Resonant frequency identification and suppression method and system of mechanical-gear-free integrated vernier motor

Abstract

The invention discloses a resonant frequency identification and suppression method and system for a mechanical-gear-free integrated vernier motor, belongs to the field of mechanical-gear-free integrated vernier motor control, and is suitable for robot joint driving, high-precision servo systems and other scenes. The invention relates to a resonant frequency on-line rapid identification method, which comprises the following steps of: introducing an additional control degree of freedom into a traditional second-order generalized integrator-frequency-locked loop, optimizing an orthogonal signal generation process, and shortening convergence time in combination with an adaptive frequency correction link; the problems that a traditional resonant frequency identification algorithm is poor in identification accuracy and low in convergence speed are solved. According to the high-performance resonance suppression strategy, disturbance is estimated in real time through a linear expansion state observer; parameter setting is simple, the problem that multi-parameter setting of a traditional wave trap is complex is solved, and the strong-robustness resonance suppression effect under load inertia changes is achieved.

Description

Technical Field

[0001] The present invention belongs to the field of mechanical gearless integrated vernier motor control, and in particular relates to a method and system for identifying and suppressing the resonant frequency of a mechanical gearless integrated vernier motor. Background Art

[0002] Robotic integrated joints utilize a cascaded servo motor and mechanical reducer. Traditional mechanical reducers suffer from vibration and frictional losses, high noise levels, and the need for regular maintenance. Coaxial magnetic gears, on the other hand, rely on the magnetic coupling of their inner and outer rotor permanent magnets to achieve torque transmission. This offers advantages such as zero mechanical contact, zero frictional losses, and zero maintenance, making them ideal for applications such as robotic joint drive and high-precision servo drives. Magnetic gears cannot provide active torque, so the proposed technology is based on a non-mechanically geared integrated vernier motor (NMGIVM) created by cascading a permanent magnet vernier motor and magnetic gears.

[0003] Currently, resonance identification algorithms for permanent magnet synchronous motors are primarily categorized into two types: offline and online. Offline identification, by applying specific excitation signals in a controlled environment, offers the advantages of high-precision identification, strong anti-interference capabilities, and a high degree of algorithmic freedom. However, this method struggles to dynamically adapt to the frequently changing operating conditions of robot joints, resulting in poor algorithmic flexibility and robustness. Online identification, by detecting the resonant frequency in real time and dynamically responding to system changes, offers the advantages of real-time performance and strong adaptability. However, it suffers from weak anti-interference capabilities, high computational resource requirements, and limited accuracy compared to offline identification methods. For integrated vernier motors without mechanical gears, the resonant frequency identification algorithm must balance identification speed and accuracy to ensure good dynamic and steady-state performance.

[0004] Strategies for suppressing electrical resonance in permanent magnet synchronous motors are primarily categorized as active and passive. Active suppression primarily achieves this by changing the controller structure. A widely used approach is the combination of PI control and state feedback, but this requires precise system modeling. Passive suppression, while not changing the system's control structure, employs techniques such as trap filters and input shapers. However, this approach can easily degrade the servo system's position control performance. For integrated vernier motors without mechanical gears, these traditional resonance suppression algorithms face higher requirements in order to effectively reduce the resonance amplitude or even completely offset the resonance peak, while also focusing on the impact of disturbances on the suppression effect. Summary of the Invention

[0005] In response to the problems existing in the prior art, the present invention provides a method and system for identifying and suppressing the resonant frequency of a vernier motor without mechanical gears, which can effectively reduce the resonance amplitude and even completely offset the resonance peak.

[0006] To solve the above technical problems, the present invention provides the following technical solution: a method for identifying and suppressing the resonant frequency of a non-mechanical gear integrated vernier motor, comprising the following steps:

[0007] S1. Collect the motor speed error signal and improve the traditional second-order generalized integrator-frequency-locked loop structure to increase the control degree of freedom g, optimize the orthogonal signal phase difference, and use the improved second-order generalized integrator-frequency-locked loop structure SOGI-FLL to identify the resonant frequency online;

[0008] S2. Based on the identified resonant frequency obtained in step S1, a linear active disturbance rejection controller (LADRC) is used to generate a compensation control variable. The linear active disturbance rejection controller (LADRC) estimates the system disturbance using a linear extended state observer (LESO) and adjusts the observer gain using a bandwidth parameterization method. A compensation signal is generated based on the speed tracking error and the observed disturbance using a linear state error feedback law (LSEF).

