Active-disturbance-rejection control optimization system and method for stepping motor servo system

Abstract

The invention discloses an active-disturbance-rejection control optimization system and method for a stepping motor servo system, and relates to the technical field of stepping motor control, and the system comprises a signal processing unit which is used for carrying out the filtering processing of an input signal of the stepping motor servo system; the state observer unit is used for calculating the disturbance compensation amount; and the control algorithm unit is used for carrying out active disturbance rejection control optimization on the stepping motor servo system. According to the method, an active-disturbance-rejection control closed loop combining feed-forward compensation and feedback adjustment is effectively constructed, the control precision, robustness and dynamic performance of the stepping motor servo system are improved, and the method has industrial application prospects and technical popularization value.

Description

Technical Field

[0001] The present invention relates to the technical field of stepping motor control, and in particular to an auto-disturbance rejection control optimization system and method for a stepping motor servo system. Background Art

[0002] In recent years, with the rapid development of industrial automation and intelligent manufacturing, stepping motor servo systems have been increasingly widely used in the field of precision control, prompting control strategies to continuously develop towards higher precision and higher robustness. Traditional control methods, such as PID control and conventional sliding mode control, although they can meet general application requirements, have obvious deficiencies in control accuracy and anti-interference ability when facing complex disturbances, load fluctuations, and non-linear factors.

[0003] For this reason, the academic and industrial communities have begun to introduce auto-disturbance rejection control (ADRC) technology. By designing an extended state observer (ESO) to estimate system disturbances in real time and adopting a strategy combining feedforward compensation and feedback regulation to improve the system's dynamic response and steady-state performance. However, although the system performance has been improved to a certain extent, the existing solutions still have limitations in signal processing accuracy, disturbance estimation real-time performance, and control signal smoothness, and it is difficult to meet the requirements of high dynamic response and high robustness at the same time. There are also problems such as low real-time disturbance estimation accuracy and obvious chattering of control signals. Therefore, how to optimize the auto-disturbance rejection control for a stepping motor servo system has become an urgent problem to be solved. Summary of the Invention

[0004] In view of the problems existing in the existing auto-disturbance rejection control optimization system for a stepping motor servo system, the present invention is proposed. Therefore, the problem to be solved by the present invention is how to provide an auto-disturbance rejection control optimization system and method for a stepping motor servo system.

[0005] To solve the above technical problems, the present invention provides the following technical solutions:

[0006] In the first aspect, the present invention provides an auto-disturbance rejection control optimization system for a stepping motor servo system, which includes a signal processing unit including a digital filter. The digital filter filters the input signal of the stepping motor servo system and transmits the filtered input signal to the control algorithm unit;

[0007] The state observer unit includes an Extended State Observer (ESO), a neural network adaptive estimation module, and a disturbance compensation corrector. The ESO is used to expand the system estimated disturbance into system state variables and output the estimated disturbance value. The neural network adaptive estimation module is used to output a disturbance prediction value based on the historical estimated disturbance sequence and the state error. The disturbance compensation corrector generates a dynamic compensation coefficient according to the estimated disturbance value output by the ESO and the disturbance prediction value output by the neural network, and calculates the disturbance compensation amount.

[0008] The control algorithm unit includes a non-linear feedback control module, an adaptive gain regulator, and a dynamic compensator. The non-linear feedback control module uses an improved sliding mode control to smooth the output of the anti-disturbance torque. The adaptive gain regulator is used to dynamically adjust the sliding mode surface parameters and the switching gain. The dynamic compensator is used to receive the output of the disturbance compensation corrector and superimpose it on the control quantity to optimize the active disturbance rejection control for the stepper motor servo system.

