Hydraulic valve transmission control system based on stepping motor and control method thereof

Abstract

The invention relates to the technical field of electromechanical control, and discloses a hydraulic valve transmission control system based on a stepping motor and a control method thereof, and the system comprises a physical parameter identification module, a health degree monitoring and compensation module, a fuzzy current adjustment module, a resonance suppression module and a driving execution module. The physical parameter identification module identifies the main resonant frequency, the real-time damping ratio and the real-time dynamic backlash of the transmission mechanism on line by injecting perturbation signals. And the health degree monitoring and compensation module fuses the real-time dynamic backlash and the long-term wear backlash predicted by the long-short-term memory network to generate an accurate compensation instruction. And the fuzzy current adjusting module dynamically reconstructs a fuzzy control membership function according to the real-time damping ratio, and adaptively adjusts the driving current. The method aims at solving the problem that an existing control method cannot adapt to physical characteristic changes of the transmission mechanism, and has the beneficial effects of being high in control accuracy, high in self-adaptive capacity and good in operation stability.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of electromechanical control, in particular to a hydraulic valve transmission control system based on a stepping motor and a control method thereof. BACKGROUND

[0002] The stepping motor has been widely used in precise positioning control occasions due to its simple structure, low cost and easy-to-implement open-loop digital control, such as driving the valve core of a hydraulic valve to perform precise displacement. In these applications, the stepping motor converts rotary motion into linear motion of the valve core through a gear-rack, screw-nut or other transmission mechanism. However, due to the existence of the mechanical transmission chain, the control accuracy and long-term stability of the entire system face significant challenges.

[0003] The transmission control method in the prior art usually relies on a fixed parameter model based on an ideal state established in the design or debugging stage. The controller generates driving instructions according to these pre-set fixed parameters. However, in actual operation, the physical characteristics of the transmission mechanism, such as load, friction, damping and structural resonance frequency, will dynamically change with working temperature, load variation and long-term mechanical wear. This gradual mismatch between the control model and the physical reality results in that the fixed parameter controller cannot continuously maintain optimal control performance, and the control accuracy and working condition adaptive ability of the system are therefore limited.

[0004] Specifically, this mismatch is particularly prominent in two key aspects. Firstly, in the transmission backlash compensation aspect, the traditional method usually adopts a fixed compensation value to offset the backlash. This approach ignores the fact that the backlash will continuously increase due to long-term operation. The fixed compensation value cannot track this gradual change, resulting in inaccurate compensation and easily causing impact and vibration when the motor reverses, which not only reduces the positioning accuracy, but also in turn aggravates the wear of the transmission mechanism, shortening its effective service life.

[0005] Secondly, in the aspect of suppressing mechanical resonance, the prior art often uses a fixed frequency notch filter or other methods to suppress the inherent mechanical resonance in the transmission chain. However, the actual resonance frequency of the system will drift due to load changes or structural parameter changes. When the resonance point deviates from the fixed frequency pre-set by the filter, the suppression effect will be greatly reduced, or even completely ineffective. This makes the system still produce harmful vibration and noise during fast positioning, seriously affecting the smoothness of the device operation and reducing the overall stability and reliability.

[0006] Therefore, how to enable the control system to real-time perceive the physical state changes of the transmission mechanism and dynamically adjust the control strategy accordingly to overcome the limitations of the traditional fixed parameter control method is a technical problem to be solved in the field. SUMMARY

[0007] In order to solve the problems of the prior art, the application provides a hydraulic valve transmission control system based on a stepping motor and a control method thereof, which solves the problem that the control precision is reduced, the gear gap compensation is inaccurate and the resonance suppression is ineffective due to the dynamic change of physical characteristics of the transmission mechanism caused by wear, load change and other factors.

[0008] In order to achieve the above object, the application provides a hydraulic valve transmission control device based on a stepping motor and a control method thereof, which realizes accurate monitoring of transmission health, adaptive adjustment of dynamic current and active suppression of system resonance by online acquisition of real-time physical parameters of the system.

