A speed fluctuation suppression disturbance control method for a low-speed permanent magnet direct drive motor

Abstract

The application belongs to the field of motor control, and discloses a speed fluctuation suppression anti-disturbance control method for a low-speed permanent magnet direct drive motor. A first-order linear active disturbance rejection controller is designed in a speed loop, an improved LESO is designed by constructing a state observation error differential term and a coefficient thereof, an 'integral+differential' parallel structure is constructed in a signal transmission path of an estimated speed loop total disturbance, a phase leading characteristic of the differential term is utilized to improve a medium-high frequency disturbance tracking speed, a second-order IIR Butterworth high-pass filter is combined, high frequency noise introduced due to an increase of an observer bandwidth can be accurately separated and filtered out, an inherent contradiction between bandwidth improvement and noise amplification in a traditional LADRC is broken through, and the speed fluctuation suppression rate under a low-speed working condition is improved. A model-free predictive control is adopted in a current loop, the dependence of the system on accurate modeling of motor parameters is reduced, current tracking errors generated under a parameter mismatch condition are reduced, torque pulsation caused by current fluctuation is reduced, and then speed fluctuation is reduced.

Description

Technical Field

[0001] This invention belongs to the field of motor control technology, and specifically relates to a speed fluctuation suppression and disturbance rejection control method for low-speed permanent magnet direct drive motors, which is particularly suitable for application scenarios of speed fluctuation suppression and disturbance rejection control for low-speed permanent magnet direct drive motors. Background Technology

[0002] Permanent magnet direct drive motors (PMDs) have been widely used in industry due to their advantages such as high power density and fast dynamic response. However, PMDs face significant periodic torque pulsations (such as cogging effect and current harmonics) and parameter sensitivity issues when operating at low speeds, resulting in severe fluctuations in the motor's output torque. These periodic torque disturbances act directly on the motor, easily causing drastic speed fluctuations under low-speed conditions, which severely restricts the system's high-precision speed tracking performance and stable operation capability.

[0003] Currently, the commonly used dual-loop control method based on Field-Oriented Control (FOC) relies on PI controllers for both speed and current loops, making its performance highly sensitive to the accuracy of motor parameters. The inherent nonlinearity and strong coupling characteristics of permanent magnet direct-drive motors make parameter mismatch and external load disturbances unavoidable. At low speeds, even small torque ripples can be significantly amplified into speed fluctuations. The limited disturbance rejection capability of PI controllers makes it difficult to effectively suppress such periodic disturbances, causing the actual speed to deviate from the setpoint, resulting in steady-state speed ripples and dynamic tracking lag, becoming a bottleneck for achieving high-precision, low-ripple, low-speed operation. To overcome the limitations of PI controllers in disturbance rejection and improve the system's robustness to changes in internal parameters and external disturbances, Active Disturbance Rejection Control (ADRC) has been introduced into the field of motor control.

[0004] The core idea of ​​ADRC is to use an Extended State Observer (ESO) to treat the total internal and external disturbances of the system (including unmodeled dynamics, parameter changes, external disturbances, etc.) as an extended state for real-time estimation and to dynamically compensate for them in the control law, thereby significantly reducing the dependence on an accurate motor model. Linear Active Disturbance Rejection Control (LADRC) simplifies the structure of traditional ADRC by combining a Linear Extended State Observer (LESO) with a linear feedback structure (such as proportional error feedback), which not only improves the disturbance tracking speed but also reduces the difficulty of parameter tuning.

[0005] However, when the existing LADRC is applied to the speed loop control of permanent magnet direct drive motors, it faces an inherent contradiction in dealing with periodic high-frequency torque pulsations under low-speed conditions: to enhance disturbance estimation and compensation capabilities, the ESO bandwidth needs to be increased, but increasing the bandwidth will amplify high-frequency noise (such as encoder noise), resulting in significantly enhanced control signal jitter and even affecting system stability; conversely, reducing the ESO bandwidth to reduce noise impact will introduce significant phase delay and amplitude attenuation, making it impossible for the ESO to effectively capture and compensate for rapidly changing disturbance signals, thus causing the speed fluctuations caused by periodic disturbances to persist, exacerbating steady-state speed fluctuations, and slowing dynamic tracking response. Summary of the Invention

[0006] The purpose of this invention is to propose a speed fluctuation suppression and disturbance rejection control method for low-speed permanent magnet direct drive motors. This method adopts a first-order linear active disturbance rejection controller. By designing an improved LESO and combining it with a second-order IIR Butterworth high-pass filter, it can accurately separate and filter out high-frequency noise introduced by the increased observer bandwidth. This helps to overcome the inherent contradiction between bandwidth improvement and noise amplification in traditional LADRC, and ultimately achieves a significant improvement in speed fluctuation suppression rate under low-speed conditions.

