Position calibration optimization method and system for limited-angle motor used in servo valve

Through characteristic analysis and the construction of deep neural network model, the problem of poor accuracy and stability of finite angle motor position calibration for servo valves is solved, and higher accuracy and stable motor position control is achieved.

CN118508823BActive Publication Date: 2024-11-12苏州海卓伺服驱动技术有限公司
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410957318.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-07-17
Publication Date
2024-11-12
Estimated Expiration
2044-07-17

AI Technical Summary

Technical Problem

The position calibration of existing servo valves for limited angle motors is difficult to accurately predict and control based on the application characteristics of the motor, resulting in poor position control accuracy and stability.

Method used

Characteristic analysis, construction of position control prediction model and calibration optimization model, and position prediction training of motor motion characteristics data flow and application characteristics through deep neural network structure, generate motor position control prediction model, and calibrate optimization based on drive control parameters to achieve Accurate position calibration control.

Benefits of technology

The position control accuracy and control stability of the limited angle motor are improved, and the reliability of the motor is enhanced.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118508823B_ABST
    Figure CN118508823B_ABST
Patent Text Reader

Abstract

The present invention discloses a position calibration optimization method and system for a limited angle motor for a servo valve, and relates to the technical field of motor control optimization. The method comprises: obtaining a motor test parameter table, performing test collection, and obtaining a motor motion characteristic data stream; performing characteristic analysis based on the motor standard motion characteristic data stream to obtain motor application characteristics; generating a motor position control prediction model; performing predictive analysis on the motor drive control parameters to obtain motor predicted position information; obtaining a motor position calibration optimization model, and performing position calibration control on the target angle motor. The method solves the technical problem that it is difficult to accurately predict and control the motor position and drive control parameters according to the application characteristics of the motor in the existing servo valve angle motor position calibration optimization, which leads to poor position control accuracy and motor control stability of the angle motor, and achieves the technical effect of improving the position control accuracy of the angle motor and the motor control stability and reliability.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field related to motor control optimization, and in particular to a position calibration optimization method and system for a limited-angle motor for a servo valve. Background Art

[0002] In the field of industrial automation, servo valves, as an important fluid control component, are widely used in aerospace, machine tools, metallurgy and other fields. With the continuous development of industrial automation and the growing demand for high-precision and high-stability position control in the field of precision control, the requirements for precise control of servo valves are getting higher and higher. As the driving element of the servo valve, the limited angle motor (LATM) has the characteristics of limited angle range, high torque / power ratio, low cost and high reliability. As the driving device of the servo valve, the accuracy and stability of the position control of the limited angle motor directly affect the performance of the entire system. In traditional servo valve systems, the position control calibration of the limited angle motor often relies on simple PID control or manual calibration, which is powerless when dealing with complex working conditions and high-precision requirements. It is difficult to ensure the accuracy and stability of the motor position control, thus affecting the performance of the entire system.

[0003] Therefore, in the current technology related to position calibration optimization of limited-angle motors for servo valves, there is a technical problem that it is difficult to accurately predict and control the motor position and drive control parameters according to the application characteristics of the motor, which in turn leads to poor position control accuracy and motor control stability of the limited-angle motor. Summary of the invention

[0004] The present application provides a position calibration optimization method and system for a limited angle motor for a servo valve, and adopts technical means such as characteristic analysis, construction of a position control prediction model and a calibration optimization model, to solve the technical problem that it is difficult to accurately predict and control the motor position and drive control parameters according to the application characteristics of the motor, which leads to poor position control accuracy and motor control stability of the angle motor in the existing servo valve angle motor position calibration optimization, thereby achieving the technical effect of improving the position control accuracy of the angle motor and the stability and reliability of the motor control.

[0005] The present application provides a position calibration optimization method for a limited-angle motor for a servo valve, the method comprising: obtaining a motor test parameter table, testing and collecting a target angle motor based on the motor test parameter table and a measuring sensor device, obtaining a motor motion characteristic data stream, the motor motion characteristic data stream including motor position, speed, acceleration, and current and voltage; preprocessing the motor motion characteristic data stream to obtain a motor standard motion characteristic data stream, performing characteristic analysis based on the motor standard motion characteristic data stream to obtain motor application characteristics, the motor application characteristics including motor nonlinear factors and motor dynamic characteristics; using a deep neural network structure to analyze the motor The motion characteristic data stream and the motor application characteristics are used for position prediction training to generate a motor position control prediction model; the motor target position is set according to the servo valve system requirements, the motor drive parameter analysis is performed based on the motor target position, the motor drive control parameters are determined, and the motor drive control parameters are predicted and analyzed based on the motor position control prediction model to obtain the motor predicted position information; based on the motor drive control parameters and the motor predicted position information, the motor position control prediction model is calibrated and optimized to obtain a motor position calibration optimization model, and the target angle motor is position calibrated and controlled based on the motor position calibration optimization model.

[0006] In the implementation method of the present application, the motor test parameter table is obtained by performing the following processing: obtaining the application scenario information and control requirement information of the target angle motor; extracting test elements from the application scenario information and control requirement information to obtain application scenario requirement elements and control requirement elements; designing motor test parameters based on the application scenario requirement elements and control requirement elements to obtain a motor scenario test parameter set and a control requirement test parameter set; and orthogonally arranging the motor scenario test parameter set and the control requirement test parameter set to obtain the motor test parameter table.