[0009] The application provides a hydraulic valve transmission control system based on a stepping motor in a first aspect.

[0010] The device comprises a physical parameter identification module, a health monitoring and compensation module, a fuzzy current adjustment module, a resonance suppression module and a driving execution module.

[0011] The physical parameter identification module is used for injecting a perturbation excitation signal into the stepping motor, collecting a response signal of the stepping motor, and performing frequency domain analysis on the response signal to identify real-time physical parameters of the transmission mechanism. The real-time physical parameters include a main resonance frequency, a real-time damping ratio and a real-time dynamic gear gap. The energy of the perturbation excitation signal is controlled to ensure that the electromagnetic torque generated thereby only excites micro-vibration of the transmission chain and does not cause macro-displacement of the valve core.

[0012] The health monitoring and compensation module is used for outputting a long-term wear prediction gear gap value through a long short-term memory network model, fusing the real-time dynamic gear gap identified by the physical parameter identification module and the long-term wear prediction gear gap value, generating a final compensation gear gap value, and generating a compensation control instruction for compensating control of the stepping motor according to the final compensation gear gap value.

[0013] The fuzzy current adjustment module is used for dynamically reconstructing a membership function in a fuzzy control model according to the real-time damping ratio identified by the physical parameter identification module, and generating a target driving current for driving the stepping motor according to the fuzzy control model.

[0014] The resonance suppression module is used for generating an active suppression signal which is the same as the main resonance frequency and opposite in phase according to the main resonance frequency identified by the physical parameter identification module, and superimposing the active suppression signal and a main driving signal to form a final driving signal.

[0015] The drive execution module is configured to receive the compensation control instruction generated by the health degree monitoring and compensation module, the target drive current generated by the fuzzy current regulation module, and the final drive signal generated by the resonance suppression module, and drive the stepper motor to perform corresponding actions.

[0016] Preferably, the physical parameter identification module obtains a frequency response function by performing fast Fourier transform on the response signal, and determines the main resonance frequency, the real-time damping ratio, and the real-time dynamic tooth gap by analyzing a peak point, a half-power bandwidth, and a high harmonic component of the frequency response function, respectively.

[0017] The frequency response function is calculated by the following formula:

[0018] ;

[0019] wherein, represents fast Fourier transform; is the collected response signal; is the injected perturbation excitation signal; is the frequency response function.

[0020] The main resonance frequency is determined by the following formula:

[0021] ;

[0022] wherein, is the resonance frequency; is a mathematical operator, read as the independent variable of the maximum value; is the frequency response function of the system; represents the input frequency; represents the imaginary unit; is a modulus operator, which represents calculating the modulus or absolute value of a complex number.

[0023] The real-time damping ratio is calculated by the half-power bandwidth method of the resonance peak, that is, if the amplitude on both sides of the resonance peak drops to the frequency at the peak value is and , then:

[0024] ;

[0025] wherein, is the real-time damping ratio; is the resonance frequency; is the lower half-power point frequency; is the higher half-power point frequency.​​ is the half-power bandwidth.

[0026] the real-time dynamic backlash by analyzing the ratio of the high-order harmonic energy caused by the backlash nonlinearity in the response signal spectrum to the fundamental energy and mapping to a pre-established model determines:

[0027] ;

[0028] wherein, is the physical backlash; is a pre-established mapping function or model; is the harmonic energy; is the fundamental energy.

[0029] In one specific embodiment, the health monitoring and compensation module fuses the real-time dynamic backlash and the long-term wear prediction backlash value through a dynamic weighting algorithm. When the real-time dynamic backlash suddenly changes, the dynamic weighting algorithm increases the weight of the real-time dynamic backlash in the fusion calculation. The final compensation backlash value is calculated by the following formula:

[0030] ;

[0031] wherein, is the real-time dynamic backlash, is the long-term wear prediction backlash value; is the final compensation backlash value; is a dynamic weight factor.