[0007] To achieve the above objectives, the present invention adopts the following technical solution:

[0008] A method for suppressing and controlling speed fluctuations in a low-speed permanent magnet direct drive motor includes the following steps:

[0009] Step 1. Real-time acquisition of the three-phase current of the motor, and conversion of the sampled current into current values ​​in the α-β coordinate system through Clark transformation; at the same time, real-time acquisition of the motor rotor position and actual speed;

[0010] Step 2. Establish a first-order linear active disturbance rejection controller based on an improved LESO and a second-order IIR Butterworth high-pass filter. The improved LESO is designed by constructing the differential term of the state estimation error and its gain coefficient. An "integral + derivative" parallel structure is constructed in the signal transmission path of the estimated total disturbance of the speed loop. The phase lead characteristic of the derivative operation is used to compensate for the phase delay of the integral link, so as to improve the tracking speed of the observer for mid-to-high frequency disturbances. At the same time, the frequency response of the disturbance estimation transfer function is optimized by adjusting the derivative gain coefficient.

[0011] The processing flow of the first-order linear active disturbance rejection controller is as follows: The actual rotational speed and the reference current of the q-axis at the previous moment are input into the improved LESO to obtain the estimated values ​​of the rotational speed and the total disturbance of the rotational loop; The estimated value of the rotational speed is input into the second-order IIR Butterworth high-pass filter to obtain the estimated value of the rotational speed after high-frequency noise compensation; Then, the estimated value of the rotational speed and the total disturbance of the rotational loop after high-frequency noise compensation are substituted into the error feedback control law to obtain the reference current of the q-axis at the current moment.

[0012] Step 3. Adopt a control strategy where the d-axis reference current is 0, use the error between the reference current and the sampled current as the control input, and generate the required control voltage by calculating the reference voltage;

[0013] Step 4. Input the calculated control voltage into the space vector pulse width modulation module to obtain the inverter's drive pulse signal, complete the motor drive, and realize the speed fluctuation suppression and disturbance rejection control of the low-speed permanent magnet direct drive motor.

[0014] Furthermore, based on the speed fluctuation suppression and disturbance rejection control method for low-speed permanent magnet direct drive motors, this invention also proposes a corresponding speed fluctuation suppression and disturbance rejection control system for low-speed permanent magnet direct drive motors, the technical solution of which is as follows:

[0015] A speed fluctuation suppression and disturbance rejection control system for low-speed permanent magnet direct drive motors includes the following modules:

[0016] The sensor acquisition module is used to acquire the three-phase current of the motor in real time and convert the sampled current into the current value in the α-β coordinate system through Clark transformation; at the same time, it can acquire the position and actual speed of the motor rotor in real time.

[0017] The active disturbance rejection control module is used to establish a first-order linear active disturbance rejection controller based on an improved LESO and a second-order IIR Butterworth high-pass filter. The improved LESO is designed by constructing the differential term of the state estimation error and its gain coefficient. An "integral + derivative" parallel structure is constructed in the signal transmission path of the estimated total disturbance of the speed loop. The phase lead characteristic of the derivative operation is used to compensate for the phase delay of the integral link, so as to improve the tracking speed of the observer for mid-to-high frequency disturbances. At the same time, the frequency response of the disturbance estimation transfer function is optimized by adjusting the derivative gain coefficient.

[0018] The processing flow of the first-order linear active disturbance rejection controller is as follows: The actual rotational speed and the reference current of the q-axis at the previous moment are input into the improved LESO to obtain the estimated values ​​of the rotational speed and the total disturbance of the rotational loop; The estimated value of the rotational speed is input into the second-order IIR Butterworth high-pass filter to obtain the estimated value of the rotational speed after high-frequency noise compensation; Then, the estimated value of the rotational speed and the total disturbance of the rotational loop after high-frequency noise compensation are substituted into the error feedback control law to obtain the reference current of the q-axis at the current moment.

[0019] The voltage prediction module is used to generate the required control voltage by using a control strategy where the d-axis reference current is 0, taking the error between the reference current and the sampled current as the control input, and calculating the reference voltage.

[0020] The system also includes a drive module, which inputs the calculated control voltage into the space vector pulse width modulation module to obtain the drive pulse signal for the inverter, thereby completing the motor drive and achieving speed fluctuation suppression and disturbance rejection control for the low-speed permanent magnet direct drive motor.

[0021] Furthermore, based on the aforementioned speed ripple suppression and disturbance rejection control method for low-speed permanent magnet direct drive motors, this invention also proposes a computer device comprising a memory and one or more processors. Executable code is stored in the memory. When the processor executes the executable code, it implements the steps of the aforementioned speed ripple suppression and disturbance rejection control method for low-speed permanent magnet direct drive motors.

[0022] Furthermore, based on the aforementioned speed fluctuation suppression and disturbance rejection control method for low-speed permanent magnet direct drive motors, this invention also proposes a computer-readable storage medium storing a program thereon. When executed by a processor, this program is used to implement the steps of the aforementioned speed fluctuation suppression and disturbance rejection control method for low-speed permanent magnet direct drive motors.