[0007] In the implementation of the present application, the motor position control prediction model is generated by performing the following processing: using a deep neural network structure to perform position control analysis on the motor motion characteristic data stream to obtain a basic motor position control model; based on the motor application characteristics, the basic motor position control model is impact updated and trained to obtain a motor position update prediction model; the motor position update prediction model is multi-dimensionally evaluated and verified to obtain model performance evaluation parameters, and a model optimization direction is determined based on the model performance evaluation parameters; based on the model optimization direction, the motor position update prediction model is optimized to generate the motor position control prediction model.

[0008] In the implementation method of the present application, the motor position control prediction model is generated and the following processing is also performed: an optimization strategy analysis is performed on the optimization direction of the model to determine the model optimization strategy parameters; model parameter information of the motor position update prediction model is extracted, and the model parameter information is optimized and updated based on the model optimization strategy parameters, and the optimized model parameter information is evaluated and iteratively updated through a fitness loss function to determine the optimal model parameters; the motor position update prediction model is configured and optimized based on the optimal model parameters to obtain the motor position control prediction model.

[0009] In the implementation method of the present application, the determination of the motor drive control parameters also performs the following processing: extracting drive characteristics from the motor motion characteristic data stream to obtain a set of motor drive associated control parameters, and using the set of motor drive associated control parameters as parameter influencing independent variables; extracting position parameters based on the motor motion characteristic data stream to obtain a set of motor position parameters, and using the set of motor position parameters as target parameter dependent variables; performing regression fitting on the parameter influencing independent variables and the target parameter dependent variables to construct a drive control parameter-motor position regression relationship; and calculating and analyzing the motor target position based on the drive control parameter-motor position regression relationship to determine the motor drive control parameters.

[0010] In the implementation method of the present application, the determination of the motor drive control parameters also performs the following processing: based on the drive control parameter-motor position regression relationship, the motor target position is threshold calculated to determine the drive control parameter selection threshold; a motor control target set is obtained, and the motor control target set is evaluated and fitted to generate a motor control effect fitness function; the motor control effect fitness function is used to perform parameter evaluation and optimization within the drive control parameter selection threshold to determine the motor drive control parameters.

[0011] In the implementation method of the present application, the motor position calibration optimization model is obtained, and the following processing is also performed: the target angle motor is controlled and monitored based on the motor drive control parameters to obtain actual motor feedback position information; the deviation between the motor predicted position information and the actual motor feedback position information is used as the motor deviation position variable; the motor position control prediction model is calibrated and optimized based on the motor deviation position variable to obtain the motor position calibration optimization model.

[0012] The present application also provides a position calibration optimization system for a limited-angle motor for a servo valve, comprising:

[0013] A motor motion characteristic data stream acquisition module, the motor motion characteristic data stream acquisition module is used to obtain a motor test parameter table, test and collect the target angle motor based on the motor test parameter table and the measurement sensor device, and obtain the motor motion characteristic data stream, the motor motion characteristic data stream includes the motor position, speed, acceleration, current and voltage; a motor application characteristic acquisition module, the motor application characteristic acquisition module is used to pre-process the motor motion characteristic data stream, obtain the motor standard motion characteristic data stream, perform characteristic analysis based on the motor standard motion characteristic data stream, and obtain the motor application characteristics, the motor application characteristics include motor nonlinear factors and motor dynamic characteristics; a motor position control prediction model generation module, the motor position control prediction model generation module is used to use a deep neural network structure to predict the motor motion characteristics The invention relates to a motor prediction model, wherein the motor prediction position information determination module is used to set the motor target position according to the requirements of the servo valve system, analyze the motor drive parameters based on the motor target position, determine the motor drive control parameters, and predict and analyze the motor drive control parameters based on the motor position control prediction model to obtain the motor prediction position information; and a motor position calibration optimization model acquisition module is used to calibrate and optimize the motor position control prediction model based on the motor drive control parameters and the motor prediction position information to obtain the motor position calibration optimization model, and perform position calibration control on the target angle motor based on the motor position calibration optimization model.

[0014] Through the position calibration optimization method and system for limited-angle motors for servo valves proposed in this application, it is intended to obtain the motor test parameter table, conduct test collection, and obtain the motor motion characteristic data stream; perform characteristic analysis based on the motor standard motion characteristic data stream to obtain the motor application characteristics; generate a motor position control prediction model; perform predictive analysis on the motor drive control parameters to obtain the motor predicted position information; obtain the motor position calibration optimization model, and perform position calibration control on the target angle motor. This solves the technical problem that it is difficult to accurately predict and control the motor position and drive control parameters according to the motor's application characteristics, which leads to poor position control accuracy and motor control stability of the angle motor, and achieves the technical effect of improving the position control accuracy, motor control stability, and reliability of the angle motor. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solution of the embodiment of the present disclosure, the accompanying drawings of the embodiment of the present disclosure will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the system according to the embodiment of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0016] Figure 1 A schematic flow chart of a position calibration optimization method for a limited-angle motor for a servo valve provided in an embodiment of the present application;

[0017] Figure 2 Schematic diagram of the structure of a position calibration optimization system for a limited-angle motor for a servo valve provided in an embodiment of the present application.

