant gain in step (3) has the following equation expression:

[0025]

[0026] in: For ADRC gain, The gain transfer function, For d-axis voltage reference value or q-axis voltage reference value, the corresponding This is the reference value for either the d-axis or q-axis current. For QR transfer function, For QR resonant gain, For QR bandwidth, The harmonic frequency to be tracked.

[0027] Furthermore, the quasi-resonance parameter co-tuning rule in step (3) is as follows: The value is 0.02. The value ranges from 10 to 1000. , .

[0028] A computer device includes a memory and a processor, wherein the memory stores a computer program and the processor executes the computer program to implement the above-described harmonic current injection active disturbance rejection control method for an electric motor.

[0029] A computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-described harmonic current injection active disturbance rejection control method for an electric motor.

[0030] Based on the above technical solution, the present invention has the following beneficial technical effects:

[0031] 1. This invention QR 2 The ADRC controller simultaneously achieves harmonic suppression and harmonic tracking, overcoming the limitation of traditional QRADRC which can only suppress but not track harmonics.

[0032] 2. This invention uses a quasi-resonant element to directly achieve harmonic suppression and tracking in the dq synchronous rotating coordinate system, without the need for additional coordinate transformation or complex harmonic decomposition and reconstruction calculations. The computational load is small when implemented digitally, and it is easy to run in real time on embedded platforms such as MCU (microcontroller unit) and DSP (digital signal processor).

[0033] 3. In terms of disturbance suppression, this invention thoroughly suppresses the 6k harmonics in the dq axis, significantly reduces THD (total harmonic distortion), and introduces almost no other higher harmonics.

[0034] 4. The present invention has high tracking accuracy. When a set of optimal gain and bandwidth is selected, the harmonic amplitude error is <5% and the phase error is <1° while ensuring fast convergence. It can be directly used for active harmonic injection.

[0035] 5. This invention retains the advantages of ADRC in resisting parameter perturbations and load mutations, and has strong robustness; at the same time, ADRC does not rely on an accurate model, and there is no need to design amplitude and phase compensation links for frequency when harmonic current is injected, and the overall structure is simple. Attached Figure Description

[0036] Figure 1 This invention QR 2 ADRC current loop control flowchart. Figure 1 middle The purpose of current loop control is to make the harmonic components in the actual current track this given harmonic component, which is the desired dq-axis current harmonic component.

[0037] Figure 2 This is a control flow diagram for linearized ESO in active disturbance rejection control.

[0038] Figure 3 This invention QR 2 A schematic diagram of the amplitude-frequency response curve of the ADRC current loop-motor control system.

[0039] Figure 4 This is a schematic diagram of the amplitude-frequency response curve of a traditional QRADRC current loop-motor system.

[0040] Figure 5 This invention QR 2 A schematic diagram illustrating the suppression effect of ADRC on the 6th harmonic current.

[0041] Figure 6 This is a schematic diagram of the process for obtaining the target harmonic current in an embodiment of the present invention. Detailed Implementation

[0042] To describe the present invention in more detail, the technical solution of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0043] This embodiment provides a permanent magnet synchronous motor QR 2 The ADRC harmonic suppression and tracking control method, specifically implemented through the following steps:

[0044] (1) Establish a first-order ADRC model of the dq-axis current of the permanent magnet synchronous motor, and unify the stator resistance, inductance change, cross-coupling terms and unmodeled dynamics as the total disturbance; taking the q-axis current as an example, the motor current equation is rearranged into the first-order standard form:

[0045]

[0046] Select State variables x 1, State variables x 2. The state equation can be obtained:

[0047]

[0048] In the formula: For unmodeled perturbations, For the critical gain, assume the total disturbance... The derivative is bounded and is denoted as . G .

[0049] (2) Based on the existing QRADRC structure, an independent quasi-resonant element is embedded in the control law to form a QR 2 The ADRC control structure enables the controller to simultaneously possess harmonic disturbance suppression capability and the ability to track harmonic currents at a specified frequency.

[0050] This invention designs a quasi-resonant linear ESO (denoted as QRESO), introducing a quasi-resonant element into the observation gain of a traditional linear ESO. G r ( s This improves the observation accuracy of 6th harmonic disturbances. The equation expression for QRESO is as follows:

[0051]

[0052] In the formula: and They are respectively and The observed values, for y The observed values, and This is the observer gain, the specific value of which is calculated using the observer bandwidth. For QR resonant gain, For QR bandwidth, ωg It is the resonant frequency.

[0053] (3) The quasi-resonant parameter coordinated tuning rule is adopted to achieve the suppression and tracking of the 6th harmonic.

[0054] This invention designs a QR code. 2 The ADRC control law improves the traditional ADRC control law into a control law with quasi-resonant gain, achieving harmonic current tracking without steady-state error; the equation of this control law is as follows:

[0055]

[0056] In the formula: This refers to the first-order ADRC controller gain (or controller bandwidth).