[0009] S3. The compensation control amount is superimposed on the current loop reference information. The current loop adopts current tracking performance and dynamic response zero-beat predictive current control to suppress speed and torque resonance.

[0010] Furthermore, in the aforementioned step S1, the open-loop transfer function G1 of the output voltage of the improved second-order generalized integrator-frequency-locked loop structure SOGI-FLL with respect to the reference voltage, and the open-loop transfer function G2 of the orthogonal voltage signal with respect to the reference voltage are respectively as follows:

[0011]

[0012] Among them, k is the control gain, g is the additional degree of freedom parameter, ω * To estimate the angular frequency, s is the Laplace operator, u and u qua are the output voltage signal and the quadrature voltage signal, u ref Indicates the input voltage signal.

[0013] Furthermore, in the aforementioned step S1, the improved second-order generalized integrator-frequency-locked loop structure SOGI-FLL adds a control degree of freedom g on the basis of the traditional second-order generalized integrator-frequency-locked loop, and converts the input error signal e u It is introduced into the calculation and generation of the orthogonal signal of the second-order generalized integrator, and the flexible control of the output signal of the second-order generalized integrator is achieved by adjusting the values of k and g at the same time.

[0014] Furthermore, in the aforementioned step S1, a frequency adaptive control link is added to the phase-locked loop control of the improved second-order generalized integrator-frequency-locked loop structure SOGI-FLL, and the added adaptive control gain expression is as follows:

[0015]

[0016] Among them, k is the input voltage error signal amplification factor, u and u qua are the output voltage and quadrature voltage signals respectively.

[0017] Furthermore, in the aforementioned step S2, a linear active disturbance rejection controller LADRC is used to generate a compensation control variable, specifically: based on the speed loop of the linear active disturbance rejection controller LADRC, a speed signal is received and a current control signal is generated; based on the predicted current control of the deadbeat current control DPCC, a current control signal is received and a corresponding voltage control signal is generated.

[0018] Furthermore, in the aforementioned step S2, the linear extended state observer LESO is used to model the internal and external disturbances of the system as an extended state variable z2, and its observer bandwidth w0 is set to the motor closed-loop bandwidth w c The preset multiple of , the LESO equation is expressed as:

[0019]

[0020] Where z1, z2, and z3 are the estimated values of x1, x2, and x3 respectively, and b 01 、b 02 、b 03 is the linear extended state observer LESO parameter, u is the system control variable, b is the control variable gain, and y is the system output;

[0021] The adjustable parameters of the linear extended state observer LESO satisfy the following formula:

[0022] β 01 =2ω0

[0023] β 02 =ω0 2

[0024] The linear extended state observer LESO also includes a linear state error feedback law LSEF, which generates a compensation signal according to the speed tracking error e1 and the observed disturbance z2. Its expression is:

[0025] u0=β 11 e1=β 11 (x1-z1)

[0026]

[0027] In the formula, u0 is the original control quantity of the system, b is the compensation gain coefficient, β 11 is the error gain coefficient, x1 is the actual value of the parameter, and z1 is the estimated value of the parameter.

[0028] The present invention also provides a resonance suppression system for a vernier motor without mechanical gear integration, comprising:

[0029] Improved second-order generalized integrator-frequency-locked loop structure SOGI-FLL for online identification of resonant frequency;

[0030] The active disturbance rejection controller LADRC includes a linear extended state observer LESO and a linear error feedback control law LSEF, which is used to generate a compensation signal;

[0031] The current loop deadbeat predictive controller is used to receive the compensation current signal output by the LADRC and generate a corresponding voltage control signal.

[0032] Furthermore, the aforementioned resonance suppression system of a vernier motor without mechanical gear integration includes:

[0033] The inner rotor is mounted on the motor shaft and has P i For permanent magnets;

[0034] The outer rotor is coaxially coupled with the outer rotor through the magnetic gear, and its surface is embedded with P o For permanent magnets, P o =P i +N, where N is the number of magnetic modulation poles of the magnetic modulation ring;

[0035] The stator winding adopts fractional slot concentrated winding with a pole-slot ratio of 2P o :Q, Q is the number of slots.