[0023] The present invention has the following advantages:

[0024] 1. This invention designs a first-order linear active disturbance rejection controller (ADC) in the speed loop to replace the traditional PI controller. Addressing the issue of phase lag and bandwidth-noise contradiction in traditional LESO disturbance estimation during low-speed operation of permanent magnet synchronous motors, an improved LESO is designed by constructing a differential term for the state estimation error and its gain coefficient. A parallel "integral + derivative" structure is built in the signal transmission path for estimating the total disturbance in the speed loop. The phase lead characteristic of the derivative operation effectively compensates for the phase delay of the integral stage, significantly improving the tracking speed of LESO for mid-to-high frequency disturbances. By adjusting the derivative gain coefficient, the frequency response of the disturbance estimation transfer function is optimized. When the derivative gain coefficient increases, the zero point of the transfer function shifts to higher frequencies, improving the gain in the mid-to-high frequency range while avoiding excessive amplification of high-frequency noise. This improvement allows the improved LESO to significantly increase the estimation speed for mid-to-high frequency disturbances such as periodic torque pulsations while retaining its ability to suppress high-frequency noise.

[0025] 2. This invention addresses the trade-off between bandwidth enhancement and noise amplification in the speed loop of a first-order linear active disturbance rejection controller (ADRC). A second-order IIR (Infinite Impulse Response) Butterworth high-pass filter is designed in the error feedback control law. Compared to a first-order filter, this filter has a steeper frequency roll-off (40dB / dec) and a flatter passband response, enabling faster and more accurate extraction of high-frequency noise from the motor speed estimation signal while avoiding spectral aliasing caused by an excessively wide transition band. Including the high-frequency noise extracted after high-pass filtering of the speed estimate using the second-order IIR Butterworth high-pass filter in the feedback loop helps to effectively suppress the amplification effect of high-frequency noise while increasing the control bandwidth. This further optimizes the estimation and disturbance rejection performance of the first-order linear ADRC while improving the system bandwidth.

[0026] 3. This invention designs a model-free predictive controller in the current loop to replace the traditional PI controller, which significantly reduces the system's dependence on accurate modeling of motor parameters (such as inductance and resistance), reduces current tracking errors caused by parameter mismatch, enhances the system's parameter robustness, and adopts a one-step prediction mechanism to predict the output direction based on the current error and input trend, correcting deviations in advance and reducing the passive characteristics of compensation by error accumulation. This improves current tracking accuracy and dynamic response speed, effectively reduces torque ripple caused by current fluctuations, and thus reduces speed fluctuations. Attached Figure Description

[0027] Figure 1 This is a block diagram of a speed fluctuation suppression and disturbance rejection control method for a low-speed permanent magnet direct drive motor in an embodiment of the present invention.

[0028] Figure 2 This is a structural diagram of the improved linear extended state observer in an embodiment of the present invention;

[0029] Figure 3 The Bode plot of the second-order IIR Butterworth high-pass filter in this embodiment of the invention; Figure 3 In the figure, (a) is the amplitude frequency response curve and (b) is the phase frequency response curve;

[0030] Figure 4 This is a structural diagram of the current loop expansion state observer in a model-free predictive controller according to an embodiment of the present invention;

[0031] Figure 5 The diagram shows a comparison of simulated speed fluctuations between the control method of this invention and the method without the improved LESO; where (a) uses the improved LESO of this invention, and (b) uses the traditional linear extended state observer.

[0032] Figure 6The figure shows a comparison of simulated speed fluctuations between the control method of the present invention and the method without a second-order IIR Butterworth high-pass filter; where (a) uses the second-order IIR Butterworth high-pass filter of the present invention; and (b) uses a traditional first-order high-pass filter.

[0033] Figure 7 The diagram shows a comparison of simulated speed fluctuations when the control method of the present invention uses a current loop model-free predictive controller; where (a) uses the current loop model-free predictive controller of the present invention; and (b) uses a traditional PI controller.

[0034] Figure 8 The diagram shows a comparison of simulated speed fluctuations between the control method of the present invention and the traditional PI control; where (a) is a schematic diagram of simulated speed fluctuations using the control method of the present invention; and (b) is a schematic diagram of simulated speed fluctuations using the traditional PI control method. Detailed Implementation

[0035] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments:

[0036] Example 1

[0037] like Figure 1 As shown in the figure, this embodiment describes a speed fluctuation suppression and disturbance rejection control method for a low-speed permanent magnet direct drive motor. The speed fluctuation suppression and disturbance rejection control method includes the following steps:

[0038] Step 1. Real-time acquisition of the three-phase current i of the motor using a current sensor. a i b i c And the sampled current i is converted through Clark transformation. a i b i c Convert the current value i to the α-β coordinate system α i β Simultaneously, the rotor electrical angle θ of the motor is acquired in real time via a photoelectric encoder. e and rotor angular velocity ω m The rotor angular velocity ω m That is, the actual rotational speed.

[0039] Step 2. Establish a first-order linear active disturbance rejection controller based on an improved LESO (i.e., an improved linear extended state observer) and a second-order IIR Butterworth high-pass filter.