[0009] As a preferred embodiment of the active disturbance rejection control optimization system for the stepper motor servo system according to the present invention, wherein: the stepper motor servo system includes the following:

[0010] The voltage equation of the stepper motor is obtained as:

[0011]

[0012] where: L0 is the winding inductance component; U A and U B are the winding voltages of phase A and phase B in the motor respectively; i A and i B are the winding currents of phase A and phase B respectively; L2 is the fundamental wave component; k e is the back electromotive force generated during the operation of the motor; r A and r B are the internal resistances of the windings of phase A and phase B respectively; w r is the rotor angular velocity; d is the axial coefficient of the motor; t is the state coefficient; is the winding power angle;

[0013] According to the voltage equation of the stepper motor, the torque equation of the stepper motor servo system is obtained as:

[0014]

[0015] where: T e is the output torque of the motor servo system, J is the system moment of inertia, β is the friction coefficient of the motor servo system, T L is the load torque during the operation of the motor, T M is the maximum static torque of the motor, is the power angle of the stepping motor rotor, is the power angle at the balanced position of the motor rotor;

[0016] Set the input magnetic energy of the stepping motor servo system to establish the mathematical dynamic model of the stepping motor servo system as:

[0017]

[0018] In the formula, γ SF is the output power of the motor servo system, and W is the input magnetic energy of the motor servo system;

[0019] According to the established model, set the state variables of the stepping motor servo system to obtain the extended state space characteristic equation of the stepping motor servo system:

[0020]

[0021] In the formula: x p1 , x p2 are the state variables of the motor servo system, y p is the output variable of the servo system; x p3 is the new state variable after the total disturbance of the system is expanded; c p , o p are the state variable coefficients.

[0022] As a preferred solution of the auto-disturbance rejection control optimization system for the stepping motor servo system described in the present invention, wherein: the digital filter works in series through an adaptive notch filter and a Kalman filter to filter high-frequency electromagnetic noise and low-frequency mechanical vibration interference respectively, and transmits the filtered signal to the control algorithm unit.

[0023] As a preferred solution of the auto-disturbance rejection control optimization system for the stepping motor servo system described in the present invention, wherein: the extended state observer ESO is based on the mathematical dynamic model of the stepping motor servo system, estimates the total disturbance of the system in real time, expands the disturbance into the system state variable, and outputs the estimated disturbance value;

[0024] During the design process of the extended state observer, set the state variable observation values as z p1 , z p2 , the total disturbance observation value is z p3 , and construct the operating equation of the servo system extended state observer as:

[0025]

[0026] In the formula: e p1 is the error between the observed position angle and the actual position angle of the extended state observation controller; e p2 is the error between the estimated rotational speed and the actual rotational speed; ep3 The error between the disturbance estimated by the observer and the actual disturbance; λ1 is the sliding mode switching function of the observer; Φ1 and Φ2 are the parameters of the observer sliding mode surface; q1 is the gain coefficient of the sliding mode switching vector; q2 is a constant;

[0027] The estimated disturbance value is output in real time, and at the same time, the state error is transmitted to the neural network module.

[0028] As a preferred solution of the auto-disturbance rejection control optimization system for the stepping motor servo system described in the present invention, wherein: the output of the disturbance prediction value based on the historical estimated disturbance sequence and the state error includes the following:

[0029] Obtain the disturbance estimation value from the extended state observer ESO as time series data, and at the same time, the system collects the state error;

[0030] The collected data is processed by normalization and denoising, and the processed historical disturbance sequence and state error information are integrated into a time series input vector;

[0031] Design an LSTM network, select ReLU as the activation function, and mean square error MSE as the loss function;

[0032] Use the historical disturbance estimation value data to perform offline training on the LSTM network, send the latest collected disturbance estimation value and state error data into the network, and update the network weights by the gradient descent method;

[0033] Based on the current input historical data sequence, the disturbance prediction value is output after being processed by the LSTM network.

[0034] As a preferred solution of the auto-disturbance rejection control optimization system for the stepping motor servo system described in the present invention, wherein: the generation of the dynamic compensation coefficient includes the following:

[0035] Predict the disturbance at the next moment based on the historical data LSTM network Perform difference calculation:

[0036]

[0037] In the formula, Δd is the difference calculation value, is the disturbance prediction value at the next moment, is the estimated disturbance value at time t;

[0038] Calculate the compensation judgment threshold according to the estimated disturbance value, and update the compensation coefficient. The compensation threshold is If |Δd| < δ, the compensation coefficient k c = 1;

[0039] When |Δd| ≥ δ, the proportional-integral strategy is adopted for adjustment, and the adjustment formula is:

[0040] k c = 1 + 0.5Δd + 0.1∫Δd dt

[0041] Calculate the final compensation amount to adjust the disturbance compensation amount in real time. The compensation amount calculation formula is:

[0042]

[0043] In the formula, Δu c is the disturbance compensation amount.