[0032] In one specific embodiment, the health monitoring and compensation module further includes using the real-time dynamic backlash as a reference to online calibrate the long short-term memory network model. The long-term wear prediction backlash value output by the long short-term memory network model is represented by the following formula:

[0033] ;

[0034] wherein, is the stepper motor operating state vector, is the vibration sensor signal; is the long-term wear prediction backlash value; represents a long short-term memory network model.

[0035] Preferably, the fuzzy current regulation module realizes the dynamic reconstruction of the membership function by setting the boundary parameters of the membership function in the fuzzy control model as a function related to the real-time damping ratio. For example, a triangular membership function is used to describe the input variable . is defined as:

[0036] ;

[0037] wherein, , , are the characteristic point parameters of the membership function, which are all functions of the real-time damping ratio ; is the input variable, i.e. the real-time load at time ; is the membership symbol, which takes a value in the interval , indicating the true degree of an input belonging to a certain fuzzy set; is the i-th fuzzy set; is a mathematical technique for constructing a triangular shape, the inner function takes the smaller one of the calculation results of two straight lines, and the outer function ensures that the value of the entire membership will never be less than 0; represents the horizontal distance of the current real-time load value relative to the left foot point of the triangle; represents the width of the rising slope; represents the position of the current load value on the rising slope; represents the horizontal distance of the current real-time load value relative to the right foot point of the triangle; represents the width of the falling slope; represents the position of the current load value on the falling slope. Preferably, the main drive signal is formed by combining the compensation control instruction and the target drive current of the stepper motor. The resonance suppression module superimposes the active suppression signal onto the main drive signal to form a final drive signal, so that the stepper motor executes the positioning instruction while suppressing the mechanical resonance of the transmission mechanism. The synthesis formula of the final drive signal

[0038] is:

[0039] ;

[0040] wherein, is the main drive signal;​ active suppression signal; final drive signal.

[0041] the active suppression signal is expressed by the following formula:

[0042] ;

[0043] wherein, is the suppression signal amplitude; is the main resonance frequency; is the suppression signal phase; is the active suppression signal; is a sinusoidal function; indicates that the suppression signal is a dynamic signal evolving over time.

[0044] Preferably, the real-time damping ratio identified by the physical parameter identification module is only used for the fuzzy current regulation module to dynamically reconstruct the membership function in the fuzzy control model; and the main resonance frequency identified by the physical parameter identification module is only used for the resonance suppression module to generate the active suppression signal.

[0045] Preferably, the real-time dynamic backlash identified by the physical parameter identification module is taken as a real-time physical state feedback of the transmission mechanism, which is used to correct the prediction drift of the long-term wear prediction backlash value by the long short-term memory network model, and to improve the response speed of the health degree monitoring and compensation module to sudden transmission state changes when fusing to generate the final compensation backlash value.

[0046] Preferably, the drive execution module generates and outputs the final drive signal to the stepper motor, the stepper motor converts the final drive signal into a step-by-step rotary motion and outputs it to the transmission mechanism, and the transmission mechanism converts the step-by-step rotary motion into a linear or angular displacement of the hydraulic valve spool to realize the control of the hydraulic valve.

[0047] The second aspect of the present application provides a hydraulic valve transmission control method based on a stepper motor.

[0048] The method comprises the following steps:

[0049] S1, injecting a perturbation excitation signal into a stepper motor, collecting a response signal of the stepper motor, and identifying real-time physical parameters of a transmission system, the real-time physical parameters including a main resonance frequency, a real-time damping ratio, and a real-time dynamic backlash;

[0050] S2, fusing the long-term wear prediction gap value output by the long short-term memory network model with the real-time dynamic gap identified in the S1 step to generate a final compensation gap value;

[0051] S3, dynamically reconstructing the membership function in the fuzzy control model according to the real-time damping ratio identified in the S1 step to generate a target drive current for constituting a main drive signal;

[0052] S4, generating an active suppression signal according to the main resonance frequency identified in the S1 step, and superimposing the active suppression signal on the main drive signal combined with the final compensation gap value generated in the S2 step and the target drive current generated in the S3 step to drive the stepper motor.