[0018] Explanation of the reference numerals: motor motion characteristic data stream acquisition module 10 , motor application characteristic acquisition module 20 , motor position control prediction model generation module 30 , motor prediction position information determination module 40 , motor position calibration optimization model acquisition module 50 . DETAILED DESCRIPTION

[0019] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0020] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0021] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0022] The present application embodiment provides a method for optimizing the position calibration of a servo valve with a limited angle motor, such as Figure 1 As shown, the method includes:

[0023] Step S100, obtain a motor test parameter table, test and collect data on the target angle motor based on the motor test parameter table and the measuring sensor device, and obtain a motor motion characteristic data stream, wherein the motor motion characteristic data stream includes motor position, speed, acceleration, current and voltage. The motor test parameter table may contain a data table of various performance parameters and indicators of the motor under specific working conditions, such as motor model (specific model and specification), rated voltage (voltage value when the motor is operating normally), rated current (current when the motor is operating normally at rated voltage), rated speed (stable speed of the motor under rated voltage and load), maximum torque (maximum torque that the motor can generate), working efficiency (ratio of motor output power to input power), temperature rise (temperature rise relative to ambient temperature after the motor has been running for a period of time), etc.; the measuring sensor device may include a position sensor for measuring the real-time position of the motor, such as an encoder; a speed sensor for measuring The motor speed can be obtained by differentiating the position signal; the acceleration sensor is used to measure the acceleration of the motor and is usually used for dynamic performance analysis; the current and voltage sensor is used to measure the current and voltage values ​​when the motor is running to understand the motor's load and energy consumption; the motor motion characteristic data stream refers to the data set obtained by testing and collecting the target angle motor through the measurement sensor equipment. The motor motion characteristic data stream includes the motor position, speed, acceleration, and current and voltage. Specifically, the motor position indicates the real-time position of the motor rotor, the speed indicates the real-time speed of the motor, the acceleration indicates the rate of change of the motor speed, and the current and voltage indicate the current and voltage values ​​when the motor is running. The motor motion characteristic data stream of information such as motor position, speed, acceleration, current and voltage is helpful to analyze the motor's motion performance. At the same time, combined with the motor test parameter table, a more comprehensive understanding of the motor's performance characteristics and working status can be obtained.

[0024] In a possible implementation, step S100 further includes step S110, obtaining the application scenario information and control requirement information of the target angle motor. The application scenario information of the target angle motor may include robot arm and leg control; precise control of the position and angle of cutting tools in CNC data processing; precise control of various mechanical arms and conveyor belt equipment on automated production lines; driving wheels to turn to the target angle in online steering systems, etc.; the control requirement information may include accuracy, stability, rapid response, intelligence and reliability. The motor angle control has a high degree of accuracy to ensure that the equipment moves accurately; it is necessary to ensure stable performance during long-term operation to avoid problems such as jitter and offset; such as the wire control steering system, the motor angle control needs to have a rapid response capability to cope with the driver's rapid steering operation; the motor angle control is gradually developing in the direction of intelligence, and by introducing advanced control algorithms and sensor technologies, more efficient energy utilization, lower operating costs and better user experience can be achieved; the motor angle control needs to have a high degree of reliability to ensure that the equipment can still work normally in harsh environments. The method further includes step S120, in which test elements are extracted from the application scenario information and the control requirement information to obtain application scenario requirement elements and control requirement elements. Application scenario requirement elements refer to the specific requirements and conditions for the target corner motor and its control system in different application scenarios, usually including environmental adaptability, size and weight, power consumption and power, safety and reliability, etc. Control requirement elements refer to the specific requirements and performance indicators for the control system of the target corner motor, including accuracy, response speed, stability, reliability, etc.

[0025] Step S100 also includes step S130, designing motor test parameters based on the application scenario demand elements and control demand elements, and obtaining a motor scenario test parameter set and a control demand test parameter set. The motor scenario test parameter set is mainly designed for the application scenario demand elements, such as environmental adaptability test parameters (temperature range, humidity range, motor vibration, etc.), size and weight parameters (external dimensions, weight), power consumption and efficiency test parameters (power consumption test, efficiency test), safety test parameters (overload protection test, overheating protection test) and reliability test (performance degradation test, life test, etc. after continuous operation for a certain period of time); the control demand test parameter set is mainly designed for the control demand elements, such as accuracy test (angle control accuracy, repeated positioning accuracy), response speed test parameters (start time, stop time), stability test parameters (steady-state error, anti-interference ability), reliability test parameters (failure rate, recovery time, etc.). It also includes step S140, orthogonally arranging the motor scenario test parameter set and the control demand test parameter set to obtain the motor test parameter table. Orthogonal arrangement refers to the use of orthogonal tables to arrange experiments. With a small number of experiments, comprehensive and balanced test data can be obtained, and then the influence of various factors on the test indicators can be analyzed. The motor scenario test parameter set and the control demand test parameter set are orthogonally arranged, and finally a motor test parameter table is obtained.