[0036] Compared with the prior art, the beneficial technical effects of the present invention using the above technical solution are as follows:

[0037] 1. The present invention makes full use of the dual-inertia model characteristics of the mechanical gearless integrated vernier motor, and obtains the system transfer function, natural resonant frequency and anti-resonant frequency by analyzing the transmission characteristics of the magnetic gear. Based on the closed-loop control system, the influence of the change of controller gain on the resonant characteristics is studied, providing a theoretical basis for the setting of the controller gain.

[0038] 2. The improved SOGI-FLL algorithm proposed in the present invention for the resonance identification of the integrated vernier motor without mechanical gears increases the one-dimensional control freedom, improves the sinusoidality and orthogonality of the output signal, and shortens the convergence time; it solves the problem that traditional algorithms are difficult to adapt to the real-time changes in the resonant frequency of the integrated vernier motor without mechanical gears with operating conditions and load conditions, and has higher identification accuracy and convergence speed.

[0039] 3. The resonance suppression strategy based on linear anti-disturbance control proposed by the present invention for the resonance suppression of the integrated vernier motor without mechanical gears is relatively simple to adjust its parameters. It regards the internal and external disturbances of the system as a disturbance term. By real-time estimation and compensation control of the disturbance, the influence of the disturbance is eliminated, and it has good steady-state and dynamic performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 This is a structural block diagram of an improved second-order generalized integrator-frequency-locked loop algorithm for resonance identification of an integrated vernier motor without mechanical gears according to the present invention.

[0041] Figure 2 It is a structural schematic diagram of the resonance suppression method of the mechanical gearless integrated vernier motor based on linear active disturbance rejection control of the present invention.

[0042] Figure 3 It is a second-order linear LADRC control block diagram of the invented resonance suppression method of the mechanical gearless integrated vernier motor.

[0043] Figure 4 This is a comparison chart of the identification and simulation results of the invented improved SOGI-FLL method and the traditional SOGI-FLL method.

[0044] Figure 5 This is a comparison chart of the simulation results of the resonance suppression effect of the invented LADRC resonance suppression method and the traditional PI controller + notch filter method. DETAILED DESCRIPTION

[0045] In order to better understand the technical content of the present invention, specific embodiments are given and described below with reference to the accompanying drawings.

[0046] Various aspects of the present invention are described herein with reference to the accompanying drawings, which show a number of illustrative embodiments. The embodiments of the present invention are not limited to those described in the accompanying drawings. It should be understood that the present invention can be implemented by any of the various concepts and embodiments described above, as well as the concepts and implementations described in detail below, because the concepts and embodiments disclosed herein are not limited to any particular implementation. In addition, some aspects disclosed herein may be used alone or in any appropriate combination with other aspects disclosed herein.

[0047] refer to Figure 1 , Figure 1 This is the improved second-order generalized integrator-frequency-locked loop algorithm structure diagram for resonance identification of the non-mechanical gear integrated vernier motor of the present invention, where u ref 、u、u qua are the input signal, the output signal of the second-order generalized integrator SOGI that is in phase with the input signal, and the output signal of the second-order generalized integrator SOGI that is orthogonal to the input signal; eu is the input error signal, k is the SOGI control gain; x1, x2, and x3 are the three integrator outputs respectively; γ is the FLL control gain; ω ref is the initial given angular frequency; ω * is the estimated angular frequency.

[0048] The present invention provides a method for identifying and suppressing the resonant frequency of a non-mechanical gear integrated vernier motor, comprising the following steps:

[0049] S1. Collect the motor speed error signal and improve the traditional second-order generalized integrator-frequency-locked loop structure to increase the control degree of freedom g, optimize the orthogonal signal phase difference, and use the improved second-order generalized integrator-frequency-locked loop structure SOGI-FLL to identify the resonant frequency online;

[0050] S2. Based on the identified resonant frequency obtained in step S1, a linear active disturbance rejection controller (LADRC) is used to generate a compensation control variable. The linear active disturbance rejection controller (LADRC) estimates the system disturbance using a linear extended state observer (LESO) and adjusts the observer gain using a bandwidth parameterization method. A compensation signal is generated based on the speed tracking error and the observed disturbance using a linear state error feedback law (LSEF).

[0051] S3. The compensation control amount is superimposed on the current loop reference information. The current loop adopts current tracking performance and dynamic response zero-beat predictive current control to suppress speed and torque resonance.

