Multi-source interference suppression method for piezoelectric driving micro-displacement system

Abstract

The invention relates to the technical field of precision control, and discloses a multi-source interference suppression method for a piezoelectric driving micro-displacement system, which comprises the following steps: S10, establishing a modeling interference model, and adaptively identifying a time-varying parameter: deeply analyzing a modeling interference source in the piezoelectric driving micro-displacement system, such as harmonic interference and the like; a generation mechanism and influence of the interference source are known, according to characteristics of the interference source, a corresponding mathematical model is established, a relation between the interference and system output is described, an adaptive algorithm is designed, time-varying parameters in an interference model are identified in real time, and it is ensured that the model can accurately reflect actual changes of the interference. The multi-source interference suppression method of the piezoelectric driving micro-displacement system aims to solve the problems that the piezoelectric driving micro-displacement system is inevitably influenced by various interference factors in actual operation, the dynamic performance and the steady-state precision of the system are reduced by the interference factors, and the instability and the maintenance cost of the system are increased.

Description

Technical Field

[0001] The invention relates to the field of precision control technology, and in particular to a multi-source interference suppression method for a piezoelectric driven micro-displacement system. Background Art

[0002] Piezoelectric ceramics, with their unique inverse piezoelectric effect, that is, the ability to produce mechanical deformation under the action of an electric field, have become an indispensable key material in the field of modern science and technology. Their nanometer-level resolution, ultra-fast response speed, high-precision displacement control and powerful driving force have brought revolutionary breakthroughs in many precision control fields.

[0003] In the actual application process, the piezoelectric driven micro-displacement system inevitably faces a series of complex multi-source interference challenges. These interference factors significantly restrict the dynamic performance and steady-state accuracy of the system. Harmonic interference mainly comes from the nonlinear waveform distortion produced by power electronic equipment such as inverters and rectifiers, which not only affects the power transmission efficiency, but also may cause overheating, vibration and noise of internal components of the system, accelerate equipment aging, and increase maintenance costs. At the same time, the common flexible structures in the system are prone to low-frequency elastic vibrations when stimulated. Vibrations that are difficult to attenuate for a long time may cause resonance, posing a serious threat to the stability and accuracy of the system. In addition, the excitation of high-frequency flexible modes increases the complexity of the dynamic response of the system, and the piezoelectric ceramic driver itself The hysteresis nonlinear characteristics further limit the improvement of its driving accuracy. External load disturbances, including uncertain factors such as environmental changes and external forces, also directly act on the system, reducing its control performance. For the above-mentioned multi-source interference problems, the existing technology shows certain limitations. First, since the system involves multiple complex factors such as flexible structure, nonlinear vibration and hysteresis characteristics, accurate modeling becomes extremely difficult, and traditional control methods are difficult to deal with comprehensively. Secondly, control strategies are mostly limited to the processing of a single interference factor and lack of comprehensive solutions, which limits the improvement of control effects. Furthermore, in application scenarios with high real-time requirements, existing control methods are often unable to respond quickly to changes in external interference, resulting in a decline in system performance. Summary of the invention

[0004] 1. Technical issues to be resolved

[0005] The purpose of the present invention is to solve the problem that a piezoelectric driven micro-displacement system is inevitably affected by multiple interference factors in actual operation. These interferences not only reduce the dynamic performance and steady-state accuracy of the system, but also increase the instability and maintenance cost of the system. Therefore, a multi-source interference suppression method for a piezoelectric driven micro-displacement system is proposed.

[0006] (II) Technical solution

[0007] The technical solution of the present invention to solve the above technical problems is as follows:

[0008] A method for suppressing multi-source interference of a piezoelectric driven micro-displacement system comprises the following steps:

[0009] S10. Establish a model of modelable interference and adaptively identify time-varying parameters: Conduct in-depth analysis of modelable interference sources in the piezoelectric drive micro-displacement system, such as harmonic interference, to understand its generation mechanism and impact. According to the characteristics of the interference source, establish a corresponding mathematical model to describe the relationship between interference and system output, design an adaptive algorithm, and identify the time-varying parameters in the interference model in real time to ensure that the model can accurately reflect the actual changes in interference.

[0010] S20. Apply the internal model-based interference suppression method to suppress the modelable interference: Based on the established interference model, design an internal model controller that can produce a control effect opposite to the interference, thereby offsetting the impact of the interference on the system, and adjust the parameters of the internal model controller to ensure that its interference suppression effect is optimal without affecting the stability and response speed of the system;

[0011] S30. Approximation and compensation of unmodeled dynamics based on neural networks: For dynamic behaviors that are difficult to model or unknown in the system, neural networks are used for approximation. Neural networks have powerful nonlinear mapping capabilities and can learn and simulate complex dynamic relationships. The neural network model is trained to accurately predict the behavior of unmodeled dynamics and incorporate it into the control strategy of the system to improve the accuracy and robustness of control.

[0012] S40. Interference estimation and compensation based on nonlinear interference observer: To achieve accurate compensation of external interference: Design a nonlinear interference observer to observe and estimate the external interference to the system in real time. The observer can use the system output and input data to accurately estimate the size and change trend of the interference, and design a compensation strategy based on the data model to incorporate the observed external interference into the control strategy. By generating corresponding compensating control effects, the impact of external interference on the system can be accurately offset.

[0013] Based on the above technical solution, the present invention can also be improved as follows.

[0014] Furthermore, the step S10 specifically includes the following steps:

[0015] S101. Analysis and identification of interference sources. Identify all possible interference sources in the system, including but not limited to mechanical vibration, electromagnetic interference, power supply fluctuations, performance changes caused by temperature changes, and harmonic interference. In-depth study of the generation mechanism, frequency characteristics, amplitude variation range of each interference and its specific impact on system performance. For harmonic interference, special attention should be paid to the stability of its frequency component, phase and amplitude.

[0016] S102, establish mathematical model, interference model construction: according to the characteristic analysis of interference source, select appropriate mathematical model to describe interference. For harmonic interference, it can be represented by superposition of sine wave or cosine wave, and it is necessary to identify the amplitude, frequency and phase of harmonic system dynamic model. In addition to the interference model, it is also necessary to establish the dynamic model of the system itself, which usually includes the transfer function or state space model of piezoelectric actuator to describe the relationship between system input such as control voltage and output such as displacement. Combine the system model and interference model to form a comprehensive model including interference influence for subsequent analysis and control design;

[0017] S103, adaptive algorithm design, design an adaptive internal model to estimate the time-varying parameters such as amplitude, frequency, and phase in the interference model in real time, consider the convergence speed, stability, computational complexity and other factors of the algorithm, optimize the algorithm, and ensure that the changes of interference parameters can be tracked quickly and accurately in practical applications;

[0018] S104. Real-time implementation and verification. Combining the adaptive internal model algorithm with the hardware and software platform of the piezoelectric drive micro-displacement system to achieve real-time estimation and update of the disturbance parameters. A series of experiments are conducted to verify the effectiveness of the established model and adaptive algorithm, including testing the performance of the system under different disturbances, observing whether the model can accurately reflect the actual changes in the disturbance, and evaluating its effect on improving the control accuracy and stability of the system.

[0019] S105. Feedback and Adjustment. Based on the experimental results, evaluate the performance of the model and algorithm, identify potential areas for improvement, and make necessary adjustments and optimizations to the model and algorithm based on the evaluation results to improve their adaptability and accuracy.

[0020] Furthermore, when there are multiple types of interference in the system, these interferences may be coupled with each other and affect the accurate estimation of parameters. The multi-source interference decoupling technology can be used to separate the impact of different interference sources on the system, thereby improving the accuracy of each interference parameter estimation. Different interference sources are identified and separated by blind source separation, independent component analysis, etc. At the same time, during the operation of the system, possible faults are monitored and diagnosed in real time so that timely measures can be taken to repair or isolate the source of the fault. Fault tree analysis, expert system or data-based fault diagnosis methods such as anomaly detection and pattern recognition are used to identify abnormal conditions in the system. As the system operating environment changes or the system itself ages, the original model may no longer be applicable. Through online learning or incremental learning technology, the model parameters can be updated in real time to keep an accurate description of the system status at all times. Sliding window technology, online support vector machine or incremental neural network and other methods are used to update the model parameters.