[0040] Traditional ESO disturbance estimation channels contain only integral terms, and their transfer functions exhibit second-order low-pass characteristics (no zeros), leading to significant phase lag in the estimation of mid-to-high frequency disturbances. The only solution is to increase the observer bandwidth, but this amplifies high-frequency noise. Inspired by the introduction of zeros in the derivative (D) term of classical PID control to compensate for phase lag, this invention constructs an estimation error derivative term in the signal transmission path for estimating the total disturbance in the speed loop. This explicitly constructs a left-half-plane zero (s = -β) for the disturbance estimation transfer function in the s-domain coordinate plane. b / β c ), through β c By selecting appropriate values, the lag effect of the integration stage can be offset without increasing the observer bandwidth, thereby improving the observer's ability to estimate mid-to-high frequency disturbances.

[0041] Specifically, to address the issues of phase lag and bandwidth-noise contradiction in traditional ESO disturbance estimation during low-speed operation of permanent magnet synchronous motors, an improved LESO is designed by constructing a differential term for the state estimation error and its coefficients. A parallel "integral + differential" structure is built in the signal transmission path for estimating the total disturbance in the speed loop. The phase lead characteristic of differential operations effectively compensates for the phase delay of the integral stage, significantly improving the observer's tracking speed for mid-to-high frequency disturbances. By adjusting the differential gain coefficient, the frequency response of the disturbance estimation transfer function is optimized. When the differential gain coefficient increases, the zero point of the transfer function shifts to higher frequencies, improving the gain in the mid-to-high frequency range while avoiding excessive amplification of high-frequency noise. This improvement allows the observer to significantly improve the estimation speed of mid-to-high frequency disturbances such as periodic torque pulsations while retaining its ability to suppress high-frequency noise.

[0042] In addition, this invention also designs a second-order IIR Butterworth high-pass filter and combines it with linear error feedback control, which effectively solves the shortcomings of the improved linear extended state observer in high-frequency noise suppression. While ensuring that the improved LESO bandwidth meets the system requirements, it significantly suppresses the problem of periodic speed fluctuations caused by high-frequency noise.

[0043] The tracking performance of the improved LESO can be enhanced by increasing its bandwidth; however, excessively high bandwidth makes the system more sensitive to high-frequency noise, leading to the introduction of high-frequency ripple in the rotational speed. Therefore, this invention improves the steady-state speed tracking accuracy of the motor by adding a second-order IIR Butterworth high-pass filter to the speed feedforward compensation stage, extracting and compensating for high-frequency noise signals from the estimated rotor angular velocity, i.e., the estimated speed value z1.

[0044] After establishing a first-order linear active disturbance rejection controller based on an improved LESO and a second-order IIR Butterworth high-pass filter, the actual rotational speed ω is... m and the q-axis reference current i at the previous momentq * (k-1) is input to the improved linear extended state observer to obtain the speed and speed loop total disturbance estimates z1 and z2; the speed estimate z1 is input into a second-order IIR Butterworth high-pass filter to obtain the speed estimate z′1 after high-frequency noise compensation; then, the high-frequency noise compensated speed estimate z′1 and the speed loop total disturbance estimate z2 are substituted into the error feedback control law to obtain the reference current i of the q-axis at the current moment. q * (k).

[0045] The following is a further detailed explanation of the specific processing procedure of the first-order linear active disturbance rejection controller in this embodiment:

[0046] in i d * Under the control strategy of 0, the equation for electromagnetic torque is:

[0047]

[0048] Where T e It is electromagnetic torque, P n It is the number of pole pairs of the motor, ψ f It is a permanent magnet flux linkage, i q It is the q-axis stator current.

[0049] The mechanical motion equation of a surface-mounted permanent magnet direct drive motor is:

[0050]

[0051] Where J is the moment of inertia, and T e It is electromagnetic torque, T L B is the load torque, and B is the viscous friction coefficient.

[0052] The mechanical motion equations of the surface-mounted permanent magnet direct drive motor are reconstructed as follows:

[0053]

[0054] Where b0 = 1.5P n ψ f / J, F s =-(T) L +Bω m ) / J.

[0055] F s Expanding to a new state variable x2, the state-space equation of the system can be written as:

[0056]

[0057] Where x1 and x2 are the state variables of the system, i.e., x1 = ω m x2 = F s Where y is the system output and u is the system control input. The expression for the second-order linear extended state observer is as follows:

[0058]

[0059] Where e s This represents the speed estimation error, where z1 and z2 are the estimated values ​​of the system state variables x1 and x2, respectively. β a It is the gain coefficient of the state estimation error term, β b It is the gain coefficient of the integral term of the state estimation error.

[0060] To simplify parameter design, this embodiment uses the bandwidth method to determine β. a β b The possible values ​​of are:

[0061] β a =2ω s ,β b =ω s 2 .

[0062] Where ω s It is the bandwidth of the linear expansion state observer of the rotational speed loop.

[0063] From formula (5), the closed-loop transfer function of the estimated total disturbance z2 of the speed loop is:

[0064]

[0065] As can be seen from the above formula (6), the traditional ESO is a low-pass filter structure, which makes it difficult to capture mid-to-high frequency periodic disturbances. Based on this, this embodiment designs an improved linear extended state observer, such as... Figure 2 As shown.