[0044] As a preferred solution of the active disturbance rejection control optimization system for the stepper motor servo system described in the present invention, wherein: the dynamic compensator receives the final compensation amount of the disturbance compensation corrector and superimposes it on the sliding mode control amount for control amount synthesis, so as to realize the active disturbance rejection control optimization for the stepper motor servo system.

[0045] In a second aspect, the present invention provides an active disturbance rejection control optimization method for a stepper motor servo system, which includes: digitally filtering the original input signal of the stepper motor servo system;

[0046] Based on the digitally filtered input signal, estimate and output the disturbance estimation value in real time, and output the disturbance prediction value based on the historical disturbance estimation value and the current system state error;

[0047] Calculate and generate a dynamic compensation coefficient according to the estimated disturbance value and the disturbance prediction value, determine the disturbance compensation amount to form a control signal;

[0048] Perform active disturbance rejection control for the stepper motor servo system according to the control signal.

[0049] In a third aspect, the present invention provides a computer device, including a memory and a processor, where: when the processor executes the computer program, the steps of the active disturbance rejection control optimization method for the stepper motor servo system are realized.

[0050] In a fourth aspect, the present invention provides a computer-readable storage medium, on which a computer program is stored, where: when the computer program is executed by the processor, the steps of the active disturbance rejection control optimization method for the stepper motor servo system are realized.

[0051] The present invention effectively constructs an active disturbance rejection control closed loop combining feedforward compensation and feedback regulation, not only realizes the comprehensive suppression of high-frequency noise, low-frequency vibration and load fluctuation, but also enables the control signal to have fast response characteristics while being smoothly output, thereby significantly improving the control accuracy, robustness and dynamic performance of the stepper motor servo system, and having industrial application prospects and technical promotion value. Brief Description of the Drawings

[0052] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0053] Figure 1 It is a structural diagram of an auto-disturbance rejection control optimization system for a stepping motor servo system. Detailed Embodiments

[0054] To make the above objects, features, and advantages of the present invention more comprehensible, the following will provide a detailed description of the specific embodiments of the present invention in conjunction with the drawings of the specification. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0055] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention can also be implemented in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0056] Secondly, the so-called "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that can be included in at least one implementation of the present invention. The appearances of "in one embodiment" in different places in this specification do not all refer to the same embodiment, nor are they separate or selectively exclusive embodiments from other embodiments.

[0057] Refer to Figure 1 , a kind of auto-disturbance rejection control optimization system for a stepping motor servo system is provided for the first embodiment of the present invention, including:

[0058] Before performing the composite auto-disturbance rejection control of the stepping motor servo system, it is necessary to clarify the structure of the mechanical drive servo system of the stepping motor and obtain the relevant state characteristic parameters of the motor servo system, and establish the dynamic model of the mechanical drive servo system of the stepping motor.

[0059] Based on the structure of the motor servo system, establish the dynamic state characteristic model of the stepping motor. Since the non-linear factors during the operation of the stepping motor are relatively serious, when establishing the dynamic state characteristic model of the stepping motor servo system, the effects of motor magnetic circuit saturation, hysteresis, and eddy current are ignored.

[0060] The motor stator windings are symmetric to each other and the electrical angle is 90°. The voltage equation of the stepper motor is obtained as follows:

[0061]

[0062]

[0063] Where: L0 is the winding inductance component; U A 、U B are the winding voltages of phase A and phase B in the motor respectively; i A 、i B are the winding currents of phase A and phase B respectively; L2 is the fundamental wave component; k e is the back electromotive force generated during the operation of the motor; r A 、r B are the internal resistances of the windings of phase A and phase B respectively; w r is the rotor angular velocity; d is the motor axial coefficient; t is the state coefficient; is the winding power angle.