[0021] Furthermore, the step S20 specifically includes the following steps:

[0022] S201, internal model controller design, integrating the interference model established in S102 into the design of the internal model controller, which includes taking into account the dynamic characteristics of the interference such as the frequency, phase and amplitude of the harmonic interference, designing the core part of the internal model controller, including the internal model, the internal model controller and the filter. The internal model is used to predict the output of the controlled object and correct it. The internal model controller adjusts the control quantity to track the given value of the control system and offset the interference. The filter is used to improve the robustness of the control system. According to the interference model and the system dynamic model, a suitable controller transfer function is designed so that it can maintain the stability and performance of the system in the presence of interference;

[0023] S202, parameter adjustment and optimization, according to the system characteristics and interference model, preliminarily set the parameters of the internal model controller, such as the time constant of the filter, the gain of the controller, etc., use MATLAB / Simulink to perform simulation experiments, verify the performance of the internal model controller in the presence of interference, and evaluate the interference suppression effect of the controller and the stability and response speed of the system through the simulation results. According to the simulation results, adjust and optimize the parameters of the controller, and gradually find the best parameter combination through the iterative process, so that the controller can suppress interference while maintaining the stability and rapid response capability of the system;

[0024] S203, real-time implementation and verification, combining the optimized internal model controller with the hardware and software platform of the system to realize real-time control, which includes writing control algorithms, configuring hardware interfaces and debugging software systems, etc., conducting field tests in the actual system to verify the performance of the internal model controller under actual working conditions, observing the suppression effect of the controller and the stability of the system by changing the operating conditions of the system or introducing different interference sources, and evaluating the performance of the controller based on the field test results. If it is found that the control effect is not ideal or the system is unstable, it is necessary to further adjust and optimize the parameters of the controller;

[0025] S204. Documentation and feedback: record the design process, parameter adjustment results and field test results of the internal model controller, write detailed technical documents and user manuals to provide reference for subsequent maintenance and upgrades, establish a feedback mechanism, collect user feedback and opinions on the use of the controller, and continuously improve and optimize the controller based on the feedback results to meet the different needs and application scenarios of users.

[0026] Furthermore, in addition to optimizing specific interference models, it is also necessary to verify the robustness of the controller to other types of interference or model uncertainties, design fault-tolerant control strategies to cope with system component failures or parameter changes, and ensure that the controller can still maintain a certain control performance in the event of a failure. According to the needs of real-time control systems, the internal model control algorithm is optimized to reduce computational complexity and improve computational efficiency. Consider using high-performance hardware platforms such as FPGA, DSP, etc. to implement the control algorithm to further improve the real-time performance of the system.

[0027] Furthermore, the step S30 specifically includes the following steps:

[0028] S301, neural network selection and construction, according to the complexity of the unmodeled dynamics and the specific needs of the system, select the appropriate neural network type, where the neural network type is RBF neural network, determine the number of nodes in the input layer, hidden layer and output layer of the neural network, as well as the connection method and activation function between the layers, the input layer contains the state and control variables of the system, and the output layer outputs the predicted value of the unmodeled dynamics;

[0029] S302, data preparation and preprocessing, collecting data containing unmodeled dynamic behaviors from the system, which can be obtained through experiments, simulations or actual operations, and cleaning, denoising and normalizing the collected data to improve the effect and convergence speed of neural network training;

[0030] S303, neural network training, dividing the preprocessed data into a training set, a validation set and a test set, the training set is used to train the neural network model, the validation set is used to adjust the model parameters to prevent overfitting, and the test set is used to evaluate the generalization ability of the model. The neural network is trained using the training set data, and the model is enabled to accurately predict the behavior of unmodeled dynamics by adjusting the center and basis width of the basis function of the RBF neural network. During the training process, a parallel learning algorithm is used to accelerate convergence and improve the training effect. The trained neural network model is evaluated using the validation set to check its prediction accuracy and generalization ability. If the model performance is poor, the model structure or training parameters need to be adjusted and retrained;

[0031] S304, neural network integration and control strategy optimization, integrating the trained neural network model into the control strategy of the system, adjusting the parameters or output of the controller according to the prediction results of the neural network to compensate for the impact of unmodeled dynamics on system performance, optimizing the control strategy of the system based on the integrated neural network model, and verifying the performance of the optimized control strategy through simulation or experiment to ensure that it can improve the control accuracy and robustness of the system;

[0032] S305, real-time implementation and verification, combining the optimized control strategy with the system's hardware and software platform to achieve real-time control, which includes writing control algorithms, configuring hardware interfaces, and debugging software systems, etc. Field tests are conducted in actual systems to verify the approximation effect of the RBF neural network model on unmodeled dynamics and the performance of the optimized control strategy. By changing the system's operating conditions or introducing different interference sources, the control effect and stability of the system are observed. Based on the field test results, the performance of the neural network model and control strategy is evaluated. If the control effect is found to be unsatisfactory or the system becomes unstable, the neural network model and control strategy need to be further adjusted and optimized.

[0033] Furthermore, during the operation of the system, the prediction accuracy and stability of the neural network model are monitored in real time, possible performance degradation problems are discovered and handled in a timely manner, and safety boundaries are set for the prediction results of the neural network model to ensure that even in the worst case, the output of the system does not exceed an acceptable safety range. At the same time, the neural network model is regularly maintained and updated, including data retraining, model structure optimization, and software platform upgrades, to ensure that it always maintains optimal performance.

[0034] Furthermore, the step S40 specifically includes the following steps:

[0035] S401. Design a nonlinear disturbance observer. According to the characteristics of the system and the characteristics of the external disturbance, select a suitable nonlinear disturbance observer. Such an observer usually has a strong nonlinear processing capability and can accurately capture and estimate the changes of the external disturbance. Based on the input and output data of the system and possible prior knowledge, establish a mathematical model of the nonlinear disturbance observer. The model should be able to reflect the relationship between the system output and the external disturbance and have the ability to be updated in real time. Adjust and optimize the parameters of the observer through simulation or experimental data to ensure that it can accurately estimate the size and change trend of the external disturbance.

[0036] S402, real-time observation and estimation of external disturbances, real-time collection of system input and output data, including system state variables measured by sensors and control signals output by controllers, etc., use nonlinear disturbance observers to process the collected data, and estimate the external disturbances to the system in real time. This step is the basis for accurate compensation, and its accuracy directly affects the effect of subsequent compensation strategies;

[0037] S403. Design a compensation strategy in combination with the data model. Establish a data model of the system based on the physical principles, empirical formulas or machine learning models of the system. The model should be able to describe the dynamic behavior of the system and the impact of external disturbances on the system. Design a corresponding compensation strategy based on the external disturbance estimated by the nonlinear disturbance observer and the data model. The compensation strategy should be able to produce a control effect opposite to the external disturbance to accurately offset its impact on the system.

[0038] S404, realize accurate compensation, embed the designed compensation strategy into the control system, generate corresponding compensation control signal in real time, the signal should be able to respond quickly to the change of external disturbance and effectively offset its influence on system performance, verify the effectiveness of compensation strategy through simulation or experiment, evaluate the performance of system under external disturbance, including stability, tracking accuracy and robustness, etc.;

[0039] S405. Continuous optimization and adjustment, establishment of feedback mechanism, continuous optimization and adjustment of nonlinear disturbance observer and compensation strategy according to system performance evaluation results, including adjustment of observer parameters, optimization of compensation strategy algorithm and improvement of data model, etc., to ensure that the system can adapt to different types of external disturbances and changes in working environment, and maintain its precise compensation capability and good performance.

[0040] Furthermore, the real-time monitoring system's performance indicators, such as response time, control accuracy, resource consumption, etc., ensure that the system can maintain stable performance under high load or complex environments. When it is found that the performance indicators have declined, the reasons are analyzed in time and corresponding measures are taken to make adjustments.