[0053] The application provides a hydraulic valve transmission control system based on a stepper motor and a control method thereof.

[0054] 1、The physical parameter identification module can identify the key physical parameters of the transmission mechanism, such as the main resonance frequency, the real-time damping ratio and the real-time dynamic gap. The fuzzy current regulation module and the resonance suppression module dynamically adjust their control strategies according to these real-time changing parameters, so that the drive current output and vibration suppression behavior of the device can always match the current physical state of the transmission mechanism, overcoming the performance degradation problem caused by parameter solidification in traditional control methods, and improving the control precision of the system under different working conditions and wear stages.

[0055] 2、The health monitoring and compensation module fuses the real-time dynamic gap reflecting the instantaneous state identified by the physical parameter identification module and the long-term wear prediction gap value reflecting the long-term trend output by the long short-term memory network model, so that the gap compensation can quickly respond to sudden changes and accurately track long-term wear, achieving more accurate compensation control, reducing the commutation impact, reducing mechanical wear and prolonging the effective service life of the transmission mechanism.

[0056] 3、The resonance suppression module generates an active suppression signal with the same system resonance frequency and opposite phase using the main resonance frequency provided by the physical parameter identification module. The signal is superimposed with the main drive signal, which can cancel the mechanical resonance generated by the transmission chain during positioning, avoid vibration, noise and positioning errors caused by resonance, ensure the smoothness of hydraulic valve control, and enhance the stability and reliability of the entire device operation. BRIEF DESCRIPTION OF DRAWINGS

[0057] Figure 1 The control device structure framework of the application is shown in the figure;

[0058] Figure 2The control method flow chart of the present application;

[0059] Figure 3 The control device structure schematic diagram of the present application;

[0060] Figure 4 The control device transmission mechanism structure schematic diagram of the present application;

[0061] Figure 5 The internal structure schematic diagram of the control device of the present application.

[0062] Wherein, 10, physical parameter identification module; 20, health degree monitoring and compensation module; 30, fuzzy current regulation module; 40, resonance suppression module; 50, drive execution module; 60, stepping motor; 70, transmission mechanism; 80, hydraulic valve. DETAILED DESCRIPTION

[0063] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the specification of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the protection scope of the present application.

[0064] Referring to the drawings Figure 1 , Figure 1 It is the control device structure framework diagram of the stepping motor based hydraulic valve transmission control device according to an embodiment of the present application. The present application provides a stepping motor based hydraulic valve transmission control system, which comprises a physical parameter identification module 10, a health degree monitoring and compensation module 20, a fuzzy current regulation module 30, a resonance suppression module 40 and a drive execution module 50.

[0065] In an embodiment, the physical parameter identification module 10 serves as the perception unit of the system, and its function is to inject a perturbation excitation signal into the stepping motor 60 and collect the response signal of the stepping motor 60. Through frequency domain analysis on the response signal, three real-time physical parameters of the transmission mechanism 70 connected to the stepping motor 60 are identified, i.e., the main resonance frequency , the real-time damping ratio and the real-time dynamic backlash .

[0066] The output of the physical parameter identification module 10 is connected to the function input ends of the health degree monitoring and compensation module 20, the fuzzy current regulation module 30 and the resonance suppression module 40. Specifically, the identified real-time dynamic backlash is transmitted to the health degree monitoring and compensation module 20, and the identified real-time damping ratio The identified main resonance frequency is transmitted to the fuzzy current regulation module 30. The identified main resonance frequency is transmitted to the resonance suppression module 40.

[0067] The health monitoring and compensation module 20 is used to generate a final compensation gap value according to the received real-time dynamic gap value , and the long-term wear prediction gap value output by the internal long-short term memory network model of the health monitoring and compensation module 20 through a dynamic weighted fusion algorithm. Subsequently, the health monitoring and compensation module 20 generates a compensation control instruction for the stepper motor 60 according to the final compensation gap value, and the output end of the health monitoring and compensation module 20 is connected to the input end of the driving execution module 50.