[0064] According to the determined motor voltage equation above, the torque equation of the motor servo system is obtained:

[0065]

[0066] Where: T e is the output torque of the motor servo system, J is the system moment of inertia, β is the friction factor of the motor servo system, T L is the motor operating load torque, T M is the maximum static torque of the motor, is the stepper motor rotor power angle, is the power angle of the motor rotor balance position;

[0067] Since the stepper motor servo system mainly generates magnetomotive force through the windings, the input magnetic energy of the motor servo system is set as W, and the mathematical dynamic model of the servo system is established:

[0068]

[0069] Where, γ SF is the output power of the motor servo system, and W is the input magnetic energy of the motor servo system;

[0070] According to the established model, the state variables of the motor servo system are set to obtain the extended state space characteristic equation of the motor servo system:

[0071]

[0072] Where: x p1 、x p2is the state variable of the motor servo system, y p is the output variable of the servo system; x p3 is the new state variable after the total disturbance expansion of the system; c p and o p are the coefficients of the state variables.

[0073] The signal processing unit includes a digital filter, which filters the input signal of the stepper motor servo system and transmits the filtered input signal to the control algorithm unit;

[0074] The main task of the signal processing module is to filter and process various signals in the stepper motor servo system to remove high-frequency noise and low-frequency interference, ensuring that the system can obtain effective control signals. The module includes two main parts: an adaptive notch filter and a Kalman filter.

[0075] The adaptive notch filter is used to filter out the high-frequency electromagnetic noise generated in the motor system, especially the high-frequency electromagnetic interference that may occur during the motor drive process. By automatically adapting to environmental changes and adjusting its parameters, it can effectively remove the interference signals at specific frequencies.

[0076] The adaptive notch filter monitors the noise components in the input signal and adjusts its frequency response characteristics in real time to adapt to different interference sources. The filter targets high-frequency noise (such as power supply noise, switching noise, etc.) and suppresses the signals within these specific frequency ranges by setting the notch frequency.

[0077] The input signal comes from the encoder signal, current sensor signal, or voltage signal of the motor, etc. It usually contains electromagnetic interference, high-frequency noise, and other unnecessary high-frequency components to avoid high-frequency noise affecting the control accuracy and stability;

[0078] The Kalman filter is used to filter out low-frequency mechanical vibration interference, the low-frequency interference generated by the inherent vibration of the mechanical system and load fluctuations during the motor startup and operation process. The Kalman filter can dynamically filter out low-frequency disturbances through recursive estimation. It can effectively handle the low-frequency noise caused by mechanical vibration, friction, or load fluctuations in the system and can dynamically adjust the gain of the filter.

[0079] The input signal comes from the control signal or sensor signal of the motor. After being processed by the Kalman filter, the low-frequency vibration interference in the signal is suppressed;

[0080] The denoised signal is used as the input signal for the subsequent control algorithm and is transmitted to the control algorithm unit to ensure that the control system can make accurate adjustments, thereby improving the accuracy and stability of the motor servo system.

[0081] The state observer unit includes an Extended State Observer (ESO), a neural network adaptive estimation module, and a disturbance compensation corrector;

[0082] Specifically, the Extended State Observer (ESO) is based on the mathematical dynamic model of the stepper motor servo system, and real-time estimates the total system disturbance (including load fluctuations, friction nonlinearity, etc.), expands the disturbance into system state variables, and outputs the estimated disturbance value.

[0083] During the design process of the Extended State Observer, the state variable observation value is set as z p1 、z p2 The total disturbance observation value is z p3 The operating equation of the servo system Extended State Observer is constructed as:

[0084]

[0085] In the formula: e p1 is the error between the observed position angle and the actual position angle of the Extended State Observer controller; e p2 is the error between the estimated rotational speed and the actual rotational speed; e p3 is the error between the observed disturbance of the observer and the actual disturbance; λ1 is the sliding mode switching function of the observer; Φ1, Φ2 are the sliding mode surface parameters of the observer; q1 is the gain coefficient of the sliding mode switching vector; q2 is a constant.

[0086] The estimated disturbance value is output in real time, and at the same time, the state error is transmitted to the neural network module. The neural network generates a gain adjustment amount based on historical disturbances and current errors, and dynamically optimizes the ESO parameters.

[0087] The neural network adaptive estimation module realizes disturbance prediction by constructing a time series prediction model based on the LSTM network, including the following steps:

[0088] Obtain the disturbance estimation values within a period of time from the Extended State Observer (ESO) as time series data, and at the same time, the system collects the state error, which reflects the current deviation of the system from the target state.

[0089] Normalize and denoise the collected data to ensure the quality and scale consistency of the input data.