[0041] (III) Beneficial effects

[0042] Compared with the prior art, the technical solution of this application has the following beneficial technical effects:

[0043] By establishing a model of modelable interference and adaptively identifying time-varying parameters, this method can deeply understand and quantify the characteristics of the main modelable interference sources in the system, such as harmonic interference, and update the model parameters in real time to accurately reflect the actual changes in interference. This adaptive modeling and identification mechanism effectively overcomes the limitations of traditional modeling methods that are difficult to deal with time-varying interference, improves the accuracy and dynamic adaptability of the model's description of interference, and provides a solid foundation for subsequent interference suppression. Secondly, the internal model-based interference suppression method is applied. By designing an internal model controller to produce a control effect opposite to the interference, it achieves accurate offset of the modelable interference. The introduction of the internal model controller not only enhances the system's ability to suppress interference, but also ensures a good balance between the interference suppression effect and the system stability and response speed through parameter adjustment. This strategy significantly improves the control accuracy and stability of the system when facing known interference. Furthermore, the approximation and compensation of unmodeled dynamics based on neural networks provide an effective means to deal with complex dynamic behaviors in the system that are difficult to directly model. With its powerful nonlinear mapping ability, neural networks can learn and simulate complex dynamic relationships and incorporate them into control strategies. This not only expands the scope of application of control strategies, but also improves the robustness and control accuracy of the system when facing unknown or difficult-to-model disturbances. Finally, interference estimation and compensation based on nonlinear disturbance observers: accurate compensation for external disturbances is achieved. Nonlinear disturbance observers can observe and estimate external disturbances in real time. The compensation strategy designed in combination with data models can accurately offset the impact of external disturbances on the system. This strategy that combines observation and compensation further enhances the system's resistance to external uncertainties and interference factors, and ensures the stability and reliability of the system in high-precision micro-operation tasks. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 It is an overall method flow chart of a multi-source interference suppression method for a piezoelectric driven micro-displacement system of the present invention;

[0045] Figure 2 The present invention is an overall block diagram of a method for suppressing multi-source interference of a piezoelectric driven micro-displacement system. DETAILED DESCRIPTION

[0046] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0047] Combination Figure 1-Figure 2 As shown, a multi-source interference suppression method for a piezoelectric driven micro-displacement system of the present invention is characterized by comprising the following steps:

[0048] S10. Establish a model of modelable interference and adaptively identify time-varying parameters: Conduct in-depth analysis of modelable interference sources in the piezoelectric drive micro-displacement system, such as harmonic interference, to understand its generation mechanism and impact. According to the characteristics of the interference source, establish a corresponding mathematical model to describe the relationship between interference and system output, design an adaptive algorithm, and identify the time-varying parameters in the interference model in real time to ensure that the model can accurately reflect the actual changes in interference.

[0049] Step S10 specifically includes the following steps:

[0050] S101. Analysis and identification of interference sources. Identify all possible interference sources in the system, including but not limited to mechanical vibration, electromagnetic interference, power supply fluctuations, performance changes caused by temperature changes, and harmonic interference. In-depth study of the generation mechanism, frequency characteristics, amplitude variation range of each interference and its specific impact on system performance. For harmonic interference, special attention should be paid to the stability of its frequency component, phase and amplitude.

[0051] S102, establish mathematical model, interference model construction: according to the characteristic analysis of interference source, select appropriate mathematical model to describe interference. For harmonic interference, it can be represented by superposition of sine wave or cosine wave, and it is necessary to identify the amplitude, frequency and phase of harmonic system dynamic model. In addition to the interference model, it is also necessary to establish the dynamic model of the system itself, which usually includes the transfer function or state space model of piezoelectric actuator to describe the relationship between system input such as control voltage and output such as displacement. Combine the system model and interference model to form a comprehensive model including interference influence for subsequent analysis and control design;

[0052] S103, adaptive algorithm design, design an adaptive internal model to estimate the time-varying parameters such as amplitude, frequency, and phase in the interference model in real time, consider the convergence speed, stability, computational complexity and other factors of the algorithm, optimize the algorithm, and ensure that the changes of interference parameters can be tracked quickly and accurately in practical applications;

[0053] S104. Real-time implementation and verification. Combining the adaptive internal model algorithm with the hardware and software platform of the piezoelectric drive micro-displacement system to achieve real-time estimation and update of the disturbance parameters. A series of experiments are conducted to verify the effectiveness of the established model and adaptive algorithm, including testing the performance of the system under different disturbances, observing whether the model can accurately reflect the actual changes in the disturbance, and evaluating its effect on improving the control accuracy and stability of the system.

[0054] S105. Feedback and Adjustment: Based on the experimental results, evaluate the performance of the model and algorithm, identify potential areas for improvement, and make necessary adjustments and optimizations to the model and algorithm based on the evaluation results to improve their adaptability and accuracy.

[0055] In step S10, a systematic analysis and suppression process is proposed for the complex multi-source interference problem faced by the piezoelectric driven micro-displacement system. First, through the interference source analysis and identification in step S101, various interference sources that may exist in the system are comprehensively sorted out and clarified, such as mechanical vibration, electromagnetic interference, power supply fluctuations, performance changes caused by temperature changes, and harmonic interference. This step provides a clear goal and direction for the subsequent model establishment and suppression strategy design. Then, step S102 establishes a mathematical model. In particular, for modelable interference sources such as harmonic interference, the superposition form of sine waves or cosine waves is used to describe its characteristics, and combined with the dynamic model of the system itself, a comprehensive model including the influence of interference is formed. This model not only reflects the basic relationship between the input and output of the system, but also fully considers the impact of interference on system performance, providing a precise mathematical basis for the subsequent control strategy design. In step S103, an adaptive algorithm is designed to realize real-time Estimate the time-varying parameters in the interference model. These algorithms, such as the least squares method and the Kalman filter, can dynamically adjust the model parameters according to the actual operating data of the system to accurately track the changes in interference. This step ensures the effectiveness and accuracy of the model in practical applications. Step S104 combines the adaptive algorithm with the hardware and software platform of the system to achieve real-time estimation and updating of interference parameters. Through a series of experimental verifications, the effectiveness of the model and algorithm is evaluated, and their effects on improving the control accuracy and stability of the system are observed. This step is a bridge between theoretical design and practical application, ensuring the feasibility and practicality of the proposed method. Finally, step S105 provides feedback and adjustments based on the experimental results. By evaluating the performance of the model and algorithm, potential room for improvement is identified, and necessary adjustments and optimizations are made accordingly. This step is an iterative process aimed at continuously improving the adaptability and accuracy of the system to better cope with complex and changing interference environments.

[0056] The implementation of steps S101 to S105 provides a comprehensive and effective multi-source interference suppression solution for the piezoelectric driven micro-displacement system. First, through in-depth analysis and identification of interference sources, the types and characteristics of interference faced by the system are clarified, which provides a scientific basis for the subsequent model establishment and suppression strategy design. Secondly, by establishing a comprehensive model including the impact of interference and designing an adaptive algorithm to estimate the interference parameters in real time, accurate description and tracking of interference are achieved, which not only improves the system's sensitivity and response speed to interference, but also enhances its adaptability and robustness. In addition, through experimental verification and feedback adjustment mechanism, the performance of the model and algorithm is continuously optimized to ensure the effectiveness and reliability of the proposed method in practical applications. This process not only improves the control accuracy and stability of the system, but also reduces the errors and failure rates caused by interference, thereby extending the service life of the system and improving its overall performance. Finally, this solution provides strong technical support for piezoelectric driven micro-displacement systems in application scenarios with high precision and high stability requirements.

[0057] When there are multiple types of interference in the system, these interferences may be coupled with each other and affect the accurate estimation of parameters. The multi-source interference decoupling technology can be used to separate the impact of different interference sources on the system, thereby improving the accuracy of each interference parameter estimation. Different interference sources are identified and separated by blind source separation and independent component analysis. At the same time, during the operation of the system, possible faults are monitored and diagnosed in real time so that timely measures can be taken to repair or isolate the source of the fault. Fault tree analysis, expert system or data-based fault diagnosis methods such as anomaly detection and pattern recognition are used to identify abnormal conditions in the system. With the changes in the system operating environment or the aging of the system itself, the original model may no longer be applicable. Through online learning or incremental learning technology, the model parameters can be updated in real time to keep an accurate description of the system status at all times. Sliding window technology, online support vector machine or incremental neural network and other methods are used to update the model parameters.