[0066] This invention designs a first-order linear active disturbance rejection controller (ALS controller) to replace the traditional PI controller. Addressing the issues of phase lag and bandwidth-noise contradiction in traditional LESO disturbance estimation during low-speed operation of permanent magnet synchronous motors, an improved LESO is designed by constructing a differential term for the state estimation error and its coefficients. A parallel "integral + derivative" structure is built in the signal transmission path for estimating the total disturbance in the speed loop. The phase lead characteristic of the derivative operation effectively compensates for the phase delay of the integral stage, significantly improving the observer's tracking speed for mid-to-high frequency disturbances. By adjusting the derivative gain coefficient, the frequency response of the disturbance estimation transfer function is optimized. As the derivative gain coefficient increases, the zeros of the transfer function shift to higher frequencies, improving the gain in the mid-to-high frequency range while avoiding excessive amplification of high-frequency noise. This improvement allows the improved LESO to significantly increase the estimation speed for mid-to-high frequency disturbances such as periodic torque pulsations while retaining its ability to suppress high-frequency noise.

[0067] Specifically, the expression for the improved LESO in this embodiment is as follows:

[0068]

[0069] Where β c It is the gain coefficient of the differential term of the state estimation error. Its value should be determined in a way that ensures the stability of the observer and maintains the characteristics of the original second-order system.

[0070] According to formula (7), the transfer function of the estimated total disturbance z2 of the speed loop of the improved LESO is written as follows:

[0071]

[0072] From equations (6) and (8), it can be seen that in order to ensure the stability of the improved LESO and maintain the characteristics of the original second-order system, the original poles must remain unchanged. Therefore, β is taken as... c =2ω s -β a .

[0073] Combination Figure 2 A comparison of formulas (6) and (8) reveals that this invention constructs a state estimation error differential term and its coefficients to design an improved LESO. In the signal transmission path for estimating the total disturbance of the speed loop, a parallel structure of "integral + differential" is constructed. The phase lead characteristic of differential operation effectively compensates for the phase delay of the integral stage, significantly improving the observer's tracking speed for mid-to-high frequency disturbances. In the s-domain coordinate plane, by adjusting the differential gain coefficient β... c By changing the zero point of the LESO speed loop disturbance estimation transfer function, the frequency response of the disturbance estimation transfer function to mid-to-high frequency disturbances is improved, thereby improving its estimation accuracy for mid-to-high frequency periodic disturbances.

[0074] The transfer function of a second-order IIR Butterworth high-pass filter is:

[0075]

[0076] Where k h It is the filter gain coefficient, ω h =2πf h It is the cutoff angular frequency (unit: rad / s), f h This is the cutoff frequency.

[0077] Bode plot of a second-order IIR Butterworth high-pass filter is shown below. Figure 3 As shown, its permissible frequency is higher than f. h Signal components pass through with almost no attenuation, at frequencies below f. h The signal is rapidly attenuated, enabling precise separation and extraction of high-frequency noise signals.

[0078] Discretize it using the Tustin transform, and rewrite the original transfer function as follows:

[0079]

[0080] in It is the pre-distortion frequency, T s It is the sampling period (seconds).

[0081] Will Substituting into the transfer function and simplifying, we get:

[0082]

[0083] in

[0084] The error feedback control law with a second-order IIR Butterworth high-pass filter is designed as follows:

[0085]

[0086] Where z′1 is the estimated rotational speed after high-frequency noise compensation, and k s It is the gain coefficient for error feedback, i q * It is the q-axis reference current.

[0087] Step 3. Adopt a control strategy where the d-axis reference current is 0, use the error between the reference current and the sampled current as the control input, and generate the required control voltage by calculating the reference voltage.

[0088] In a preferred embodiment, this embodiment establishes a model-free predictive controller based on a current loop extended state observer in the current loop, and adopts a control strategy with the d-axis reference current being 0.

[0089] First, sample the current i a i b i c i is obtained after Clark transformation α i β Combined with the predicted voltage u from the previous moment α u β Stator current is estimated using a current loop expansion state observer. and total disturbance of current loop The estimated With reference current i α * i β * The voltage is input into the voltage prediction equation to obtain the predicted voltage u at the next time step. α * u β * .

[0090] The specific processing flow of the model-free predictive controller based on the current loop extended state observer is as follows:

[0091] The hyperlocal model of the permanent magnet direct drive motor is established as follows:

[0092]

[0093] in These represent the stator current, control voltage, and total current loop disturbance in the α-β coordinate system, respectively, where α is the input gain coefficient.

[0094] with i αβ and F αβ Assuming the state variables are the current loop extended state observer, construct the state-space form as follows:

[0095]

[0096] in and They are i αβ and F αβ The estimated value, where I is a 2×2 identity matrix, β d and β e This is the error feedback gain coefficient of the current loop extended state observer. To simplify system design, this embodiment also uses the bandwidth method to configure the gain coefficient as β. d =2ω c ,β e =ω c 2 , where ω c It is the bandwidth of the current loop expansion state observer.