[0057] The rules for coordinated tuning of quasi-resonant parameters include: using a small resonant gain for the quasi-resonance within the extended state observer to ensure system dynamic stability and disturbance suppression; using a moderate resonant gain for the quasi-resonance within the control law to ensure harmonic tracking accuracy; and ensuring that the quasi-resonant bandwidth within the extended state observer is greater than that within the control law to avoid resonant coupling. For the QR within the ESO, a small resonant gain is used ( K r =0.02), providing strong disturbance suppression capability while maintaining control system stability; for QR within the ADRC control law, a larger resonant gain is used ( To ensure harmonic tracking accuracy, it is also recommended that the QR bandwidth within the ESO be increased. ω c The bandwidth greater than the QR within the control law This avoids resonance conflicts and improves convergence speed.

[0058] The following example illustrates the suppression and injection of 6th harmonic current disturbances generated by a permanent magnet synchronous motor under conditions such as inverter dead zone, using the QR method of this invention. 2 The ADRC control method achieves the functions of suppressing disturbances of the 6th harmonic and tracking the target amplitude and phase. The controlled object is a surface-mounted permanent magnet synchronous motor. Control, current loop adopts as follows Figure 1 The QR code shown 2 The ADRC controller uses a conventional first-order ADRC controller for the speed loop.

[0059] The system has disturbances such as inverter dead zone and power device voltage drop. The motor stator current mainly contains 6k-1 and 6k+1 harmonics, which are uniformly represented as 6k harmonic components in the dq synchronous rotating coordinate system. These are the main components causing torque pulsation and speed fluctuation, with the 6th harmonic current having the most significant impact.

[0060] The specific steps for injecting and suppressing the 6th harmonic in this embodiment are as follows:

[0061] Step S1: Embed a quasi-resonant element in the extended state observer (ESO) to accurately observe and compensate for the total disturbance of the 6th harmonic frequency. At the same time, embed an independent quasi-resonant element in the control law gain to achieve zero steady-state error tracking of the 6th harmonic command current.

[0062] Step S2: The resonant frequencies of both quasi-resonant elements are set to 6 times the electric angular velocity. ω g =6 ω e To match the main harmonic frequencies of the system, targeted suppression and tracking are achieved.

[0063] Step S3: [The text appears to be incomplete and contains several grammatical errors. A more accurate translation would require the full context.] Figure 1 The motor in this example is considered as a first-order inertial element, and its transfer function is: G motor Then according to Figure 1 and Figure 2 From the control flow, we can obtain the amplitude-frequency response curve of the current loop-motor system, such as... Figure 3 As shown.

[0064]

[0065] Figure 4 The amplitude-frequency response curve of the traditional QRADRC is given, and compared with... Figure 3 and Figure 4 The present invention QR 2 While retaining the ability to suppress disturbances at the resonant frequency, the ADRC controller increases the bandwidth and gain of the input-output function at the resonant frequency, enabling it to simultaneously suppress and track harmonic currents near the resonant frequency.

[0066] Step S4: Directly inject the 6th harmonic command into the q-axis current setting:

[0067]

[0068] in: This is the DC reference value for the q-axis (output of the speed loop or manually set). For the target harmonic current amplitude, The phase of the target harmonic current.

[0069] Step S5: For the d-axis current, use the DC target value output from the pre-speed loop as the input to fully utilize QR. 2 ADRC's harmonic disturbance suppression capability ensures i d For control precision of =0, the d-axis reference current is as follows:

[0070]

[0071] Step S6: Obtain the motor mechanical speed through encoder and current sampling. ω m Instantaneous values ​​of three-phase current i a , i b , i c After coordinate transformation, we obtain i d and i q For the extreme logarithm is P The motor has an electrical angular velocity of ω e This is the fundamental frequency of the current.

[0072] When it is necessary to actively suppress 6 in the dq axis ω e When dealing with harmonic currents at a certain frequency, the current controller uses the QR of this invention. 2 The ADRC control structure has a resonant input frequency of 6 for QR. ω e By actively adding a harmonic current reference signal to the given dq-axis DC target current, the current controller can adaptively track the dq-axis reference current containing harmonic current throughout the entire motor operation process, and can actively observe and compensate for the 6% harmonic current caused by inverter dead time, etc. ω e Frequency perturbations result in higher tracking accuracy.

[0073] Design according to the above steps and use Simulink simulation. In the simulation model, the inverter switching frequency is 20kHz and the dead time is 10μs. The actual harmonic current tracking and suppression waveforms are as follows. Figure 5 As shown in the figure, the horizontal axis represents the captured simulation time, and the vertical axis represents the harmonic current amplitude. The expected values ​​of d6 and q6 are the expected sixth harmonic current waveforms given on the d-axis and q-axis of the dq0 coordinate system, respectively. The actual values ​​of d6 and q6 are the actual sixth harmonic current waveforms on the d-axis and q-axis of the dq0 coordinate system. It can be seen that in the QR of this invention... 2 Under ADRC control, the actual harmonic current can accurately track the given harmonic current. At the same time, for the 6th harmonic current of the d-axis with a given value of 0, it can strongly suppress harmonic disturbances.