[0026] Step S200, preprocessing the motor motion characteristic data stream to obtain the motor standard motion characteristic data stream, performing characteristic analysis based on the motor standard motion characteristic data stream to obtain the motor application characteristics, wherein the motor application characteristics include motor nonlinear factors and motor dynamic characteristics. Preprocessing the motor motion characteristic data stream refers to removing unfavorable factors such as noise and abnormal values ​​in the motor motion characteristic data, thereby improving data quality to more accurately analyze the motor motion characteristics. Through preprocessing, a clean, accurate and reliable motor motion characteristic data stream is obtained, that is, the motor standard motion characteristic data stream. The data stream should be able to accurately reflect the actual motion state of the motor, including key parameters such as motor position, speed, acceleration, etc. Then, based on the motor standard motion characteristic data stream, the application characteristics of the motor are analyzed, including motor nonlinear factors and motor dynamic characteristics. Specifically, motor nonlinear factors may include magnetic saturation, inductance non-decoupling, temperature changes of magnetic field and resistance, etc. Due to the influence of magnetic saturation, the inductance decreases with the increase of current, affecting the torque output of the motor, and there is mutual inductance between the inductance and the magnetic field, and the current change and the magnetic field affect each other. At the same time, the magnetic field and resistance of the motor will decrease or increase with the increase of temperature, affecting the output and efficiency of the motor; the motor dynamic characteristics may include acceleration and deceleration characteristics, continuous non-steady-state operation and dynamic response, as well as the motor's response time, damping ratio, natural frequency, etc., which reflect the motor's response ability when subjected to external excitation.

[0027] Step S300, using a deep neural network structure to perform position prediction training on the motor motion characteristic data stream and the motor application characteristics, and generate a motor position control prediction model. Using the motor motion characteristic data stream and the motor application characteristics, a model capable of predicting the future position of the motor is constructed and trained based on the deep neural network structure. Specifically, a recurrent neural network (RNN), a long short-term memory network (LSTM) or a convolutional neural network (CNN) can be used to determine the number of layers, the number of neurons, and the activation function of the network. The model is trained using the prepared motor motion characteristic data stream. During the training process, the motor motion characteristic data stream and the motor application characteristics are used as input, and the future position of the motor is used as output. The network parameters are adjusted through the back propagation algorithm so that the model can accurately predict the future position of the motor. The model is evaluated and optimized, and finally a motor position control prediction model is obtained, which can predict the future position of the motor based on the input motor motion characteristic data stream and the motor application characteristics, and provide guidance for the position control of the servo valve system.

[0028] In a possible implementation, step S300 further includes step S310, using a deep neural network structure to perform position control analysis on the motor motion characteristic data stream to obtain a basic motor position control model. The real-time data stream generated by the motor during movement is processed and parsed by a deep neural network structure to achieve precise control of the motor position and obtain a basic motor position control model. It also includes step S320, based on the motor application characteristics, the basic motor position control model is impact updated and trained to obtain a motor position update prediction model. For specific motor application scenarios and requirements, the existing basic motor position control model is impact updated and trained, that is, the model will adjust its internal parameters and weights according to the new data samples to adapt to the motor control requirements in specific application scenarios, and obtain a motor position update prediction model, which can more accurately predict the position change of the motor and achieve more efficient and stable position control.

[0029] Step S300 also includes step S330, performing multi-dimensional evaluation and verification on the motor position update prediction model, obtaining model performance evaluation parameters, and determining the model optimization direction according to the model performance evaluation parameters. The model is tested using motor motion characteristic data streams from different application scenarios, different working conditions and different time periods to evaluate the adaptability of the model under different conditions, thereby obtaining a series of model performance evaluation parameters, such as average prediction error, maximum prediction error, prediction response time, position prediction stability under abnormal conditions, etc. Determining the model optimization direction based on the model performance evaluation parameters means improving prediction accuracy, accelerating response speed, improving robustness or optimizing computational efficiency, etc. For example, if the prediction accuracy is insufficient, it may be necessary to increase the complexity of the model (such as increasing the number of network layers, the number of neurons, etc.) or improve the model structure (such as using a more advanced neural network architecture); if the response speed is too slow, it may be necessary to optimize the computational efficiency of the model or adopt a more efficient algorithm; if the robustness is poor, it may be necessary to increase the ability to handle abnormal conditions or improve the stability of the model. It also includes step S340, optimizing the motor position update prediction model based on the model optimization direction to generate the motor position control prediction model.

[0030] In a possible implementation, step S340 further includes step S341, performing optimization strategy analysis on the optimization direction of the model and determining the model optimization strategy parameters. To improve the prediction accuracy, the network structure can be adjusted (increasing the number of network layers and neurons), the training algorithm can be optimized, data enhancement can be performed, etc. To speed up the response speed, model compression or hardware optimization can be performed; to enhance robustness, regularization can be performed or an exception handling mechanism can be designed for abnormal situations; to optimize the computational efficiency, algorithm optimization or parallelization can be performed. After determining the optimization strategy, specific optimization strategy parameters are further determined, such as network structure parameters: such as the number of network layers, the number of neurons, the type of activation function, etc.; training algorithm parameters: such as learning rate, batch size, number of iterations, etc.; exception handling parameters: such as exception threshold, exception handling strategy, etc. It also includes step S342, extracting the model parameter information of the motor position update prediction model, optimizing and updating the model parameter information based on the model optimization strategy parameters, and evaluating and iterating the optimized model parameter information through the fitness loss function to determine the optimal model parameters. That is, the corresponding fitness of each model optimization strategy parameter is calculated respectively, and the model parameter with the largest fitness is selected to determine it as the model optimal parameter. It also includes step S343, configuring and optimizing the motor position update prediction model based on the model optimal parameter to obtain the motor position control prediction model.