[0068] The fuzzy current regulation module 30 is used to dynamically reconstruct the membership function in its internal fuzzy control model according to the received real-time damping ratio , and generate a target driving current for driving the stepper motor 60 according to the reconstructed fuzzy control model, and the output end of the fuzzy current regulation module 30 is connected to the input end of the driving execution module 50.

[0069] The resonance suppression module 40 is used to generate an active suppression signal according to the received main resonance frequency , and the frequency of the active suppression signal is the same as the main resonance frequency , and the phase is opposite, and the output end of the resonance suppression module 40 is connected to the input end of the driving execution module 50.

[0070] The driving execution module 50 is the final execution unit of the system, and its input end receives the compensation control instruction from the health monitoring and compensation module 20, the target driving current from the fuzzy current regulation module 30, and the active suppression signal from the resonance suppression module 40, and the driving execution module 50 synthesizes the three signals to form a final driving signal , and outputs it to the stepper motor 60.

[0071] After receiving the final driving signal , the stepper motor 60 converts it into precise stepping rotary motion, which is converted into linear displacement or angular displacement of the spool of the hydraulic valve 80 through the transmission mechanism 70, thereby realizing precise control of the opening degree of the hydraulic valve 80.

[0072] The general working principle of the embodiment is as follows: the physical parameter identification module 10 periodically or under specific instructions performs online identification on the transmission system, to provide real-time and accurate physical parameter references for the other three parallel control modules, the health degree monitoring and compensation module 20, the fuzzy current regulation module 30 and the resonance suppression module 40, based on the real-time parameters, the three modules respectively generate their own control strategies from the three dimensions of cogging compensation, drive current self-adaption and resonance suppression, and finally, the drive execution module 50 unifies the multiple-source and cooperative control strategies into the final drive signal, to realize high-precision, high-stability and high-adaptability control of the stepper motor 60 and the hydraulic valve 80.

[0073] Referring to the drawings Figure 1 , the following will be a detailed description of the core modules constituting the transmission control device based on the stepper motor 60 of the hydraulic valve 80 in the embodiment of the application.

[0074] In a specific embodiment, the physical parameter identification module 10 obtains the real-time physical parameters of the transmission mechanism 70 through an active and non-invasive online identification process, and the identification process of the physical parameter identification module 10 includes four steps of signal injection, signal collection, frequency domain analysis and parameter extraction. In the signal injection step, the physical parameter identification module 10 injects an energy-controlled perturbation excitation signal into the stepper motor 60 at rest or in a specific working condition , preferably, the perturbation excitation signal is a swept sine signal or a broadband white noise signal, and the amplitude is set to ensure that the electromagnetic torque generated thereby can only excite micro-vibration of the transmission mechanism 70, but is insufficient to cause macro-displacement of the hydraulic valve 80 spool. In the signal collection step, the physical parameter identification module 10 synchronously collects the phase current of the stepper motor 60 as a response signal through the current sensor connected with the drive circuit of the stepper motor 60 .

[0075] In the frequency domain analysis step, the physical parameter identification module 10 performs fast Fourier transform on the collected response signal and the injected perturbation excitation signal , to obtain the frequency response function of the system, and the calculation formula of the frequency response function is as follows:

[0076] ;

[0077] Wherein, represents fast Fourier transform; is the collected response signal; is the injected perturbation excitation signal; is the frequency response function.

[0078] In the parameter extraction step, the physical parameter identification module 10 analyzes the frequency response function, and the main resonance frequency is determined by finding the frequency corresponding to the peak point of the amplitude of the frequency response function The real-time damping ratio is calculated by using the half-power bandwidth method on the resonance peak The real-time dynamic backlash is determined by analyzing the ratio of the high-order harmonic energy to the fundamental energy caused by the nonlinear effect of the backlash in the response signal spectrum, and mapping it to a pre-established physical model The physical parameter identification module 10 finally identifies the main resonance frequency , real-time damping ratio and real-time dynamic backlash , which are transmitted to the corresponding functional modules in the subsequent.