[0074] When injecting and tracking harmonic currents in a motor, it is often necessary to distinguish the positive and negative sequences of the harmonic currents. This invention can also be used for injecting harmonic currents of specific positive and negative sequences. Taking the injection of a negative-sequence 5th harmonic current and a positive-sequence 7th harmonic current in a three-phase coordinate system as an example... Figure 6 As shown in the figure i d5dc and i q5dc These are the d-axis DC component and q-axis DC component of the fifth harmonic current to be injected in the fifth synchronous rotating coordinate system, respectively. i d7dc and i q7dc These represent the d-axis DC component and q-axis DC component of the 7th harmonic current to be injected in the 7th synchronous rotating coordinate system, respectively. d5q5-dq and d7q7-dq represent the transformations from the 5th and 7th synchronous rotating coordinate systems to the dq0 coordinate system, respectively. After the transformation, the negative sequence harmonic current in the dq0 coordinate system is obtained. i dh -and i qh -and positive sequence harmonic currents i dh + and i qh +, if they are combined into a single vector, that is Figure 1 In Ultimately, QR code was used. 2 ADRC can inject positive and negative sequence harmonic currents.

[0075] In summary, the present invention QR 2 The ADRC control structure can achieve precise tracking directly in the dq coordinate system without additional coordinate transformations, harmonic positive and negative sequence separation, or multiple rotation transformations. It has low computational complexity, high real-time performance, and can quickly and accurately output the 6th harmonic current with the desired amplitude and phase. QR 2 ADRC simultaneously possesses the capabilities of harmonic current tracking and harmonic disturbance suppression. When the target harmonic current is given to be 0 (primarily performing suppression), compared to the traditional QRADRC, QR... 2 ADRC has stronger harmonic suppression capabilities and does not cause a significant increase in harmonic currents at other frequencies.

[0076] The core of this invention lies in the dual-quasi-resonant synergistic structure of ESO and control law, and the integrated design concept of disturbance suppression and command tracking. For those skilled in the art, simply increasing the number of parallel QRs and setting different resonant frequencies to achieve tracking or suppression of multiple frequency points is a direct replacement of conventional techniques and does not change the core control architecture and synergistic mechanism of this invention. Therefore, it cannot be considered an innovative contribution of this invention, nor does it exceed the scope of protection of this invention.

Claims

1. A method for harmonic current injection active disturbance rejection control of an electric motor, characterized in that, Includes the following steps: (1) Establish the first-order ADRC model of the dq-axis current of the permanent magnet synchronous motor and its state equation; the expression of the first-order ADRC model of the dq-axis current of the permanent magnet synchronous motor is as follows: in: and These are the d-axis current and q-axis current of the motor, respectively. and These are the d-axis voltage and q-axis voltage of the motor, respectively. and These are the d-axis inductance and q-axis inductance of the motor, respectively. and These are the d-axis fundamental flux linkage and the q-axis fundamental flux linkage of the motor, respectively. and These represent the total disturbance along the d-axis and the total disturbance along the q-axis, respectively. and These represent unmodeled perturbations along the d-axis and q-axis, respectively. This is the stator resistance of the motor. The electric angular velocity of the motor. t Indicates time; (2) Design a quasi-resonant linear ESO based on the state equation, that is, introduce a quasi-resonant element into the observation gain of a traditional linear ESO to generate observations of current and total disturbance; the equation expression of the quasi-resonant linear ESO is as follows: in: and State variables and The observed values, y The output of the state equation. For output quantity y The observed values, and They are respectively and The first derivative, and For observer gain, The gain transfer function, For QR transfer function, For QR resonant gain, For QR bandwidth, The harmonic frequency to be suppressed s For the Laplace operator, This is the critical gain. u For control variables; (3) The traditional ADRC control law is improved into a control law with quasi-resonant gain by adopting the quasi-resonant parameter cooperative tuning rule. The observed values ​​of current and total disturbance are input into the control law to generate voltage reference values ​​and apply them to the motor control system to realize harmonic current injection and active disturbance rejection control. The control law with quasi-resonant gain is expressed by the following equation: in: For ADRC gain, The gain transfer function, For d-axis voltage reference value or q-axis voltage reference value, the corresponding This is the reference value for either the d-axis or q-axis current. For QR transfer function, For QR resonant gain, For QR bandwidth, The harmonic frequency to be tracked.

2. The method for harmonic current injection active disturbance rejection control for motors according to claim 1, characterized in that, The expression for the state equation in step (1) is as follows: in: and For state variables, G For the perturbation derivative, and They are respectively and The first derivative; for d-axis state observation, , , u , They correspond to as , , , For q-axis state observation, , , u , They correspond to as , , , .

3. The harmonic current injection active disturbance rejection control method for motors according to claim 1, characterized in that, The quasi-resonance parameter co-tuning rule in step (3) is as follows: The value is 0.

02. The value ranges from 10 to 1000. , .

4. A computer device, comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: The processor is used to execute the computer program to implement the harmonic current injection active disturbance rejection control method for motors as described in any one of claims 1 to 3.

5. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by the processor, it implements the harmonic current injection active disturbance rejection control method for motors as described in any one of claims 1 to 3.

Images