[0097] like Figure 4 As shown, the state equations of the current loop extended state observer are discretized using the forward Euler method:

[0098]

[0099] in, Represents i at time k+1 and k. αβ The estimated value, F represents time k+1 and k. αβ The estimated value, T s This refers to the sampling time. After discretizing and rearranging the hyperlocal model of the permanent magnet direct drive motor using the forward Euler method, the expression for the reference voltage is obtained as follows:

[0100]

[0101] in Let Δθ = ω represent the rotation matrix. m T s , Represent the reference current at time k+1; obtain Then the predicted voltage u is obtained. α * u β * .

[0102] Of course, in this embodiment, the current loop is not limited to the model-free predictive controller based on the current loop extended state observer mentioned above. For example, a PI controller can also be used. The structure of the current loop using a PI controller is more conventional. Specifically, the sampled current is transformed to obtain its value in the dq coordinate system. The difference between the reference current and the sampled current is used to obtain the current error signal. The control voltage can be obtained by performing proportional and integral operations on the signal and then summing them.

[0103] Step 4. Calculate the predicted voltage u α * u β * The input is fed into the space vector pulse width modulation module to obtain the inverter's drive pulse signal, which completes the PMSM motor drive. This reduces the impact of periodic disturbances on the speed fluctuation amplitude without introducing additional high-frequency noise, thus achieving speed fluctuation suppression and disturbance rejection control for low-speed permanent magnet direct drive motors.

[0104] contrast Figure 5 As can be seen from (a) and (b) in the present invention, the improved LESO design significantly improves the observer's estimation accuracy for periodic disturbances, reduces torque pulsation caused by periodic disturbances, and makes the output speed fluctuation more stable.

[0105] contrast Figure 6 As can be seen from (a) and (b) in the present invention, by designing a second-order IIR Butterworth high-pass filter, has a steeper attenuation characteristic than a first-order high-pass filter, which can more quickly suppress the high-frequency ripple and measurement noise mixed in the motor speed signal, and further improve the high-frequency noise sensitivity problem caused by the increase in bandwidth.

[0106] contrast Figure 7 As can be seen from (a) and (b) in the paper, compared with the traditional PI control used in the current loop, the model-free predictive control of this invention significantly reduces the system's dependence on accurate modeling of motor parameters (such as inductance, resistance, etc.), reduces the current tracking error caused by parameter mismatch, and reduces the speed fluctuation caused by current fluctuation.

[0107] contrast Figure 8 As can be seen from (a) and (b) in the figure, compared with the traditional PI control method, the speed fluctuation amplitude of the present invention is significantly reduced under low-speed operating conditions, by about 0.271 rpm, thus verifying the effectiveness of the control method.

[0108] Example 2

[0109] This embodiment 2 describes a speed fluctuation suppression and disturbance rejection control system for a low-speed permanent magnet direct drive motor. This system is based on the same inventive concept as the speed fluctuation suppression and disturbance rejection control method for a low-speed permanent magnet direct drive motor in embodiment 1 above.

[0110] A speed fluctuation suppression and disturbance rejection control system for low-speed permanent magnet direct drive motors includes the following modules:

[0111] The sensor acquisition module is used to acquire the three-phase current of the motor in real time and convert the sampled current into the current value in the α-β coordinate system through Clark transformation; at the same time, it can acquire the position and actual speed of the motor rotor in real time.

[0112] The active disturbance rejection control module is used to establish a first-order linear active disturbance rejection controller based on an improved LESO and a second-order IIR Butterworth high-pass filter. The improved LESO is designed by constructing the differential term of the state estimation error and its gain coefficient. An "integral + derivative" parallel structure is constructed in the signal transmission path of the estimated total disturbance of the speed loop. The phase lead characteristic of the derivative operation is used to compensate for the phase delay of the integral link, so as to improve the tracking speed of the observer for mid-to-high frequency disturbances. At the same time, the frequency response of the disturbance estimation transfer function is optimized by adjusting the derivative gain coefficient.

[0113] The processing flow of the first-order linear active disturbance rejection controller is as follows: The actual rotational speed and the reference current of the q-axis at the previous moment are input into the improved LESO to obtain the estimated values ​​of the rotational speed and the total disturbance of the rotational loop; The estimated value of the rotational speed is input into the second-order IIR Butterworth high-pass filter to obtain the estimated value of the rotational speed after high-frequency noise compensation; Then, the estimated value of the rotational speed and the total disturbance of the rotational loop after high-frequency noise compensation are substituted into the error feedback control law to obtain the reference current of the q-axis at the current moment.

[0114] The voltage prediction module is used to generate the required control voltage by using a control strategy where the d-axis reference current is 0, taking the error between the reference current and the sampled current as the control input, and calculating the reference voltage.

[0115] The system also includes a drive module, which inputs the calculated control voltage into the space vector pulse width modulation module to obtain the drive pulse signal for the inverter, thereby completing the motor drive and achieving speed fluctuation suppression and disturbance rejection control for the low-speed permanent magnet direct drive motor.