[0058] In a piezoelectric driven micro-displacement system, when multiple types of interference coexist and couple with each other, the traditional single interference suppression method often fails to achieve the desired effect. In order to solve this problem, the dependent claims propose a solution of using multi-source interference decoupling technology. This technology can identify and separate the influence of different interference sources on the system through advanced signal processing methods such as blind source separation and independent component analysis, thereby achieving decoupling of interference signals. This process helps to eliminate the mutual influence between interferences and improve the accuracy of each interference parameter estimation. At the same time, in order to ensure the stable operation of the system, the dependent claims also emphasize the importance of real-time monitoring and fault diagnosis. Through fault tree analysis, expert system or data-based fault diagnosis methods (such as anomaly detection, modeling, etc.), the system can effectively eliminate the influence of interference on the system. The system can monitor and identify abnormal conditions in operation, such as performance degradation, parameter deviation, etc. in real time. Once signs of failure are found, the system can quickly take measures to repair or isolate the source of the failure to prevent the failure from spreading and causing more serious damage to the system. In addition, considering the changes in the system operating environment or the aging of the system itself, the original model may gradually lose its accuracy. In order to address this problem, the dependent claims propose online learning or incremental learning technologies, which allow the system to update model parameters in real time during operation to reflect the latest changes in the system status. Through methods such as sliding window technology, online support vector machines or incremental neural networks, the system can continuously learn and optimize its internal model to ensure its accurate description of the system status.

[0059] The use of multi-source interference decoupling technology can significantly improve the accuracy of interference parameter estimation in piezoelectric driven micro-displacement systems. By separating the influence of different interference sources, the coupling effect between interferences is eliminated, so that each interference parameter can be estimated more accurately. This not only improves the control accuracy and stability of the system, but also enhances the system's adaptability to complex interference environments. The introduction of real-time monitoring and fault diagnosis mechanisms provides a strong guarantee for the stable operation of the system. By promptly detecting and handling abnormal situations, the system can avoid the occurrence of faults or reduce the impact of faults on system performance. This not only improves the reliability and availability of the system, but also reduces maintenance costs and downtime. The application of online learning or incremental learning technology enables the system to continuously adapt to changes in the operating environment and the aging process of the system. By updating the model parameters in real time, the system can always maintain an accurate description of the system status, thereby ensuring the effectiveness and accuracy of the control strategy. This adaptive capability is of great significance for improving the long-term operating performance of the system and extending the system life.

[0060] S20. Apply the internal model-based interference suppression method to suppress the modelable interference: Based on the established interference model, design an internal model controller that can produce a control effect opposite to the interference, thereby offsetting the impact of the interference on the system, and adjust the parameters of the internal model controller to ensure that its interference suppression effect is optimal without affecting the stability and response speed of the system;

[0061] Step S20 specifically includes the following steps:

[0062] S201, internal model controller design, integrating the interference model established in S102 into the design of the internal model controller, which includes taking into account the dynamic characteristics of the interference such as the frequency, phase and amplitude of the harmonic interference, designing the core part of the internal model controller, including the internal model, the internal model controller and the filter. The internal model is used to predict the output of the controlled object and correct it. The internal model controller adjusts the control quantity to track the given value of the control system and offset the interference. The filter is used to improve the robustness of the control system. According to the interference model and the system dynamic model, a suitable controller transfer function is designed so that it can maintain the stability and performance of the system in the presence of interference;

[0063] S202, parameter adjustment and optimization, according to the system characteristics and interference model, preliminarily set the parameters of the internal model controller, such as the time constant of the filter, the gain of the controller, etc., use MATLAB / Simulink to perform simulation experiments, verify the performance of the internal model controller in the presence of interference, and evaluate the interference suppression effect of the controller and the stability and response speed of the system through the simulation results. According to the simulation results, adjust and optimize the parameters of the controller, and gradually find the best parameter combination through the iterative process, so that the controller can suppress interference while maintaining the stability and rapid response capability of the system;

[0064] S203, real-time implementation and verification, combining the optimized internal model controller with the hardware and software platform of the system to realize real-time control, which includes writing control algorithms, configuring hardware interfaces and debugging software systems, etc., conducting field tests in the actual system to verify the performance of the internal model controller under actual working conditions, observing the suppression effect of the controller and the stability of the system by changing the operating conditions of the system or introducing different interference sources, and evaluating the performance of the controller based on the field test results. If it is found that the control effect is not ideal or the system is unstable, it is necessary to further adjust and optimize the parameters of the controller;

[0065] S204. Documentation and feedback: record the design process, parameter adjustment results and field test results of the internal model controller, write detailed technical documents and user manuals to provide reference for subsequent maintenance and upgrades, establish a feedback mechanism, collect user feedback and opinions on the use of the controller, and continuously improve and optimize the controller based on the feedback results to meet the different needs and application scenarios of users.

[0066] In step S20, in view of the interference problem faced by the piezoelectric driven micro-displacement system, an internal model controller is designed and its parameters are optimized to achieve effective suppression of interference and significant improvement of system performance. First, in step S201, the system model containing interference characteristics established in S102 is integrated into the design of the internal model controller. The internal model controller predicts the output of the controlled object through its internal model, and uses the internal model controller and the filter to adjust the control quantity to track the given value and offset the interference. This design method fully considers the dynamic characteristics of the interference, such as the frequency, phase and amplitude of the harmonic interference, so that the controller can more accurately predict and respond to the interference, thereby improving the stability and performance of the system. Subsequently, in step S202, the performance of the internal model controller in the presence of interference is verified by preliminarily setting the controller parameters and conducting simulation experiments using MATLAB / Simulink. The simulation results not only show the interference suppression effect of the controller, but also provide an evaluation of the system stability and response speed. Based on these evaluation results, the controller parameters are iteratively adjusted and optimized to gradually find the best parameter combination to ensure that the controller can effectively suppress interference while maintaining the stability and rapid response capability of the system. In step S203, the optimized internal model controller is combined with the hardware and software platform of the system to achieve real-time control. This step includes writing control algorithms, configuring hardware interfaces, and debugging software systems to ensure that the controller can run stably in the actual system. Through field tests, the performance of the controller under different working conditions and interference sources is verified, and its suppression effect and system stability are further evaluated. Finally, in step S204, the design process of the controller, parameter adjustment results, and field test results are recorded, and detailed technical documents and user manuals are written. These documents not only provide references for subsequent maintenance and upgrades, but also establish a feedback mechanism to collect user feedback and opinions. Based on these feedback results, the controller can be continuously improved and optimized to meet the needs of different users and application scenarios.

[0067] By implementing the various measures in step S20, the performance of the piezoelectric driven micro-displacement system has been significantly improved. First, the design of the internal model controller fully considers the dynamic characteristics of the interference, so that the controller can more accurately predict and respond to interference, thereby significantly improving the stability and anti-interference ability of the system. Secondly, the controller parameters are optimized and adjusted through simulation experiments and field tests to ensure that the controller can maintain good control performance under various working conditions. This parameter optimization process not only improves the response speed and control accuracy of the system, but also reduces the system's sensitivity to parameter changes and enhances the robustness of the system. In addition, the documentation of the design process and test results provides an important reference for subsequent maintenance and upgrades, which helps to reduce maintenance costs and improve the maintainability of the system. At the same time, the established feedback mechanism enables the controller to continuously receive user feedback and make improvements and optimizations, so as to continuously meet the user's new needs and changes in application scenarios. This continuous improvement process helps to maintain the advancement and competitiveness of the controller and provides strong support for the widespread application of piezoelectric driven micro-displacement systems.

[0068] In addition to optimizing specific disturbance models, it is also necessary to verify the robustness of the controller to other types of disturbances or model uncertainties, and to design fault-tolerant control strategies to cope with situations such as system component failures or parameter changes, to ensure that the controller can still maintain a certain control performance in the event of a failure. In response to the needs of real-time control systems, the internal model control algorithm is optimized to reduce computational complexity and improve computational efficiency. Consider using high-performance hardware platforms such as FPGA and DSP to implement the control algorithm to further improve the real-time performance of the system.