[0031] Step S400, set the motor target position according to the requirements of the servo valve system, analyze the motor drive parameters based on the motor target position, determine the motor drive control parameters, predict and analyze the motor drive control parameters based on the motor position control prediction model, and obtain the motor predicted position information. According to the specific requirements of the servo valve system, set the accurate target position of the motor, analyze and determine the drive control parameters of the motor based on the set motor target position, for example, calculate and set the speed that the motor needs to reach based on the difference between the target position and the current position; determine whether the motor needs to rotate forward or reverse to reach the target position; calculate and set the current or voltage required for the motor drive based on the type, load and speed of the motor; ensure that the motor can move accurately according to the set target position. The trained motor position control prediction model is used to predict and analyze the motor drive control parameters. Specifically, the motor drive control parameters (such as speed, direction, drive current / voltage, etc.) are input into the motor position control prediction model. The model predicts the future position of the motor based on the input parameters, that is, the motor predicted position information is obtained. According to the predicted position information, the control strategy of the servo valve system is adjusted in real time to ensure that the motor can accurately reach the target position. Alternatively, by analyzing the difference between the predicted position information and the actual position, possible faults or abnormal conditions of the motor can be predicted, and corresponding measures can be taken in advance to prevent or repair them, thereby achieving accurate prediction and optimization of motor position control.

[0032] In a possible implementation, step S400 further includes step S410, extracting driving features from the motor motion characteristic data stream, obtaining a motor drive associated control parameter set, and using the motor drive associated control parameter set as a parameter affecting an independent variable. Extracting features directly related to the motor drive performance from the motor motion characteristic data stream, which may include the dynamic response characteristics, stability, efficiency, etc. of the motor, to form a motor drive associated control parameter set, which is used to adjust and optimize the control strategy of the motor, and the motor drive associated control parameter set is regarded as an independent variable, that is, they are variables that can be adjusted or optimized, and by changing the values ​​of these parameters, the driving performance of the motor can be affected, such as improving accuracy, reducing noise, optimizing efficiency, etc. It also includes step S420, extracting position parameters based on the motor motion characteristic data stream, obtaining a motor position parameter set, and using the motor position parameter set as a target parameter dependent variable. Specially extracting information directly related to the motor position from the motor motion characteristic data stream, which may include the current position of the motor, the angle turned, the speed of rotation, etc., to form a motor position data set, which describes the position and motion state of the motor at different times or different working conditions.

[0033] Step S400 also includes step S430, which performs regression fitting on the parameter-affecting independent variable and the target parameter dependent variable to construct a drive control parameter-motor position regression relationship. Regression fitting is used to determine the relationship between the independent variable and the dependent variable, that is, a regression algorithm is used to fit the relationship between the control drive control parameter and the motor position. Specifically, by collecting a series of experimental data or historical data, a series of observation values ​​of independent variables (control parameters) and dependent variables (motor position) can be obtained. Then, a regression algorithm (such as linear regression, polynomial regression, neural network regression, etc.) is used to estimate a function to obtain a drive control parameter-motor position regression relationship, which describes how to adjust the control parameter (independent variable) to affect the position of the motor (dependent variable). It also includes step S440, which calculates and analyzes the motor target position based on the drive control parameter-motor position regression relationship to determine the motor drive control parameter. Starting from the initial position, the drive control parameter is gradually adjusted, the response of the motor position is observed, and the next drive control parameter is adjusted according to the regression relationship until the position of the motor approaches or reaches the target position, and the motor drive control parameter required to reach the target position is determined.

[0034] In a possible implementation, step S440 further includes step S441, performing threshold calculation on the motor target position based on the drive control parameter-motor position regression relationship, and determining the drive control parameter selection threshold. Based on the performance requirements of the system (such as accuracy, stability, response time, etc.), define the allowable deviation range of the motor position relative to the target position, that is, the threshold, use the regression relationship, predict the position of the motor according to the current drive control parameters, compare the predicted position with the target position, calculate the position deviation, and determine whether the position deviation exceeds the set threshold range, and then determine the drive control parameter selection threshold to ensure that the appropriate control parameters can be found quickly and accurately in the subsequent control process. It also includes step S442, obtaining a motor control target set, evaluating and fitting the motor control target set, and generating a motor control effect fitness function. The motor control target set may include motor control cost, control rate, etc., which are evaluated and fitted to generate a motor control effect fitness function, which is used to evaluate the motor control effect. The smaller the value of the fitness function, the better the performance of the motor under the current control parameters, and the closer it is to the requirements of the control target set. The method further includes step S443, wherein the motor control effect fitness function is used to perform parameter evaluation and optimization within the drive control parameter selection threshold to determine the motor drive control parameters. Within the drive control parameter selection threshold, the effect of different drive control parameter combinations is evaluated by calculating the value of the fitness function, and the optimal parameter combination is selected as the motor drive control parameter.