[0052] The signal transmission process of the frequency locked loop is as follows: the orthogonal signal u output by the second-order generalized integrator SOGI is converted into qua and the input error signal e u After multiplication, the angular frequency correction signal e is obtained by filtering with a low-pass filter. f , e f The angular frequency correction is then obtained through the adaptive adjustment link of the frequency-locked loop (FLL) control gain and the accumulation action of the integrator, and is superimposed on the set reference angular frequency to obtain the estimated angular frequency value. In the entire adjustment process of the improved second-order generalized integrator-frequency-locked loop structure SOGI-FLL, the estimated angular frequency also participates in the process of the second-order generalized integrator SOGI generating orthogonal signals.

[0053] Improved second-order generalized integrator-frequency locked loop structure SOGI-FLL resonant frequency identification algorithm adds a control degree of freedom g on the basis of the traditional second-order generalized integrator-frequency locked loop structure, and converts the input error signal e u It is introduced into the calculation and generation of SOGI quadrature signals. By adjusting the values of k and g simultaneously, the control of the SOGI output signal of the second-order generalized integrator is made more flexible.

[0054] According to the improved SOGI-FLL structure diagram, its transfer functions are:

[0055]

[0056] Where G1 and G2 are the open-loop transfer functions of the output voltage and quadrature voltage signals with respect to the reference voltage, respectively, and s is the Laplace operator. Introducing the g value changes the zero-pole distribution of the transfer function, thereby affecting the output response performance of the system.

[0057] In addition, in order to accelerate the frequency convergence to a stable value and keep the convergence value unchanged, this method adds a frequency adaptive control link to the frequency locked loop FLL control. The additional adaptive control gain expression is:

[0058]

[0059] Among them, k is the input voltage error signal amplification factor, u and u qua are the output voltage and quadrature voltage signals respectively.

[0060] Figure 2 This is a schematic diagram of the resonance suppression method and system structure of a mechanical gearless integrated vernier motor based on linear anti-disturbance control of the present invention. Compared with the traditional proportional integral (PI) vector control link, the present invention is characterized in that linear anti-disturbance control and zero-beat predictive current control are used to replace the traditional PI speed control and current control respectively. The system consists of a mechanical gearless integrated vernier motor, a three-phase inverter, a space vector pulse width modulation (SVPWM) module, a sampling circuit, a linear anti-disturbance control speed controller, a zero-beat current predictive controller, Clarke coordinate transformation, Park coordinate transformation, inverse Park coordinate transformation and other links.

[0061] First, the motor current information is obtained through the sampling circuit, and the angle and speed information is obtained by the sensor. The speed error signal is sent to the invented linear active disturbance rejection controller LADRC, and the current error signal is sent to the zero-beat predictive current controller. The dq-axis voltage output by the current controller is subjected to inverse Park transformation and space voltage vector pulse width modulation to obtain the on-off signal for controlling the power devices of the three-phase inverter. Finally, the mechanical gearless integrated vernier motor is driven to operate through the power conversion circuit.

[0062] Figure 3This is the control block diagram of the linear active disturbance rejection controller (LADRC) for the resonance suppression method of the mechanical gearless integrated vernier motor of the present invention. It includes a linear tracking differentiator (LTD), a linear extended state observer (LESO), and a linear state error feedback control law (LSEF). The LTD's main function is to pre-arrange the transient process and extract the input signal containing random noise and its differential signal, thereby resolving the contradiction between overshoot and rapidity. Its linear discrete expression is:

[0063]

[0064] Where v(k) is the system input signal, e(k) is the estimation error, and r is the speed factor.

[0065] The Linear Extended State Observer (LESO) treats the unknown internal and external disturbances of the system as a disturbance term and expands it into a new state variable. It observes each state variable in real time through the input and output of the system and dynamically estimates and compensates for the disturbance term. The LESO equation is expressed as:

[0066]

[0067] Where, e1 is the tracking error between LESO and the output signal, z1, z2, and z3 are the estimated values of x1, x2, and x3 respectively, and b 01 、b 02 、b 03 is the LESO parameter, u is the system control quantity, b is the control quantity gain, and y is the system output.

[0068] The linear state error feedback control law LSEF adopts a linear combination of error values, and the system becomes a linear integral series link, avoiding nonlinear functions. Its expression is:

[0069]

[0070] Where, e2 is the parameter estimation error, β 11 and β 22 are error gain coefficients.