[0069] In the process of further improving the control performance of the piezoelectric driven micro-displacement system, the dependent claims emphasize several key aspects, aiming to enhance the robustness, fault tolerance and real-time performance of the controller. First, in order to verify the robustness of the controller to other types of interference or model uncertainty, different types of interference and model changes are introduced through simulation or experiment, and the response of the controller is observed and recorded. This process not only evaluates the performance of the controller under known interference, but also explores its performance in unknown or complex interference environments, ensuring that the controller has wide adaptability and stability. Secondly, the fault-tolerant control strategy is designed to deal with emergencies such as system component failures or parameter changes. These strategies usually include fault detection, isolation and recovery mechanisms, which can quickly adjust the control strategy when a fault is detected to avoid the fault from causing serious impact on the overall performance of the system. Through pre-set fault-tolerant logic and algorithms, the controller can Maintaining a certain control performance and ensuring the continuous operation and safety of the system, it is also crucial to optimize the internal model control algorithm in response to the needs of real-time control systems. By improving the algorithm structure, reducing unnecessary calculation steps or adopting more efficient numerical methods, the computational complexity of the control algorithm can be reduced and the computational efficiency can be improved. This optimization enables the controller to respond to system changes more quickly, reduce delays and errors, and thus improve the real-time performance of the system. Finally, considering using high-performance hardware platforms such as FPGA (field programmable gate array) or DSP (digital signal processor) to implement the control algorithm, it can further accelerate the calculation process and reduce execution time. These hardware platforms have powerful parallel processing capabilities and customized design flexibility, which can better meet the stringent requirements of real-time control systems. By deploying the control algorithm on these hardware platforms, the real-time performance and processing capabilities of the system can be significantly improved.

[0070] First, verifying the robustness of the controller to other types of disturbances or model uncertainties enhances the adaptability and stability of the controller, enabling it to maintain high performance under a wider range of operating conditions. This improvement in robustness helps reduce the system's dependence on specific environments and conditions and improves the system's reliability and availability. Second, designing a fault-tolerant control strategy provides additional safety for the system. In emergencies such as component failures or parameter changes, the fault-tolerant control strategy can respond quickly and adjust the control strategy to avoid serious impacts of the failure on the system. This fault-tolerant capability helps extend the service life of the system and reduce downtime and maintenance costs caused by failures. Optimizing the internal model control algorithm and combining it with the implementation of a high-performance hardware platform significantly improves the system's real-time performance and processing capabilities. Faster computing speeds and lower latency enable the controller to more accurately track system changes and respond in a timely manner, thereby improving the control accuracy and stability of the system.

[0071] S30. Approximation and compensation of unmodeled dynamics based on neural networks: For dynamic behaviors that are difficult to model or unknown in the system, neural networks are used for approximation. Neural networks have powerful nonlinear mapping capabilities and can learn and simulate complex dynamic relationships. The neural network model is trained to accurately predict the behavior of unmodeled dynamics and incorporate it into the control strategy of the system to improve the accuracy and robustness of control.

[0072] Step S30 specifically includes the following steps:

[0073] S301, neural network selection and construction, according to the complexity of the unmodeled dynamics and the specific needs of the system, select the appropriate neural network type, where the neural network type is RBF neural network, determine the number of nodes in the input layer, hidden layer and output layer of the neural network, as well as the connection method and activation function between the layers, the input layer contains the state and control variables of the system, and the output layer outputs the predicted value of the unmodeled dynamics;

[0074] S302, data preparation and preprocessing, collecting data containing unmodeled dynamic behaviors from the system, which can be obtained through experiments, simulations or actual operations, and cleaning, denoising and normalizing the collected data to improve the effect and convergence speed of neural network training;

[0075] S303, neural network training, dividing the preprocessed data into a training set, a validation set and a test set, the training set is used to train the neural network model, the validation set is used to adjust the model parameters to prevent overfitting, and the test set is used to evaluate the generalization ability of the model. The neural network is trained using the training set data, and the model is enabled to accurately predict the behavior of unmodeled dynamics by adjusting the center and basis width of the basis function of the RBF neural network. During the training process, a parallel learning algorithm is used to accelerate convergence and improve the training effect. The trained neural network model is evaluated using the validation set to check its prediction accuracy and generalization ability. If the model performance is poor, the model structure or training parameters need to be adjusted and retrained;

[0076] S304, neural network integration and control strategy optimization, integrating the trained neural network model into the control strategy of the system, adjusting the parameters or output of the controller according to the prediction results of the neural network to compensate for the impact of unmodeled dynamics on system performance, optimizing the control strategy of the system based on the integrated neural network model, and verifying the performance of the optimized control strategy through simulation or experiment to ensure that it can improve the control accuracy and robustness of the system;

[0077] S305, real-time implementation and verification, combining the optimized control strategy with the system's hardware and software platform to achieve real-time control, which includes writing control algorithms, configuring hardware interfaces, and debugging software systems, etc. Field tests are conducted in actual systems to verify the approximation effect of the RBF neural network model on unmodeled dynamics and the performance of the optimized control strategy. By changing the system's operating conditions or introducing different interference sources, the control effect and stability of the system are observed. Based on the field test results, the performance of the neural network model and control strategy is evaluated. If the control effect is found to be unsatisfactory or the system becomes unstable, the neural network model and control strategy need to be further adjusted and optimized.

[0078] In step S30, in view of the unmodeled dynamics problem existing in the piezoelectric driven micro displacement system, a neural network model is constructed and trained, and integrated into the control strategy of the system to achieve effective compensation for the unmodeled dynamics and improve the control performance. First, in step S301, according to the complexity of the unmodeled dynamics and the system requirements, a suitable neural network type is selected and its structure is designed, including determining the number of nodes in the input layer, hidden layer and output layer, and selecting an appropriate activation function. This process ensures that the neural network can capture the key features of the unmodeled dynamics and lays the foundation for subsequent training and prediction. Then, in step S302, by collecting system data and preprocessing it, a high-quality data set is provided for the training of the neural network. Data cleaning, denoising and normalization processing help to improve the training efficiency and prediction accuracy of the neural network. In step S303, the preprocessed data is used to train the neural network model. By continuously adjusting the model parameters, the model can accurately predict the behavior of the unmodeled dynamics. During the training process, optimization algorithms such as gradient descent and back propagation algorithm are used to accelerate the convergence of the model and improve the training effect. The use of validation sets and test sets ensures the generalization ability and prediction accuracy of the model. In step S304, the trained neural network model is integrated into the control strategy of the system. The prediction results of the neural network are used to guide the parameter adjustment or output compensation of the controller, thereby achieving effective compensation for the unmodeled dynamics. At the same time, the control strategy is optimized to further improve the control accuracy and robustness of the system. This process is verified by simulation or experiment to ensure the effectiveness of the optimized control strategy in practical applications. Finally, in step S305, the optimized control strategy is combined with the hardware and software platform of the system to achieve real-time control. Field tests verify the approximation effect of the neural network model on the unmodeled dynamics and the performance of the optimized control strategy. By changing the system operating conditions or introducing different interference sources, the control effect and stability of the system are evaluated, providing a basis for subsequent adjustments and optimizations.

[0079] By constructing and training the neural network model, effective prediction and compensation for unmodeled dynamics are achieved, and the control accuracy and stability of the system are significantly improved. This process not only solves the problem that traditional control methods are difficult to accurately describe and compensate for unmodeled dynamics, but also provides the possibility for further optimization of the system. Secondly, the neural network model is integrated into the control strategy, and the robustness and adaptability of the system are further improved by optimizing the control strategy. This enables the system to maintain a stable control effect when facing complex and changeable operating environments and external interferences, and reduces the risk of failures. In addition, the real-time implementation and verification process ensures the effectiveness and reliability of the optimized control strategy in practical applications. The results of the field test provide an important reference basis for the subsequent operation and maintenance of the system, and also provide direction for possible further adjustments and optimizations.

[0080] During system operation, the prediction accuracy and stability of the neural network model are monitored in real time, possible performance degradation issues are discovered and handled in a timely manner, and safety boundaries are set for the prediction results of the neural network model to ensure that even in the worst case, the system output does not exceed an acceptable safety range. At the same time, the neural network model is regularly maintained and updated, including data retraining, model structure optimization, and software platform upgrades, to ensure that it always maintains optimal performance.