[0090] Integrate the preprocessed historical disturbance sequence and state error information into a time series input vector as the input sample.

[0091] Design an LSTM network, which includes one or more LSTM layers for capturing the time dependence of the data, and then a fully connected layer for outputting the disturbance prediction value.

[0092] Select ReLU as the activation function and the Mean Squared Error (MSE) as the loss function to measure the error between the predicted disturbance and the actual disturbance.

[0093] In the initial stage of the system, the LSTM network is offline trained using historical data so that it can initially master the time series characteristics of the disturbance.

[0094] During the operation of the system, the latest collected disturbance and state error data are sent into the network in real time, and the network weights are updated by the gradient descent method for adaptive learning.

[0095] Based on the current input historical data sequence, after being processed by the LSTM network, the module outputs the predicted disturbance value.

[0096] Disturbance compensation corrector: According to the difference between the output of the ESO and the disturbance predicted by the neural network, generate a dynamic compensation coefficient, adjust the disturbance compensation amount in real time, and the corrected compensation amount is input to the control algorithm unit;

[0097] By comparing the estimated disturbance value of the ESO with the predicted value of the neural network, the compensation amount is dynamically corrected to suppress the model mismatch error.

[0098] Based on historical data, the LSTM network predicts the disturbance at the next moment Perform difference calculation:

[0099]

[0100] Where, Δd is the difference calculation value, is the predicted disturbance at the next moment, is the estimated disturbance value output by the extended state observer at time t;

[0101] Calculate the compensation judgment threshold based on the estimated disturbance value, generate a compensation coefficient, and the compensation threshold is If |Δd| < δ, the compensation coefficient k c = 1;

[0102] If |Δd| ≥ δ, the proportional-integral (PI) strategy is adopted for adjustment, and the adjustment formula is:

[0103] k c = 1 + 0.5Δd + 0.1∫Δddt

[0104] Calculate the final compensation amount and directly superimpose it into the control amount to adjust the disturbance compensation amount in real time. The compensation amount calculation formula is:

[0105]

[0106] Where, Δu c is the disturbance compensation amount;

[0107] The control algorithm unit includes a non - linear feedback control module, an adaptive gain regulator, and a dynamic compensator. The non - linear feedback control unit suppresses disturbances through the strong robustness of sliding - mode control (SMC), and uses the hyperbolic tangent function to replace the traditional sign function to reduce chattering, achieving smooth control output.

[0108] Control law design: The control quantity consists of an equivalent control term u e and a switching control term u sw :

[0109] u s = T e + u sw

[0110] Switching control term: The hyperbolic tangent function is used to achieve smooth switching:

[0111]

[0112] where φ is the observer sliding - mode surface parameter.

[0113] Dynamically adjust the sliding - mode surface parameter φ and the sliding - mode switching vector gain coefficient q1 to adapt to the changes in load inertia and the intensity of external disturbances, ensuring the global stability of the system. Construct the energy function as:

[0114]

[0115] where q1 * is the ideal gain, and γ is the learning rate;

[0116] Gain update law: By ensuring Derive the adaptive law: Switching gain adaptation

[0117]

[0118] In the formula, is the updated sliding - mode switching vector gain coefficient;

[0119] Sliding - mode surface parameter adaptation:

[0120]

[0121] In the formula, is the updated sliding - mode surface parameter; η is the adaptation parameter;

[0122] The dynamic compensator receives the compensation signal from the disturbance compensation corrector, superimposes it on the sliding - mode control quantity, forming a composite disturbance - rejection structure of "feed - forward compensation + feedback suppression" to accelerate the disturbance - suppression process.

[0123] Perform the synthesis of the control quantity. The total control quantity is the linear superposition of the sliding - mode control quantity and the disturbance compensation quantity.

[0124] Furthermore, this embodiment also provides an active disturbance rejection control optimization method for a stepper motor servo system, including:

[0125] Digital filter the original input signal of the stepper motor servo system to generate a filtered clean signal, and transmit this signal to the state observer unit.

[0126] Utilize the filtered input signal and the system model to expand the actual disturbances in the system into state variables, and estimate and output the disturbance estimation value in real time.