[0071] The parameter design principle of the linear extended state observer LESO is:

[0072] ① The observer bandwidth w0 is set to the motor closed-loop bandwidth w c 3-5 times of LESO, the adjustable parameters satisfy:

[0073] β 01 =2ω0

[0074] β 02 =ω0 2

[0075] ② The linear state error feedback law used in the invented LESO generates a compensation signal based on the speed tracking error e1 and the observed disturbance z2, and its expression is:

[0076] u0=β 11 e1=β 11 (x1-z1)

[0077]

[0078] In the formula, u0 is the original control quantity of the system, b is the compensation gain coefficient, β 11 is the error gain coefficient, x1 is the actual value of the parameter, and z1 is the estimated value of the parameter.

[0079] The present invention also provides a resonance suppression system for a vernier motor without mechanical gear integration, comprising:

[0080] Improved second-order generalized integrator-frequency-locked loop structure SOGI-FLL for online identification of resonant frequency;

[0081] The active disturbance rejection controller LADRC includes a linear extended state observer LESO and a linear error feedback control law LSEF, which is used to generate a compensation signal;

[0082] The current loop deadbeat predictive controller is used to receive the compensation current signal output by the LADRC and generate a corresponding voltage control signal.

[0083] As a resonance suppression system for an integrated vernier motor without mechanical gears, it also includes:

[0084] The inner rotor is mounted on the motor shaft and has P i For permanent magnets;

[0085] The outer rotor is coaxially coupled with the outer rotor through the magnetic gear, and its surface is embedded with P o For permanent magnets, P o =P i +N, where N is the number of magnetic modulation poles of the magnetic modulation ring;

[0086] The stator winding adopts fractional slot concentrated winding with a pole-slot ratio of 2P o :Q, Q is the number of slots.

[0087] Figure 4This is a comparison chart of the simulation results of the identification results of the invented improved SOGI-FLL method and the traditional SOGI-FLL method. In the simulation, the frequency identification process during starting, loading, and unloading is simulated at 0s, 1s, and 3s respectively. Among them, the parameters of the improved SOGI-FLL are set as: k=0.8, g=-0.1, γ=6.5, and the motor reference speed is 100rpm. According to the simulation results, it can be seen from the speed error signal that speed resonance exists during starting, loading, and unloading; the two signals output by SOGI always maintain good orthogonality; the angular frequency identification results show that after a short-time convergence process of about 0.3s, the angular frequency gradually converges to a stable value of 47rad / s and remains unchanged. Since resonance is an inherent property of the integrated vernier motor without mechanical gears, under the premise that the motor structure and parameters remain unchanged, the resonant frequency caused by starting, loading, and unloading is the same value. Compared with the traditional SOGI-FLL, the convergence deviation and convergence fluctuation of the improved SOGI-FLL are effectively reduced, and the convergence speed is accelerated at startup, which reflects the superiority of the resonance identification method proposed in the present invention.

[0088] Figure 5 This chart compares the resonance suppression effects of the invented LADRC resonance suppression method with those of a traditional PI controller + notch filter method. In the simulation, the motor reference speed was set to 100 rpm. Starting, adding a 5 Nm load, and removing a 5 Nm load were simulated at 0s, 1s, and 3s. Both the invented LADRC method and the traditional PI + notch filter method were used to suppress motor resonance. Figure 5 The torque and speed waveforms of the motor under the two resonance suppression methods are shown. The simulation results show that the LADRC method significantly reduces the resonance amplitude and convergence time during startup and load addition and reduction, both for the torque and speed waveforms, compared to the traditional PI control + notch filter method. This demonstrates the effectiveness of the proposed LADRC.

[0089] While the present invention has been described above with reference to preferred embodiments, this is not intended to limit the present invention. Persons skilled in the art will readily appreciate that various modifications and variations can be made without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention shall be determined by the appended claims.

Claims

1. A method for identifying and suppressing the resonant frequency of a non-mechanical gear integrated vernier motor, characterized in that: The following steps are involved: S1. Collect the motor speed error signal and improve the traditional second-order generalized integrator-frequency-locked loop structure to increase the control degree of freedom g, optimize the orthogonal signal phase difference, and use the improved second-order generalized integrator-frequency-locked loop structure SOGI-FLL to identify the resonant frequency online; S2. Based on the identified resonant frequency obtained in step S1, a linear active disturbance rejection controller (LADRC) is used to generate a compensation control variable. The linear active disturbance rejection controller (LADRC) estimates the system disturbance using a linear extended state observer (LESO) and adjusts the observer gain using a bandwidth parameterization method. A compensation signal is generated based on the speed tracking error and the observed disturbance using a linear state error feedback law (LSEF). S3. The compensation control amount is superimposed on the current loop reference information. The current loop adopts current tracking performance and dynamic response zero-beat predictive current control to suppress speed and torque resonance.