[0081] The performance of the model when processing real-time data ensures that the model's prediction results always meet the needs of system control. When it is found that the prediction accuracy decreases or the model stability fluctuates, the system can quickly identify and trigger the corresponding processing flow. Setting a safety boundary for the prediction results of the neural network model is a preventive measure that ensures that even in the most unfavorable circumstances, such as temporary deterioration of model performance or extreme input conditions, the system's output can be limited to an acceptable safety range, thereby avoiding potential harm to the system or the external environment. In addition, regular maintenance and updating of neural network models are key to maintaining their optimal performance. This includes retraining the model with new or updated data sets to improve the adaptability and accuracy of the model; optimizing the model structure, such as adding or removing hidden layers, adjusting the number of nodes, or changing the activation function to better match the system characteristics and control requirements; and upgrading the software platform to take advantage of the latest computing resources and algorithm optimization to improve the operating efficiency and real-time performance of the model. These maintenance and update activities not only help maintain the long-term effectiveness of the model, but also enable it to continuously adapt to changes in the system environment and improve the control performance of the overall system.

[0082] Real-time monitoring of prediction accuracy and stability can timely detect and deal with possible performance degradation problems, and prevent model failure from adversely affecting system control. This preventive maintenance measure helps to improve the reliability and stability of the system and reduce system failures and downtime caused by model performance problems. Secondly, setting a safety boundary for the prediction results of the neural network model provides additional security for system control. When the model performance temporarily declines or faces extreme conditions, the safety boundary can limit the output range of the system to prevent it from exceeding the acceptable safety threshold, thereby protecting the system itself and the surrounding environment from potential hazards. This safety mechanism is of great significance to improving the safety and risk resistance of the system. Finally, regular maintenance and updating of the neural network model can ensure that the model always maintains optimal performance. Through measures such as retraining, structural optimization and software platform upgrades, the model can continuously adapt to changes in the system environment and increased control needs, thereby improving the overall control performance and efficiency of the system.

[0083] S40. Interference estimation and compensation based on nonlinear interference observer: To achieve accurate compensation of external interference: Design a nonlinear interference observer to observe and estimate the external interference to the system in real time. The observer can use the system output and input data to accurately estimate the size and change trend of the interference, and design a compensation strategy based on the data model to incorporate the observed external interference into the control strategy. By generating corresponding compensating control effects, the impact of external interference on the system can be accurately offset.

[0084] Step S40 specifically includes the following steps:

[0085] S401. Design a nonlinear disturbance observer. According to the characteristics of the system and the characteristics of the external disturbance, select a suitable nonlinear disturbance observer. Such an observer usually has a strong nonlinear processing capability and can accurately capture and estimate the changes of the external disturbance. Based on the input and output data of the system and possible prior knowledge, establish a mathematical model of the nonlinear disturbance observer. The model should be able to reflect the relationship between the system output and the external disturbance and have the ability to be updated in real time. Adjust and optimize the parameters of the observer through simulation or experimental data to ensure that it can accurately estimate the size and change trend of the external disturbance.

[0086] S402, real-time observation and estimation of external disturbances, real-time collection of system input and output data, including system state variables measured by sensors and control signals output by controllers, etc., use nonlinear disturbance observers to process the collected data, and estimate the external disturbances to the system in real time. This step is the basis for accurate compensation, and its accuracy directly affects the effect of subsequent compensation strategies;

[0087] S403. Design a compensation strategy in combination with the data model. Establish a data model of the system based on the physical principles, empirical formulas or machine learning models of the system. The model should be able to describe the dynamic behavior of the system and the impact of external disturbances on the system. Design a corresponding compensation strategy based on the external disturbance estimated by the nonlinear disturbance observer and the data model. The compensation strategy should be able to produce a control effect opposite to the external disturbance to accurately offset its impact on the system.

[0088] S404, realize accurate compensation, embed the designed compensation strategy into the control system, generate corresponding compensation control signal in real time, the signal should be able to respond quickly to the change of external disturbance and effectively offset its influence on system performance, verify the effectiveness of compensation strategy through simulation or experiment, evaluate the performance of system under external disturbance, including stability, tracking accuracy and robustness, etc.;

[0089] S405. Continuous optimization and adjustment, establishment of feedback mechanism, continuous optimization and adjustment of nonlinear disturbance observer and compensation strategy according to system performance evaluation results, including adjustment of observer parameters, optimization of compensation strategy algorithm and improvement of data model, etc., to ensure that the system can adapt to different types of external disturbances and changes in working environment, and maintain its precise compensation capability and good performance.

[0090] In step S40, a precise compensation mechanism based on a nonlinear disturbance observer is designed and implemented to solve the external interference problem that the piezoelectric driven micro-displacement system may face. First, in step S401, a suitable nonlinear disturbance observer is selected and designed by deeply analyzing the characteristics of the system and the characteristics of the external interference. This observer uses the input and output data of the system and possible prior knowledge to construct a mathematical model that can reflect the relationship between the system output and the external interference in real time. The model not only has powerful nonlinear processing capabilities, but also can update itself according to real-time data to ensure that it can accurately capture and estimate the changes in external interference. Next, in step S402, the system collects input and output data in real time, including system state variables measured by sensors and control signals output by controllers. These data are sent to the nonlinear disturbance observer for processing to estimate the external interference suffered by the system in real time. The accuracy of this step is directly related to the effect of the subsequent compensation strategy and is the basis for precise compensation. In step S403, based on the physical principles, empirical formulas or machine learning models of the system, a mathematical model of the system is established. According to the model, the model comprehensively describes the dynamic behavior of the system and the impact of external disturbances on the system, providing a solid foundation for designing compensation strategies. Subsequently, according to the external disturbance estimation value and data model provided by the nonlinear disturbance observer, a corresponding compensation strategy is designed. The strategy aims to produce a control effect opposite to the external disturbance to accurately offset its impact on the system performance. In step S404, the designed compensation strategy is embedded in the control system to generate compensation control signals in real time. These signals can quickly respond to changes in external disturbances and effectively offset their adverse effects on system performance. Through simulation or experimental verification, the effectiveness of the compensation strategy under external disturbances, as well as key performance indicators such as system stability, tracking accuracy and robustness are evaluated. Finally, in step S405, a feedback mechanism is established to continuously optimize and adjust the nonlinear disturbance observer and compensation strategy. According to the system performance evaluation results, the observer parameters are continuously adjusted, the compensation strategy algorithm is optimized, and the data model is improved to ensure that the system can adapt to different types of external disturbances and changes in the working environment, and maintain its accurate compensation capability and good performance.

[0091] By designing a nonlinear disturbance observer and estimating external disturbances in real time, the system can accurately capture and predict the changing trend of disturbances, providing a reliable basis for subsequent precise compensation, which greatly improves the system's anti-interference ability and stability in complex environments. Secondly, the compensation strategy designed based on the data model can produce a control effect opposite to the external disturbance, effectively offsetting its impact on system performance, which not only improves the tracking accuracy and response speed of the system, but also enhances the robustness of the system, enabling it to maintain stable performance under various interference conditions. In addition, the continuous optimization and adjustment mechanism ensures that the system can continuously adapt to changes in the external environment and the diversity of interference types. By continuously adjusting the observer parameters, optimizing the compensation strategy algorithm and improving the data model, the system can always maintain its precise compensation capability and good performance, extending the service life of the system and reducing maintenance costs.

[0092] Real-time monitoring of system performance indicators, such as response time, control accuracy, resource consumption, etc., to ensure that the system can maintain stable performance under high load or complex environment. When performance indicators are found to decline, timely analyze the reasons and take corresponding measures to make adjustments.