[0035] Step S500, based on the motor drive control parameters and the motor predicted position information, the motor position control prediction model is calibrated and optimized to obtain a motor position calibration optimization model, and the target angle motor is position calibrated and controlled based on the motor position calibration optimization model. By combining the motor drive control parameters (such as speed, direction, drive current / voltage, etc.) and the motor predicted position information, the accuracy and effectiveness of the current motor position control prediction model are evaluated, the actual motor drive control parameters are input into the motor position control prediction model to obtain the predicted position information, the predicted position information is compared with the actual motor position information, the difference and error between the two are analyzed, and the motor position control prediction model is calibrated and optimized to improve its prediction accuracy and robustness. Specifically, the source of the error between the predicted position information and the actual position information is analyzed, which may be inaccurate model parameter settings, unreasonable model structure, data noise, etc. According to the error analysis results, the model parameters, such as the weights and bias terms of the neural network, are adjusted to improve the prediction performance of the model; if the model structure is unreasonable, it is possible to consider increasing or decreasing the number of network layers, changing the number of neurons, or adopting a more advanced network structure; the adjusted model is reapplied to the motor position prediction, and iterative optimization is continuously performed until the predetermined performance indicators are met or the upper limit of the number of iterations is reached, and finally a motor position calibration optimization model is obtained, which has higher accuracy and stability in predicting the motor position. The motor position control model after calibration and optimization is used to perform precise position calibration control on the target angle motor. That is, the target position that the target angle motor needs to reach is set, and the target position is converted into corresponding motor drive control parameters, such as speed, direction, and drive current / voltage, and input into the motor position calibration optimization model. The motor position calibration optimization model predicts the future position of the motor based on the input control parameters, and performs real-time calibration control on the motor based on the difference between the predicted position and the actual position to ensure that the motor can accurately reach the target position. The model predictive control technology and real-time calibration control strategy are combined to achieve high-precision control of the motor position.

[0036] In a possible implementation, step S500 further includes step S510, controlling and monitoring the target angle motor based on the motor drive control parameters to obtain actual motor feedback position information. The target angle motor is driven to move by using the predetermined motor drive control parameters, and the actual position information of the motor during the movement is obtained by corresponding monitoring means. The optimal motor drive control parameters determined by the parameter evaluation optimization are set as the drive control parameters of the target angle motor, and the set motor drive control parameters are sent to the motor controller. The controller generates corresponding control signals according to these parameters, drives the target angle motor to move according to a preset trajectory or mode, and collects the position information of the motor in real time through the built-in encoder, sensor or other monitoring equipment of the motor. The position information can be absolute position, relative position, angle change, etc. The collected motor position information is processed and analyzed to obtain the actual motor feedback position information, which is used for comparison with the preset target position to evaluate the control accuracy and performance of the motor. It also includes step S520, taking the deviation between the motor predicted position information and the actual motor feedback position information as the motor deviation position variable. Compare the key position of the motor predicted by the regression relationship of the drive control parameters with the actual position measured by the sensor and other equipment to obtain the difference between the two, that is, the motor deviation position variable. It also includes step S530, calibrating and optimizing the motor position control prediction model based on the motor deviation position variable to obtain the motor position calibration optimization model. According to the motor deviation position variable, the motor position control prediction model is calibrated, including adjusting model parameters, optimizing model structure, etc., for example, using the least squares method or gradient descent, etc., to optimize the model parameters according to the motor deviation position variable, so that the position predicted by the model is closer to the actual position.

[0037] In the above, refer to Figure 1 The position calibration optimization method for a servo valve with a limited angle motor according to an embodiment of the present invention is described in detail. Figure 2 A position calibration optimization system for a limited-angle motor for a servo valve according to an embodiment of the present invention is described.

[0038] The position calibration optimization system for a limited angle motor for a servo valve according to an embodiment of the present invention is used to solve the technical problem that it is difficult to accurately predict and control the motor position and drive control parameters according to the application characteristics of the motor, which leads to poor position control accuracy and motor control stability of the angle motor, and achieves the technical effect of improving the position control accuracy of the angle motor and the stability and reliability of the motor control. The position calibration optimization system for a limited angle motor for a servo valve includes: a motor motion characteristic data stream acquisition module 10, a motor application characteristic acquisition module 20, a motor position control prediction model generation module 30, a motor prediction position information determination module 40, and a motor position calibration optimization model acquisition module 50.