[0116] In this embodiment, the sensor acquisition module includes a current sensor and a photoelectric encoder. The current sensor is used to acquire the three-phase current of the motor in real time; the photoelectric encoder is used to acquire the position and actual speed of the motor rotor in real time.

[0117] Taking the voltage prediction module using the aforementioned model-free predictive controller based on the current loop extended state observer as an example, the active disturbance rejection control module analyzes the reference rotational speed ω. m * With actual rotational speed ω m The q-axis current reference value is obtained after high-frequency noise suppression. The voltage prediction module adopts a control strategy with the d-axis stator current being 0. It estimates the stator current and the total current loop disturbance through the current loop expansion state observer. The estimated values ​​of the stator current and the total current loop disturbance are then fed into the predicted voltage calculation module to obtain the voltage reference value at the next moment. The voltage reference value output by the voltage prediction module is then input into the space vector pulse width modulation module (SVPWM) to generate the inverter's switching signal, thereby driving the motor.

[0118] It should be noted that the implementation process of the functions and roles of each functional module in the speed fluctuation suppression and disturbance rejection control system in this embodiment 2 is detailed in the implementation process of the corresponding steps of the method in the above embodiment 1, and will not be repeated here.

[0119] Example 3

[0120] This embodiment 3 describes a computer device. The computer device includes a memory and one or more processors. Executable code is stored in the memory. When the processor executes the executable code, it implements the steps of the speed fluctuation suppression and disturbance rejection control method for a low-speed permanent magnet direct drive motor described in embodiment 1 above.

[0121] In this embodiment, the computer device can be any device or apparatus with data processing capabilities, and will not be described in detail here.

[0122] Example 4

[0123] This embodiment 4 describes a computer-readable storage medium storing a program that, when executed by a processor, is used to implement the steps of the speed fluctuation suppression and disturbance rejection control method for a low-speed permanent magnet direct drive motor described in embodiment 1 above.

[0124] The computer-readable storage medium can be an internal storage unit of any device or apparatus with data processing capabilities, such as a hard disk or memory, or an external storage device of any device with data processing capabilities, such as a plug-in hard disk, smart media card (SMC), SD card, flash card, etc.

[0125] Of course, the above description is only a preferred embodiment of the present invention. The present invention is not limited to the above-described embodiments. It should be noted that any equivalent substitutions or obvious modifications made by those skilled in the art under the guidance of this specification fall within the scope of this specification and should be protected by the present invention.

Claims

1. A speed fluctuation suppression disturbance control method for a low-speed permanent magnet direct drive motor, characterized by, Comprising the following steps: Step 1. Real-time acquisition of motor three-phase current, and conversion of the sampled current into current values in the alpha-beta coordinate system through Clark transformation; at the same time, real-time acquisition of the position and actual speed of the motor rotor; Step 2. Establishing a first-order linear active disturbance rejection controller based on an improved LESO and a second-order IIR Butterworth high-pass filter; wherein the improved LESO is designed by constructing a state estimation error differential term and its gain coefficient, and an "integral+differential" parallel structure is constructed in the signal transfer path of the estimated speed loop total disturbance, the phase lead characteristic of the differential operation is utilized to compensate for the phase delay of the integral element, so as to improve the tracking speed of the observer for the medium and high frequency disturbance; at the same time, the differential gain coefficient is adjusted to optimize the frequency response of the disturbance estimation transfer function; The processing flow of the first-order linear active disturbance rejection controller is as follows: the acquired actual speed and the reference current of the q-axis at the last moment are input into the improved LESO to obtain the estimated values of the speed and the total disturbance of the speed loop; the speed estimation value is input into the second-order IIR Butterworth high-pass filter to obtain the speed estimation value after high-frequency noise compensation, and then the speed estimation value after high-frequency noise compensation and the estimated value of the total disturbance of the speed loop are substituted into the error feedback control law to obtain the reference current of the q-axis at the current moment; Step 3. Adopting a control strategy with the d-axis reference current being 0, taking the error between the reference current and the sampled current as the control input, and generating the required control voltage through calculation of the reference voltage; Step 4. Inputting the calculated control voltage into a space vector pulse width modulation module to obtain the driving pulse signal of the inverter, completing the motor driving, and realizing the speed fluctuation suppression and disturbance rejection control of the low-speed permanent magnet direct drive motor.

2. The speed fluctuation suppression disturbance control method for a low-speed permanent magnet direct drive motor according to claim 1, characterized by, In step 2, the calculation expression of the improved LESO is as follows: where e s is the speed estimation error, ω m denotes the actual speed, z1 is the estimated value of the speed, z2 is the estimated value of the total disturbance of the speed loop, β a , β b and β c are gain coefficients of the improved LESO, β a is the gain coefficient of the state estimation error term, β b is the gain coefficient of the state estimation error integral term, and β c is the gain coefficient of the state estimation error differential term; u is the system output, i.e., the reference current b0=1.5P n ψ f / J, P n is the number of motor pole pairs, ψ f is the permanent magnet flux linkage, and J is the rotational inertia.