[0127] Based on the historical disturbance estimation sequence and the current state error, adopt a time series prediction model such as LSTM to output the predicted value of future disturbances and capture the time series characteristics of the disturbances.

[0128] By comparing the disturbance estimation value of the ESO with the neural network predicted value, calculate and generate a dynamic compensation coefficient, and determine the disturbance compensation amount to be applied.

[0129] Based on the improved sliding mode control, generate a smooth disturbance rejection torque output according to the system error.

[0130] According to the real-time system state, error and load changes, dynamically adjust the sliding mode surface parameters and switching gain to ensure that the controller has sufficient robustness under different working conditions.

[0131] Superimpose the compensation signal generated by the disturbance compensation corrector and the disturbance rejection torque output by the non-linear feedback control to form a control input signal.

[0132] Transmit the control signal after dynamic compensation to the motor drive system to perform active disturbance rejection control optimization for the stepper motor servo system.

[0133] This embodiment also provides a computer device, which is applicable to the case of an active disturbance rejection control optimization system for a stepper motor servo system, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement all or part of the steps of the method described in the embodiment of the present invention as proposed in the above embodiment.

[0134] This embodiment also provides a storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the method in any optional implementation manner of the above embodiment. Among them, the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (abbreviated as SRAM), electrically erasable programmable read-only memory (abbreviated as EEPROM), erasable programmable read-only memory (abbreviated as EPROM), programmable read-only memory (abbreviated as PROM), read-only memory (abbreviated as ROM), magnetic memory, flash memory, a magnetic disk or an optical disc.

[0135] The storage medium proposed in this embodiment and the data storage method proposed in the above embodiment belong to the same inventive concept. For technical details not described in detail in this embodiment, reference can be made to the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.

[0136] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered by the scope of the claims of the present invention.

Claims

1. An auto-disturbance rejection control optimization system for a stepping motor servo system, characterized in that: Comprising: A signal processing unit, including a digital filter, which filters the input signal of the stepper motor servo system and transmits the filtered input signal to the control algorithm unit; A state observer unit, including an extended state observer ESO, a neural network adaptive estimation module, and a disturbance compensation corrector. The extended state observer ESO is used to expand the system estimated disturbance into system state variables and output the estimated disturbance value; The neural network adaptive estimation module is used to output a disturbance prediction value based on the historical estimated disturbance sequence and the state error; The disturbance compensation corrector generates a dynamic compensation coefficient according to the estimated disturbance value output by the extended state observer ESO and the disturbance prediction value output by the neural network, and calculates the disturbance compensation amount; A control algorithm unit, including a non-linear feedback control module, an adaptive gain regulator, and a dynamic compensator. The non-linear feedback control module uses an improved sliding mode control to smooth the anti-disturbance torque output; the adaptive gain regulator is used to dynamically adjust the sliding mode surface parameters and the switching gain; the dynamic compensator is used to receive the output of the disturbance compensation corrector and superimpose it on the control quantity to perform active disturbance rejection control optimization for the stepper motor servo system.

2. The auto-disturbance rejection control optimization system for the stepper motor servo system according to claim 1, wherein: The stepper motor servo system includes the following: The voltage equation of the stepper motor is obtained as: Where: L0 is the winding inductance component; U A and U B are the winding voltages of phase A and phase B in the motor respectively; i A and i B are the winding currents of phase A and phase B respectively; L2 is the fundamental wave component; k e is the back electromotive force generated during the operation of the motor; r A and r B are the internal resistances of the windings of phase A and phase B respectively; w r is the angular velocity of the rotor; d is the axial coefficient of the motor; t is the state coefficient; is the winding power angle; According to the voltage equation of the stepper motor, the torque equation of the stepper motor servo system is obtained as: Where: T e is the output torque of the motor servo system, J is the system moment of inertia, β is the friction coefficient of the motor servo system, T L is the load torque during motor operation, T M is the maximum static torque of the motor, is the power angle of the stepping motor rotor, is the power angle at the balanced position of the motor rotor; The input magnetic energy of the stepper motor servo system is set to establish the mathematical dynamic model of the stepper motor servo system as: where γ SF the motor servo system outputs power, and W is the input magnetic energy of the motor servo system; According to the established model, the state variables of the stepper motor servo system are set to obtain the extended state space characteristic equation of the stepper motor servo system: where: x p1 , x p2 are the state variables of the motor servo system, and y p is the output variable of the servo system; x p3 is the new state variable after the total disturbance expansion of the system; c p , o p are the state variable coefficients.