[0093] In a real-time monitoring system, continuous monitoring of performance indicators such as response time, control accuracy, and resource consumption is the key to ensuring stable operation of the system. This process runs through all links from step S401 to S405. First, when designing a nonlinear disturbance observer in S401, it is necessary to consider its demand for system resources and its impact on the system response time to ensure that the design of the observer matches the overall performance requirements of the system. Subsequently, when observing and estimating external disturbances in real time in S402, the system needs to efficiently process input and output data, which requires the system to have fast response capabilities and low resource consumption. When designing a compensation strategy in combination with a data model in S403, not only the compensation effect is concerned, but also the impact of the compensation strategy on system performance, such as whether it will increase response time or resource consumption. Therefore, trade-offs and optimizations are required when designing the strategy. When S404 achieves precise compensation, the generation of the compensation control signal needs to be rapid and accurate to avoid additional burden on system performance. Finally, in the S405 continuous optimization and adjustment phase, the system monitors changes in performance indicators to promptly identify potential problems, such as increased response time, reduced control accuracy, or increased resource consumption, and then analyzes the causes and takes targeted measures to make adjustments, such as optimizing algorithms, upgrading hardware, or adjusting system configurations, to ensure the stability and continuous optimization of system performance.

[0094] By continuously monitoring key performance indicators such as response time, control accuracy and resource consumption, the system can promptly detect and resolve potential performance issues, thereby ensuring stable performance under high load or complex environments. This greatly improves the reliability and availability of the system and reduces the risk of failure due to system performance degradation. Secondly, when a performance indicator is found to have declined, the system can quickly analyze the cause and take appropriate measures to make adjustments. This rapid response and adaptive capability helps to reduce system downtime and maintenance costs and improve the overall operating efficiency of the system. At the same time, by continuously optimizing and adjusting system configurations and algorithms, the system can gradually improve its performance to better meet changes in user needs and application scenarios.

[0095] In the S10 stage, through in-depth analysis of modelable interference sources in the system, such as harmonic interference, the corresponding mathematical models are constructed, and the adaptive algorithm is used to identify the time-varying parameters in these models in real time. This process ensures that the model can closely follow the actual changes in the interference and provide an accurate basis for subsequent interference suppression. Then, in the S20 stage, based on the established interference model, an internal model-based interference suppression strategy is designed and implemented. The internal model controller directly offsets the impact of interference on the system output by generating a control action opposite to the interference. By fine-tuning the parameters of the internal model controller, it not only ensures the maximization of the interference suppression effect, but also takes into account the stability and response speed of the system, avoiding the negative impact that may be caused by excessive suppression. For those dynamic behaviors in the system that are difficult to model or completely unknown, the S30 stage introduces a neural network model for approximation. With its powerful nonlinear mapping ability, the neural network can learn and simulate complex dynamic relationships, thereby accurately predicting the behavior of unmodeled dynamics. By training the neural network The network model is incorporated into the control strategy of the system, which effectively improves the accuracy and robustness of the control and further enhances the system's ability to cope with unknown interference. Finally, in the S40 stage, an accurate compensation mechanism for external interference based on nonlinear interference observer and data model is designed and implemented. The nonlinear interference observer uses the system output and input data to observe and estimate the external interference of the system in real time, accurately capture the size and change trend of the interference, and design and implement the compensation strategy in combination with the data model. The observed external interference is incorporated into the control strategy, and the influence of external interference on the system is accurately offset by generating corresponding compensation control effects. This process not only improves the system's ability to suppress external interference, but also ensures the stable operation of the system in various complex environments. In summary, this method achieves effective suppression of multi-source interference in the piezoelectric driven micro-displacement system by comprehensively using technical means such as mathematical modeling, internal model control, neural network approximation and nonlinear interference observation, and significantly improves the overall performance and stability of the system.

[0096] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0097] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for suppressing multi-source interference of a piezoelectric driven micro-displacement system, characterized in that: The following steps are involved: S10. Establish a model of modelable interference and adaptively identify time-varying parameters: Conduct in-depth analysis of modelable interference sources in the piezoelectric drive micro-displacement system, such as harmonic interference, to understand its generation mechanism and impact. According to the characteristics of the interference source, establish a corresponding mathematical model to describe the relationship between interference and system output, design an adaptive algorithm, and identify the time-varying parameters in the interference model in real time to ensure that the model can accurately reflect the actual changes in interference. S20. Apply the internal model-based interference suppression method to suppress the modelable interference: Based on the established interference model, design an internal model controller that can produce a control effect opposite to the interference, thereby offsetting the impact of the interference on the system, and adjust the parameters of the internal model controller to ensure that its interference suppression effect is optimal without affecting the stability and response speed of the system; S30. Approximation and compensation of unmodeled dynamics based on neural networks: For dynamic behaviors that are difficult to model or unknown in the system, neural networks are used for approximation. Neural networks have powerful nonlinear mapping capabilities and can learn and simulate complex dynamic relationships. The neural network model is trained to accurately predict the behavior of unmodeled dynamics and incorporate it into the control strategy of the system to improve the accuracy and robustness of control. S40. Interference estimation and compensation based on nonlinear interference observer: To achieve accurate compensation of external interference: Design a nonlinear interference observer to observe and estimate the external interference to the system in real time. The observer can use the system output and input data to accurately estimate the size and change trend of the interference, and design a compensation strategy based on the data model to incorporate the observed external interference into the control strategy. By generating corresponding compensating control effects, the impact of external interference on the system can be accurately offset.

2. The multi-source interference suppression method of a piezoelectric driven micro-displacement system according to claim 1 is characterized in that: The step S10 specifically includes the following steps: S101. Analysis and identification of interference sources. Identify all possible interference sources in the system, including but not limited to mechanical vibration, electromagnetic interference, power supply fluctuations, performance changes caused by temperature changes, and harmonic interference. In-depth study of the generation mechanism, frequency characteristics, amplitude variation range of each interference and its specific impact on system performance. For harmonic interference, special attention should be paid to the stability of its frequency component, phase and amplitude. S102, establish mathematical model, interference model construction: according to the characteristic analysis of interference source, select appropriate mathematical model to describe interference. For harmonic interference, it can be represented by superposition of sine wave or cosine wave, and it is necessary to identify the amplitude, frequency and phase of harmonic system dynamic model. In addition to the interference model, it is also necessary to establish the dynamic model of the system itself, which usually includes the transfer function or state space model of piezoelectric actuator to describe the relationship between system input such as control voltage and output such as displacement. Combine the system model and interference model to form a comprehensive model including interference influence for subsequent analysis and control design; S103, adaptive algorithm design, design an adaptive internal model to estimate the time-varying parameters such as amplitude, frequency, and phase in the interference model in real time, consider the convergence speed, stability, computational complexity and other factors of the algorithm, optimize the algorithm, and ensure that the changes of interference parameters can be tracked quickly and accurately in practical applications; S104. Real-time implementation and verification. Combining the adaptive internal model algorithm with the hardware and software platform of the piezoelectric drive micro-displacement system to achieve real-time estimation and update of the disturbance parameters. A series of experiments are conducted to verify the effectiveness of the established model and adaptive algorithm, including testing the performance of the system under different disturbances, observing whether the model can accurately reflect the actual changes in the disturbance, and evaluating its effect on improving the control accuracy and stability of the system. S105. Feedback and Adjustment. Based on the experimental results, evaluate the performance of the model and algorithm, identify potential areas for improvement, and make necessary adjustments and optimizations to the model and algorithm based on the evaluation results to improve their adaptability and accuracy.

3. The multi-source interference suppression method of a piezoelectric driven micro-displacement system according to claim 2 is characterized in that: When there are multiple types of interference in the system, these interferences may be coupled with each other and affect the accurate estimation of parameters. The multi-source interference decoupling technology can be used to separate the impact of different interference sources on the system, thereby improving the accuracy of each interference parameter estimation. Different interference sources are identified and separated by methods such as blind source separation and independent component analysis. At the same time, during the operation of the system, possible faults are monitored and diagnosed in real time so that timely measures can be taken to repair or isolate the source of the fault. Fault tree analysis, expert system or data-based fault diagnosis methods such as anomaly detection and pattern recognition are used to identify abnormal conditions in the system. As the system operating environment changes or the system itself ages, the original model may no longer be applicable. Through online learning or incremental learning technology, the model parameters can be updated in real time to keep an accurate description of the system status at all times. Sliding window technology, online support vector machine or incremental neural network and other methods are used to update the model parameters.