2. The method for identifying and suppressing the resonant frequency of a non-mechanical gear integrated vernier motor according to claim 1, characterized in that: In step S1, the open-loop transfer function G1 of the output voltage of the improved second-order generalized integrator-frequency-locked loop structure SOGI-FLL with respect to the reference voltage and the open-loop transfer function G2 of the orthogonal voltage signal with respect to the reference voltage are respectively as follows: Among them, k is the control gain, g is the additional degree of freedom parameter, ω * To estimate the angular frequency, s is the Laplace operator, u and u qua are the output voltage signal and the quadrature voltage signal, u ref Indicates the input voltage signal.

3. The method for identifying and suppressing the resonant frequency of a mechanical gearless integrated vernier motor according to claim 1, characterized in that: In step S1, the improved second-order generalized integrator-frequency-locked loop structure SOGI-FLL adds a control degree of freedom g on the basis of the traditional second-order generalized integrator-frequency-locked loop, and converts the input error signal e u It is introduced into the calculation and generation of the orthogonal signal of the second-order generalized integrator, and the flexible control of the output signal of the second-order generalized integrator is achieved by adjusting the values of k and g at the same time.

4. The method for identifying and suppressing the resonant frequency of a mechanical gearless integrated vernier motor according to claim 1, characterized in that: In step S1, a frequency adaptive control link is added to the phase-locked loop control of the improved second-order generalized integrator-frequency-locked loop structure SOGI-FLL. The added adaptive control gain expression is as follows: Among them, k is the input voltage error signal amplification factor, u and u qua are the output voltage and quadrature voltage signals respectively.

5. The method for identifying and suppressing the resonant frequency of a mechanical gearless integrated vernier motor according to claim 1, characterized in that: In step S2, a linear active disturbance rejection controller (LADRC) is used to generate a compensation control variable. Specifically, the speed loop of the linear active disturbance rejection controller (LADRC) receives a speed signal and generates a current control signal. The predicted current control of the deadbeat current control (DPCC) receives a current control signal and generates a corresponding voltage control signal.

6. The method for identifying and suppressing the resonant frequency of a mechanical gearless integrated vernier motor according to claim 1, characterized in that: In step S2, the linear extended state observer LESO is used to model the system internal and external disturbances as extended state variables z2, and its observer bandwidth w0 is set to the motor closed-loop bandwidth w c The preset multiple of The LESO equation is expressed as: Where z1, z2, and z3 are the estimated values of x1, x2, and x3 respectively, and b 01 、b 02 、b 03 is the linear extended state observer LESO parameter, u is the system control variable, b is the control variable gain, and y is the system output; The adjustable parameters of the linear extended state observer LESO satisfy the following formula: b 01 =2ω0 b 02 =ω0 2 The linear extended state observer LESO also includes a linear state error feedback law LSEF, which generates a compensation signal according to the speed tracking error e1 and the observed disturbance z2. Its expression is: u0=β 11 e1=β 11 (x1-z1) In the formula, u0 is the original control quantity of the system, b is the compensation gain coefficient, β 11 is the error gain coefficient, x1 is the actual value of the parameter, and z1 is the estimated value of the parameter.

7. A resonance suppression system for a non-mechanical gear integrated vernier motor, characterized in that: include: Improved second-order generalized integrator-frequency-locked loop structure SOGI-FLL for online identification of resonant frequency; The active disturbance rejection controller LADRC includes a linear extended state observer LESO and a linear error feedback control law LSEF, which is used to generate a compensation signal; The current loop deadbeat predictive controller is used to receive the compensation current signal output by the LADRC and generate a corresponding voltage control signal.

8. The resonance suppression system of a non-mechanical gear integrated vernier motor according to claim 7, characterized in that: include: The inner rotor is mounted on the motor shaft and has P i For permanent magnets; The outer rotor is coaxially coupled with the outer rotor through the magnetic gear, and its surface is embedded with P o For permanent magnets, P o =P i +N, where N is the number of magnetic modulation poles of the magnetic modulation ring; The stator winding adopts fractional slot concentrated winding with a pole-slot ratio of 2P o :Q, Q is the number of slots.

Images