[0079] The real-time dynamic backlash identified by the physical parameter identification module 10 is transmitted to the health monitoring and compensation module 20, and the function of the health monitoring and compensation module 20 is to fuse the backlash information of two different time scales to generate a final compensation backlash value that can quickly respond to mutations and smoothly track long-term wear .

[0080] In one embodiment, the health monitoring and compensation module 20 maintains a long short-term memory network model inside, which receives the real-time running state vector of the stepper motor 60 (including real-time current, voltage, and speed) and external vibration sensor signals as input, and outputs the long-term wear prediction backlash value of the transmission mechanism 70 due to long-term wear In order to enable the long short-term memory network model to make effective predictions, it needs to be trained offline.

[0081] In a preferred embodiment, its training process includes: first, by accelerating the wear test of the same type of transmission mechanism 70, or collecting historical running data of the equipment throughout its life cycle, a training data set is constructed, in which each data sample contains the motor running state vector , vibration sensor signals , and the physical backlash true value measured at that time point by high-precision measurement equipment Then, taking the minimization of the mean square error between the model prediction value and the physical backlash true value as the optimization goal, the network weights of the long short-term memory network model are iteratively optimized using the back propagation algorithm until the model converges. The trained model is solidified in the health monitoring and compensation module 20.

[0082] The health monitoring and compensation module 20 fuses the received real-time dynamic backlash with the long-term wear prediction backlash value output by the model to ultimately compensate the backlash value which is calculated by the following formula:

[0083] ;

[0084] wherein, is the real-time dynamic backlash, is the long-term wear prediction backlash value; is the ultimately compensated backlash value; is the dynamic weight factor.

[0085] The value of the weight factor is dynamically adjusted according to the change rate of the real-time dynamic backlash When the change rate of the real-time dynamic backlash exceeds a preset threshold, it indicates that the transmission state has changed abruptly, the value of the weight factor will increase, thereby increasing the proportion of the real-time measurement value in the fusion calculation, to ensure the rapid response of the compensation, and in stable working conditions, the value of the weight factor decreases, to increase the proportion of the long-term prediction value, thereby filtering out measurement noise and ensuring the smoothness of the compensation. In addition, the real-time dynamic backlash is also used as a reference value to periodically calibrate the output of the LSTM model online to correct the prediction drift that may occur. The health monitoring and compensation module 20 ultimately generates corresponding compensation control instructions according to the calculated and transmits them to the drive execution module 50.

[0086] At the same time, the real-time damping ratio identified by the physical parameter identification module 10 is transmitted to the fuzzy current regulation module 30, which is responsible for adaptively regulating the drive current of the stepper motor 60 according to the real-time changes in the system load characteristics.

[0087] In one embodiment, the fuzzy current regulation module 30 internally contains a fuzzy control model whose input is the real-time motor load and whose output is the adjustment amount for the target drive current. The core of this fuzzy control model lies in that its membership function is dynamically reconfigurable.

[0088] Specifically, the characteristic point parameters of the membership function used to describe fuzzy variables such as load are designed as functions related to the real-time damping ratio , so that a triangular membership function For example, it defines the three feature point parameters of a triangle shape. , , All are about The function, when the system damping ratio When the value increases, it means that the system load or resistance increases, and the boundary of the membership function will expand accordingly, so that the reasoning of the fuzzy controller can adapt to the physical characteristics of the current system. Based on this adaptive fuzzy control model, the fuzzy current adjustment module 30 generates the target driving current and transmits it to the driving execution module 50.