3. The auto-disturbance rejection control optimization system for the stepper motor servo system according to claim 2, wherein: The digital filter works in series with an adaptive notch filter and a Kalman filter to filter the high-frequency electromagnetic noise and low-frequency mechanical vibration interference respectively, and transmits the filtered signal to the control algorithm unit.

4. The auto-disturbance rejection control optimization system for the stepper motor servo system according to claim 3, characterized in that: The extended state observer ESO estimates the total system disturbance in real time based on the mathematical dynamic model of the stepper motor servo system, expands the disturbance into system state variables, and outputs the estimated disturbance value; During the design process of the extended state observer, the observed values of the state variables are set as z p1 , z p2 , and the observed value of the total disturbance is z p3 . The operating equation of the extended state observer for the servo system is constructed as follows: where: e p1 is the error between the observed position angle of the extended state observer controller and the actual position angle; e p2 is the error between the estimated rotational speed and the actual rotational speed; e p3 For the observer to estimate the disturbance and the actual disturbance error; λ1 is the sliding mode switching function of the observer; Φ1, Φ2 are the sliding mode surface parameters of the observer; q1 is the sliding mode switching vector gain coefficient; q2 is a constant; The estimated disturbance value is output in real time, and at the same time, the state error is transmitted to the neural network module.

5. The auto-disturbance rejection control optimization system for the stepper motor servo system according to claim 4, wherein: The output of the disturbance prediction value based on the historical estimated disturbance sequence and the state error includes the following: The disturbance estimation value is obtained from the extended state observer ESO as time series data, and at the same time, the system collects the state error; The collected data is processed by normalization and denoising, and the processed historical disturbance sequence and state error information are integrated into a time series input vector; An LSTM network is designed, ReLU is selected as the activation function, and the mean square error MSE is used as the loss function; The LSTM network is offline trained using the historical disturbance estimation value data, and the latest collected disturbance estimation value and state error data are sent into the network, and the network weights are updated by the gradient descent method; Based on the current input historical data sequence, the disturbance prediction value is output after being processed by the LSTM network.

6. The auto-disturbance rejection control optimization system for the stepper motor servo system according to claim 5, characterized in that: The generation of the dynamic compensation coefficient includes the following: Next - moment disturbance prediction based on historical data using LSTM network Perform difference calculation: where Δd is the difference calculation value, is the predicted disturbance value at the next moment, is the estimated disturbance value at time t; The compensation judgment threshold is calculated based on the estimated disturbance value, and the compensation coefficient is updated. The compensation threshold is When |Δd| < δ, the compensation coefficient k c = 1; When |Δd|≥δ, the proportional-integral strategy is adopted for adjustment, and the adjustment formula is: k c = 1 + 0.5Δd + 0.1∫Δd dt Calculate the final compensation amount to adjust the disturbance compensation amount in real time. The compensation amount calculation formula is: where Δu c is the disturbance compensation amount.

7. The auto-disturbance rejection control optimization system for a stepper motor servo system according to claim 6, wherein: The dynamic compensator receives the final compensation amount of the disturbance compensation corrector and superimposes it on the sliding mode control amount for control amount synthesis, so as to realize the active disturbance rejection control optimization for the stepper motor servo system.

8. An auto-disturbance rejection control optimization system for a stepper motor servo system, based on the auto-disturbance rejection control optimization method for a stepper motor servo system according to any one of claims 1 to 7, characterized in that: Including, Perform digital filtering on the original input signal of the stepper motor servo system; Based on the input signal after digital filtering, estimate and output the disturbance estimation value in real time, and output the disturbance prediction value based on the historical disturbance estimation value and the current system state error; Calculate and generate the dynamic compensation coefficient according to the estimated disturbance value and the disturbance prediction value, and determine the disturbance compensation amount to form the control signal; Perform the active disturbance rejection control for the stepper motor servo system according to the control signal.

9. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the active disturbance rejection control optimization method for the stepper motor servo system described in claim 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the active disturbance rejection control optimization method for the stepper motor servo system described in claim 8.

Images