4. The method for suppressing multi-source interference of a piezoelectric driven micro-displacement system according to claim 2, characterized in that: The step S20 specifically includes the following steps: S201, internal model controller design, integrating the interference model established in S102 into the design of the internal model controller, which includes taking into account the dynamic characteristics of the interference such as the frequency, phase and amplitude of the harmonic interference, designing the core part of the internal model controller, including the internal model, the internal model controller and the filter. The internal model is used to predict the output of the controlled object and correct it. The internal model controller adjusts the control quantity to track the given value of the control system and offset the interference. The filter is used to improve the robustness of the control system. According to the interference model and the system dynamic model, a suitable controller transfer function is designed so that it can maintain the stability and performance of the system in the presence of interference; S202, parameter adjustment and optimization, according to the system characteristics and interference model, preliminarily set the parameters of the internal model controller, such as the time constant of the filter, the gain of the controller, etc., use MATLAB / Simulink to perform simulation experiments, verify the performance of the internal model controller in the presence of interference, and evaluate the interference suppression effect of the controller and the stability and response speed of the system through the simulation results. According to the simulation results, adjust and optimize the parameters of the controller, and gradually find the best parameter combination through the iterative process, so that the controller can suppress interference while maintaining the stability and rapid response capability of the system; S203, real-time implementation and verification, combining the optimized internal model controller with the hardware and software platform of the system to realize real-time control, which includes writing control algorithms, configuring hardware interfaces and debugging software systems, etc., conducting field tests in the actual system to verify the performance of the internal model controller under actual working conditions, observing the suppression effect of the controller and the stability of the system by changing the operating conditions of the system or introducing different interference sources, and evaluating the performance of the controller based on the field test results. If it is found that the control effect is not ideal or the system is unstable, it is necessary to further adjust and optimize the parameters of the controller; S204. Documentation and feedback: record the design process, parameter adjustment results and field test results of the internal model controller, write detailed technical documents and user manuals to provide reference for subsequent maintenance and upgrades, establish a feedback mechanism, collect user feedback and opinions on the use of the controller, and continuously improve and optimize the controller based on the feedback results to meet the different needs and application scenarios of users.

5. The method for suppressing multi-source interference of a piezoelectric driven micro-displacement system according to claim 4, characterized in that: In addition to optimizing specific interference models, it is also necessary to verify the robustness of the controller to other types of interference or model uncertainties, design fault-tolerant control strategies to cope with system component failures or parameter changes, and ensure that the controller can still maintain a certain control performance in the event of a failure. According to the needs of real-time control systems, the internal model control algorithm is optimized to reduce computational complexity and improve computational efficiency. Consider using high-performance hardware platforms such as FPGA and DSP to implement the control algorithm to further improve the real-time performance of the system.

6. The multi-source interference suppression method of a piezoelectric driven micro-displacement system according to claim 1, characterized in that: The step S30 specifically includes the following steps: S301, neural network selection and construction, according to the complexity of the unmodeled dynamics and the specific needs of the system, select the appropriate neural network type, where the neural network type is RBF neural network, determine the number of nodes in the input layer, hidden layer and output layer of the neural network, as well as the connection method and activation function between the layers, the input layer contains the state and control variables of the system, and the output layer outputs the predicted value of the unmodeled dynamics; S302, data preparation and preprocessing, collecting data containing unmodeled dynamic behaviors from the system, which can be obtained through experiments, simulations or actual operations, and cleaning, denoising and normalizing the collected data to improve the effect and convergence speed of neural network training; S303, neural network training, dividing the preprocessed data into a training set, a validation set and a test set, the training set is used to train the neural network model, the validation set is used to adjust the model parameters to prevent overfitting, and the test set is used to evaluate the generalization ability of the model. The neural network is trained using the training set data, and the model is enabled to accurately predict the behavior of unmodeled dynamics by adjusting the center and basis width of the basis function of the RBF neural network. During the training process, a parallel learning algorithm is used to accelerate convergence and improve the training effect. The trained neural network model is evaluated using the validation set to check its prediction accuracy and generalization ability. If the model performance is poor, the model structure or training parameters need to be adjusted and retrained; S304, neural network integration and control strategy optimization, integrating the trained neural network model into the control strategy of the system, adjusting the parameters or output of the controller according to the prediction results of the neural network to compensate for the impact of unmodeled dynamics on system performance, optimizing the control strategy of the system based on the integrated neural network model, and verifying the performance of the optimized control strategy through simulation or experiment to ensure that it can improve the control accuracy and robustness of the system; S305, real-time implementation and verification, combining the optimized control strategy with the system's hardware and software platform to achieve real-time control, which includes writing control algorithms, configuring hardware interfaces, and debugging software systems, etc. Field tests are conducted in actual systems to verify the approximation effect of the RBF neural network model on unmodeled dynamics and the performance of the optimized control strategy. By changing the system's operating conditions or introducing different interference sources, the control effect and stability of the system are observed. Based on the field test results, the performance of the neural network model and control strategy is evaluated. If the control effect is found to be unsatisfactory or the system becomes unstable, the neural network model and control strategy need to be further adjusted and optimized.

7. The method for suppressing multi-source interference of a piezoelectric driven micro-displacement system according to claim 6, characterized in that: During the operation of the system, the prediction accuracy and stability of the neural network model are monitored in real time, possible performance degradation problems are discovered and handled in a timely manner, and a safety boundary is set for the prediction results of the neural network model to ensure that even in the worst case, the output of the system does not exceed an acceptable safety range. At the same time, the neural network model is regularly maintained and updated, including data retraining, model structure optimization, and software platform upgrades, to ensure that it always maintains optimal performance.

8. The method for suppressing multi-source interference of a piezoelectric driven micro-displacement system according to claim 1, characterized in that: The step S40 specifically includes the following steps: S401. Design a nonlinear disturbance observer. According to the characteristics of the system and the characteristics of the external disturbance, select a suitable nonlinear disturbance observer. Such an observer usually has a strong nonlinear processing capability and can accurately capture and estimate the changes of the external disturbance. Based on the input and output data of the system and possible prior knowledge, establish a mathematical model of the nonlinear disturbance observer. The model should be able to reflect the relationship between the system output and the external disturbance and have the ability to be updated in real time. Adjust and optimize the parameters of the observer through simulation or experimental data to ensure that it can accurately estimate the size and change trend of the external disturbance. S402, real-time observation and estimation of external disturbances, real-time collection of system input and output data, including system state variables measured by sensors and control signals output by controllers, etc., use nonlinear disturbance observers to process the collected data, and estimate the external disturbances to the system in real time. This step is the basis for accurate compensation, and its accuracy directly affects the effect of subsequent compensation strategies; S403. Design a compensation strategy in combination with the data model. Establish a data model of the system based on the physical principles, empirical formulas or machine learning models of the system. The model should be able to describe the dynamic behavior of the system and the impact of external disturbances on the system. Design a corresponding compensation strategy based on the external disturbance estimated by the nonlinear disturbance observer and the data model. The compensation strategy should be able to produce a control effect opposite to the external disturbance to accurately offset its impact on the system. S404, realize accurate compensation, embed the designed compensation strategy into the control system, generate corresponding compensation control signal in real time, the signal should be able to respond quickly to the change of external disturbance and effectively offset its influence on system performance, verify the effectiveness of compensation strategy through simulation or experiment, evaluate the performance of system under external disturbance, including stability, tracking accuracy and robustness, etc.; S405. Continuous optimization and adjustment, establishment of feedback mechanism, continuous optimization and adjustment of nonlinear disturbance observer and compensation strategy according to system performance evaluation results, including adjustment of observer parameters, optimization of compensation strategy algorithm and improvement of data model, etc., to ensure that the system can adapt to different types of external disturbances and changes in working environment, and maintain its precise compensation capability and good performance.

9. The method for suppressing multi-source interference of a piezoelectric driven micro-displacement system according to claim 1, characterized in that: The real-time monitoring system's performance indicators, such as response time, control accuracy, resource consumption, etc., are used to ensure that the system can maintain stable performance under high load or complex environments. When performance indicators are found to have declined, the reasons are analyzed in a timely manner and corresponding measures are taken to make adjustments.

Images