3. The speed fluctuation suppression disturbance control method for a low-speed permanent magnet direct drive motor according to claim 2, characterized by, In step 2, the gain coefficient β is determined by bandwidth method a and β b is the gain coefficient of the improved LESO a = 2ω s , β b = ω s 2 , ω s is the bandwidth of the improved LESO; the value of β c must ensure the stability of the observer and maintain the original second-order system characteristics, β c = 2ω s - β a .

4. The speed fluctuation suppression disturbance control method for a low-speed permanent magnet direct drive motor according to claim 2, characterized by, In step 2, the closed-loop transfer function z2(s) of the estimated value z2 of the total disturbance of the speed loop is: where F s represents the mathematical representation of the total disturbance of the speed loop in the Laplace domain.

5. The speed fluctuation suppression disturbance control method for a low-speed permanent magnet direct drive motor according to claim 1, characterized by, In step 2, the expression of the transfer function of the second-order IIR Butterworth high-pass filter is as follows: where k h is a filter gain coefficient, ω h = 2πf h is a cutoff angular frequency, in rad / s; f h is a cutoff frequency; The original transfer function is rewritten as: wherein is the pre-distortion frequency, T s is the sampling period in seconds.

6. The speed fluctuation suppression disturbance control method for a low-speed permanent magnet direct drive motor according to claim 5, characterized by, In step 2, the error feedback control law with the second-order IIR Butterworth high-pass filter is designed as follows: where z1' is the speed estimation value after high-frequency noise compensation; k s is the gain coefficient of error feedback, ω m represents the reference speed, z1 is the estimation value of the speed, and z2 is the estimation value of the total disturbance of the speed loop; b0=1.5P n ψ f / J, is the q-axis reference current.

7. The speed fluctuation suppression disturbance control method for a low-speed permanent magnet direct drive motor according to claim 1, characterized by, In step 3, a model-free predictive controller based on a current loop extended state observer and a predictive voltage equation is established; the model-free predictive controller adopts a control strategy with the d-axis stator current being 0, and obtains the values of the d-axis and q-axis reference currents in the alpha-beta coordinate system at the current moment through inverse Park transformation; the current values in the alpha-beta coordinate system obtained by Clark transformation of the sampled current and the predicted voltage at the last moment are input into the current loop extended state observer to obtain the estimated values of the reference current and the total disturbance; the estimated values of the reference current and the total disturbance and the values of the d-axis and q-axis reference currents in the alpha-beta coordinate system at the current moment obtained through inverse Park transformation are input into the predictive voltage equation to obtain the predicted voltage at the next moment, i.e. the control voltage.

8. A speed fluctuation suppression disturbance control system for a low-speed permanent magnet direct drive motor, characterized by, Comprising the following modules: The sensor acquisition module is configured to acquire three-phase currents of the motor in real time, and convert the sampled currents into current values in an α-β coordinate system through Clark transformation; meanwhile, the position and actual speed of the rotor of the motor are acquired in real time; The active disturbance rejection control module is configured to establish a first-order linear active disturbance rejection controller based on an improved LESO and a second-order IIR Butterworth high-pass filter; wherein the improved LESO is designed by constructing a state observation error differential term and a gain coefficient thereof, and an "integral+differential" parallel structure is constructed in a signal transmission path of the estimated speed loop total disturbance, the phase lead characteristic of differential operation is utilized to compensate for the phase delay of the integral element, so as to improve the tracking speed of the observer for the medium-high frequency disturbance; meanwhile, the differential gain coefficient is adjusted to optimize the frequency response of the disturbance estimation transfer function; The processing procedure of the first-order linear active disturbance rejection controller is as follows: the acquired actual speed and the reference current of the q-axis at the previous moment are input into the improved LESO to obtain the estimated values of the speed and the total disturbance of the speed loop; the speed estimation value is input into the second-order IIR Butterworth high-pass filter to obtain the speed estimation value after high-frequency noise compensation, and then the speed after high-frequency noise compensation and the estimated value of the total disturbance of the speed loop are substituted into the error feedback control law to obtain the reference current of the q-axis at the current moment; The voltage prediction module is configured to adopt a control strategy in which the d-axis reference current is 0, take the error between the reference current and the sampled current as a control input, and generate the required control voltage by calculating a reference voltage; The driving module is configured to input the calculated control voltage into a space vector pulse width modulation module to obtain driving pulse signals of the inverter, complete motor driving, and implement speed fluctuation rejection anti-disturbance control for the low-speed permanent magnet direct drive motor.

9. A computer device comprising a memory and one or more processors; having stored in the memory executable code; characterized in that, When the processor executes the executable code, the steps of the speed fluctuation rejection anti-disturbance control method for the low-speed permanent magnet direct drive motor as claimed in any one of claims 1 to 7 are implemented.

10. A computer readable storage medium having stored thereon a program; characterized in that, When the processor executes the program, the steps of the speed fluctuation rejection anti-disturbance control method for the low-speed permanent magnet direct drive motor as claimed in any one of claims 1 to 7 are implemented.

Images