[0089] The main resonant frequency identified by the physical parameter identification module 10 The signal is then transmitted to the resonance suppression module 40. The function of the resonance suppression module 40 is to actively suppress the mechanical resonance generated by the transmission mechanism 70 during the precise positioning process. In one embodiment, the resonance suppression module 40 determines the resonance frequency based on the received main resonance frequency. This generates a sinusoidal signal with the same frequency but opposite phase, which serves as the active suppression signal. Active suppression signal Generate using the following formula:

[0090] ;

[0091] in, To suppress signal amplitude; The main resonant frequency; To suppress signal phase; This is an active suppression signal; It is a sine function; This indicates that the suppression signal is a dynamic signal that evolves over time.

[0092] This active suppression signal The control components generated by the health monitoring and compensation module 20, the fuzzy current adjustment module 30, and the resonance suppression module 40 are transmitted to the drive execution module 50. These components are then converged to the drive execution module 50 for signal synthesis and execution. In one embodiment, the drive execution module 50 first combines the compensation control command output by the health monitoring and compensation module 20 with the target drive current output by the fuzzy current adjustment module 30 to form the main drive signal. .

[0093] Subsequently, the active suppression signal output by the resonance suppression module 40 is... With the main drive signal The signals are superimposed to form the final driving signal. The final drive signal The signal is fed into the stepper motor 60 and converted into a physical electrical signal to drive the stepper motor 60, thereby completing the multi-dimensional, adaptive, and coordinated control of the entire transmission system.

[0094] See attached document Figure 2 , Figure 2 This is a flowchart of a hydraulic valve transmission control method based on a stepper motor according to an embodiment of the present invention. The present invention provides a hydraulic valve transmission control method based on a stepper motor, which achieves adaptive control of the transmission system through a coordinated, multi-step process. The method specifically includes the following steps:

[0095] S1, a perturbation excitation signal is injected into the stepper motor 60, the response signal of the stepper motor 60 is collected, and the real-time physical parameters of the transmission system are identified. In one embodiment, this step is performed by the physical parameter identification module 10. This step first injects an energy-controlled perturbation excitation signal into the stepper motor 60. Simultaneously, the phase current of the stepper motor 60 is collected as a response signal. Subsequently, through the analysis of and Perform a Fast Fourier Transform to obtain the system's frequency response function. Finally, by analyzing the frequency response function, the real-time physical parameters of the transmission mechanism 70 are identified. These real-time physical parameters include the main resonant frequency. Real-time damping ratio and real-time dynamic tooth backlash .

[0096] S2, based on the real-time dynamic backlash identified in step S1, is fused with the long-term wear prediction backlash value output by the Long Short-Term Memory (LSTM) network model to generate the final compensated backlash value. In one embodiment, this step is performed by the health monitoring and compensation module 20. This module first uses its internal LSM network model to output the long-term wear prediction backlash value of the transmission mechanism 70 caused by long-term wear, based on the real-time operating state vector of the stepper motor 60 and the external vibration sensor signal. Subsequently, the module employs a dynamic weighted algorithm to calculate the long-term wear prediction backlash value. Real-time dynamic backlash identified in step S1 The fusion is performed, and the calculation formula is as follows:

[0097] ;

[0098] in, For real-time dynamic tooth backlash, Predict backlash values ​​for long-term wear; To compensate for the backlash value; This is a dynamic weighting factor.

[0099] S3, based on the real-time damping ratio identified in step S1, dynamically reconstruct the membership function in the fuzzy control model to generate the target drive current used to constitute the main drive signal. In one embodiment, this step is performed by the fuzzy current adjustment module 30, which sets the feature point parameters of the membership function in the fuzzy control model used to describe fuzzy variables such as motor load to the real-time damping ratio identified in step S1. The relevant functions enable the fuzzy controller's inference process to adapt to the real-time changing physical characteristics of the transmission mechanism 70. The fuzzy current adjustment module 30 performs fuzzy inference based on this adaptive fuzzy control model, and finally generates a target drive current that matches the current operating conditions.