[0039] A motor motion characteristic data stream acquisition module 10, wherein the motor motion characteristic data stream acquisition module 10 is used to acquire a motor test parameter table, and based on the motor test parameter table and a measurement sensor device, test and collect the target angle motor to acquire the motor motion characteristic data stream, wherein the motor motion characteristic data stream includes the motor position, speed, acceleration, and current and voltage;

[0040] A motor application characteristic acquisition module 20, the motor application characteristic acquisition module 20 is used to pre-process the motor motion characteristic data stream, obtain the motor standard motion characteristic data stream, perform characteristic analysis based on the motor standard motion characteristic data stream, and obtain the motor application characteristics, the motor application characteristics including motor nonlinear factors and motor dynamic characteristics;

[0041] A motor position control prediction model generation module 30, wherein the motor position control prediction model generation module 30 is used to perform position prediction training on the motor motion characteristic data stream and the motor application characteristics using a deep neural network structure to generate a motor position control prediction model;

[0042] A motor predicted position information determination module 40, the motor predicted position information determination module 40 is used to set the motor target position according to the servo valve system requirements, perform motor drive parameter analysis based on the motor target position, determine the motor drive control parameters, and perform predictive analysis on the motor drive control parameters based on the motor position control prediction model to obtain motor predicted position information;

[0043] The motor position calibration optimization model acquisition module 50 is used to calibrate and optimize the motor position control prediction model based on the motor drive control parameters and the motor predicted position information, obtain the motor position calibration optimization model, and perform position calibration control on the target angle motor based on the motor position calibration optimization model.

[0044] The specific configuration of the motor motion characteristic data stream acquisition module 10 will be described in detail below. The motor motion characteristic data stream acquisition module 10 further includes: obtaining the application scenario information and control requirement information of the target angle motor; extracting test elements from the application scenario information and control requirement information to obtain application scenario requirement elements and control requirement elements; designing motor test parameters based on the application scenario requirement elements and control requirement elements to obtain a motor scenario test parameter set and a control requirement test parameter set; orthogonally arranging the motor scenario test parameter set and the control requirement test parameter set to obtain the motor test parameter table.

[0045] The specific configuration of the motor position control prediction model generation module 30 will be described in detail below. The motor position control prediction model generation module 30 may further include: using a deep neural network structure to perform position control analysis on the motor motion characteristic data stream to obtain a basic motor position control model; performing influence update training on the basic motor position control model based on the motor application characteristics to obtain a motor position update prediction model; performing multi-dimensional evaluation and verification on the motor position update prediction model to obtain model performance evaluation parameters, and determining the model optimization direction according to the model performance evaluation parameters; optimizing the motor position update prediction model based on the model optimization direction to generate the motor position control prediction model.

[0046] The specific configuration of the motor position control prediction model generation module 30 will be described in detail below. The motor position control prediction model generation module 30 may further include: performing an optimization strategy analysis on the model optimization direction to determine the model optimization strategy parameters; extracting the model parameter information of the motor position update prediction model, optimizing and updating the model parameter information based on the model optimization strategy parameters, and evaluating and iterating the optimized model parameter information through the fitness loss function to determine the optimal model parameters; configuring and optimizing the motor position update prediction model based on the optimal model parameters to obtain the motor position control prediction model.

[0047] The specific configuration of the motor predicted position information determination module 40 will be described in detail below. The motor predicted position information determination module 40 further includes: extracting driving features from the motor motion characteristic data stream to obtain a motor drive associated control parameter set, and using the motor drive associated control parameter set as a parameter influencing independent variable; extracting position parameters based on the motor motion characteristic data stream to obtain a motor position parameter set, and using the motor position parameter set as a target parameter dependent variable; performing regression fitting on the parameter influencing independent variable and the target parameter dependent variable to construct a drive control parameter-motor position regression relationship; performing calculation and analysis on the motor target position based on the drive control parameter-motor position regression relationship to determine the motor drive control parameter.

[0048] The specific configuration of the motor prediction position information determination module 40 will be described in detail below. The motor prediction position information determination module 40 further includes: performing threshold calculation on the motor target position based on the drive control parameter-motor position regression relationship to determine the drive control parameter selection threshold; obtaining a motor control target set, evaluating and fitting the motor control target set, and generating a motor control effect fitness function; using the motor control effect fitness function to perform parameter evaluation and optimization within the drive control parameter selection threshold to determine the motor drive control parameters.

[0049] The specific configuration of the motor position calibration optimization model acquisition module 50 will be described in detail below. The motor position calibration optimization model acquisition module 50 may further include: controlling and monitoring the target angle motor based on the motor drive control parameters to obtain actual motor feedback position information; using the deviation between the motor prediction position information and the actual motor feedback position information as a motor deviation position variable; calibrating and optimizing the motor position control prediction model based on the motor deviation position variable to obtain the motor position calibration optimization model.

[0050] The position calibration optimization system of the limited-angle motor for a servo valve provided in the embodiment of the present invention can execute the position calibration optimization method of the limited-angle motor for a servo valve provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0051] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0052] The above specific implementations do not constitute a limitation on the protection scope of this application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of this application should be included in the protection scope of this application.

Claims

1. A method for optimizing position calibration of a limited angle motor for a servo valve, characterized in that: The method comprises: Obtain a motor test parameter table, test and collect the target angle motor based on the motor test parameter table and the measurement sensor device, and obtain the motor motion characteristic data stream, wherein the motor motion characteristic data stream includes the motor position, speed, acceleration, current and voltage; Preprocessing the motor motion characteristic data stream to obtain a motor standard motion characteristic data stream, performing characteristic analysis based on the motor standard motion characteristic data stream to obtain motor application characteristics, wherein the motor application characteristics include motor nonlinear factors and motor dynamic characteristics; Performing position prediction training on the motor motion characteristic data stream and the motor application characteristics using a deep neural network structure to generate a motor position control prediction model; The motor target position is set according to the servo valve system requirements, the motor drive parameter analysis is performed based on the motor target position, the motor drive control parameter is determined, and the motor drive control parameter is predicted and analyzed based on the motor position control prediction model to obtain the motor predicted position information; Based on the motor drive control parameters and the motor predicted position information, the motor position control prediction model is calibrated and optimized to obtain a motor position calibration optimization model, and the target angle motor is position calibrated and controlled based on the motor position calibration optimization model; Wherein, generating the motor position control prediction model includes: Performing position control analysis on the motor motion characteristic data stream using a deep neural network structure to obtain a basic motor position control model; Performing influence update training on the basic motor position control model based on the motor application characteristics to obtain a motor position update prediction model; Performing multi-dimensional evaluation and verification on the motor position update prediction model to obtain model performance evaluation parameters, and determining the model optimization direction according to the model performance evaluation parameters; Optimizing the motor position update prediction model based on the model optimization direction to generate the motor position control prediction model; Wherein, generating the motor position control prediction model includes: Performing optimization strategy analysis on the optimization direction of the model and determining model optimization strategy parameters; Extracting and obtaining model parameter information of the motor position update prediction model, optimizing and updating the model parameter information based on the model optimization strategy parameters, and evaluating and iteratively updating the optimized model parameter information through a fitness loss function to determine the optimal model parameters; Based on the optimal parameters of the model, the motor position update prediction model is configured and optimized to obtain the motor position control prediction model; Wherein, determining the motor drive control parameters includes: Extracting driving characteristics from the motor motion characteristic data stream to obtain a motor drive associated control parameter set, and using the motor drive associated control parameter set as a parameter influencing independent variable; Extracting position parameters based on the motor motion characteristic data stream to obtain a motor position parameter set, and using the motor position parameter set as a target parameter dependent variable; Perform regression fitting on the parameter-affecting independent variable and the target parameter-dependent variable to construct a drive control parameter-motor position regression relationship; Calculating and analyzing the motor target position based on the drive control parameter-motor position regression relationship to determine the motor drive control parameter; Wherein, determining the motor drive control parameter includes: Performing threshold calculation on the motor target position based on the drive control parameter-motor position regression relationship to determine a drive control parameter selection threshold; Acquire a motor control target set, evaluate and fit the motor control target set, and generate a motor control effect fitness function; The motor control effect fitness function is used to perform parameter evaluation and optimization within the drive control parameter selection threshold to determine the motor drive control parameters.

2. The method for optimizing the position calibration of a limited-angle motor for a servo valve according to claim 1, characterized in that: The step of obtaining the motor test parameter table includes: Acquiring application scenario information and control requirement information of the target angle motor; Extracting test elements from the application scenario information and the control requirement information to obtain application scenario requirement elements and control requirement elements; Design motor test parameters based on the application scenario requirement elements and control requirement elements to obtain a motor scenario test parameter set and a control requirement test parameter set; The motor scenario test parameter set and the control demand test parameter set are orthogonally arranged to obtain the motor test parameter table.

3. The method for optimizing the position calibration of a limited-angle motor for a servo valve according to claim 1, characterized in that: The step of obtaining the motor position calibration optimization model comprises: Control and monitor the target angle motor based on the motor drive control parameters to obtain actual motor feedback position information; Taking the deviation between the predicted motor position information and the actual motor feedback position information as a motor deviation position variable; The motor position control prediction model is calibrated and optimized based on the motor deviation position variable to obtain the motor position calibration optimization model.

4. Position calibration optimization system for limited angle motor for servo valve, characterized in that: The system is used to implement the position calibration optimization method for a limited-angle motor for a servo valve according to any one of claims 1 to 3, and the system comprises: A motor motion characteristic data stream acquisition module, wherein the motor motion characteristic data stream acquisition module is used to acquire a motor test parameter table, and based on the motor test parameter table and a measurement sensor device, test and collect the target angle motor to acquire the motor motion characteristic data stream, wherein the motor motion characteristic data stream includes the motor position, speed, acceleration, and current and voltage; A motor application characteristic acquisition module, the motor application characteristic acquisition module is used to pre-process the motor motion characteristic data stream, obtain the motor standard motion characteristic data stream, perform characteristic analysis based on the motor standard motion characteristic data stream, and obtain the motor application characteristics, the motor application characteristics including motor nonlinear factors and motor dynamic characteristics; A motor position control prediction model generation module, the motor position control prediction model generation module is used to use a deep neural network structure to perform position prediction training on the motor motion characteristic data stream and the motor application characteristics to generate a motor position control prediction model; A motor predicted position information determination module, the motor predicted position information determination module is used to set the motor target position according to the servo valve system requirements, perform motor drive parameter analysis based on the motor target position, determine the motor drive control parameters, and perform predictive analysis on the motor drive control parameters based on the motor position control prediction model to obtain the motor predicted position information; A motor position calibration optimization model acquisition module is used to calibrate and optimize the motor position control prediction model based on the motor drive control parameters and the motor predicted position information, obtain a motor position calibration optimization model, and perform position calibration control on the target angle motor based on the motor position calibration optimization model.