[0100] S4, based on the main resonant frequency identified in step S1, an active suppression signal is generated, and this active suppression signal is superimposed on the main drive signal, which combines the final compensation backlash value generated in step S2 and the target drive current generated in step S3, to drive the stepper motor 60. In one embodiment, this step is executed collaboratively by the resonance suppression module 40 and the drive execution module 50. First, the resonance suppression module 40 generates an active suppression signal based on the main resonant frequency identified in step S1. This generates a sinusoidal signal with the same frequency but opposite phase, which serves as the active suppression signal. Meanwhile, the drive execution module 50 will use the final compensation backlash value generated in step S2. (Converted into a compensation control command) and combined with the target drive current generated in step S3, forming the main drive signal. Finally, the drive execution module 50 will actively suppress the signal. With the main drive signal The signals are superimposed to form the final driving signal. It outputs the signal to the stepper motor 60, thereby achieving precise, stable, and adaptive control of the hydraulic valve 80.

Claims

1. A hydraulic valve actuation control system based on a stepper motor, characterized by, The physical parameter identification module injects a perturbation excitation signal into the stepper motor, collects a response signal of the stepper motor, and performs frequency domain analysis on the response signal to identify real-time physical parameters of the transmission mechanism, including a main resonance frequency, a real-time damping ratio, and a real-time dynamic backlash. The health monitoring and compensation module outputs a long-term wear prediction backlash value through a long short-term memory network model, fuses the real-time dynamic backlash identified by the physical parameter identification module and the long-term wear prediction backlash value to generate a final compensation backlash value, and generates a compensation control instruction for compensation control of the stepper motor according to the final compensation backlash value. The fuzzy current regulation module dynamically reconstructs a membership function in a fuzzy control model according to the real-time damping ratio identified by the physical parameter identification module, and generates a target drive current for driving the stepper motor according to the fuzzy control model. The resonance suppression module generates an active suppression signal that is the same as and opposite in phase to the main resonance frequency identified by the physical parameter identification module, and superimposes the active suppression signal on a main drive signal to form a final drive signal. The drive execution module receives the compensation control instruction, the target drive current of the stepper motor, and the final drive signal, and drives the stepper motor to perform corresponding actions. The physical parameter identification module obtains a frequency response function by performing fast Fourier transform on the response signal, and determines the main resonance frequency, the real-time damping ratio, and the real-time dynamic backlash by analyzing peak points, half-power bandwidth, and high harmonic components of the frequency response function.

2. A hydraulic valve actuation control system based on a stepper motor as set forth in claim 1, characterized in that, The health monitoring and compensation module fuses the real-time dynamic backlash and the long-term wear prediction backlash value through a dynamic weighting algorithm, wherein when the real-time dynamic backlash mutates, the weight of the real-time dynamic backlash in the fusion calculation is increased.

3. The hydraulic valve actuation control system based on a stepper motor of claim 1, wherein, The health monitoring and compensation module further includes the real-time dynamic backlash as a reference for online calibration of the long short-term memory network model.

4. A hydraulic valve actuation control system based on a stepper motor as set forth in claim 3, wherein The fuzzy current regulation module dynamically reconstructs the membership function in the fuzzy control model by setting boundary parameters of the membership function in the fuzzy control model as a function related to the real-time damping ratio.

5. The hydraulic valve actuation control system based on a stepper motor of claim 1, wherein, The main drive signal is formed after combining the compensation control instruction and the target drive current of the stepper motor; the resonance suppression module superimposes the active suppression signal on the main drive signal to form a final drive signal, so that the stepper motor executes a positioning instruction while suppressing mechanical resonance of the transmission mechanism.

6. A hydraulic valve actuation control system based on a stepper motor as set forth in claim 1, wherein, The real-time damping ratio identified by the physical parameter identification module is only used by the fuzzy current regulation module to dynamically reconstruct the membership function in the fuzzy control model; and the main resonance frequency identified by the physical parameter identification module is only used by the resonance suppression module to generate the active suppression signal.

7. The hydraulic valve actuation control system based on a stepper motor of claim 1, wherein, ​ 8. The hydraulic valve actuation control system based on a stepper motor of claim 1, wherein, The physical parameter identification module identifies the real-time dynamic backlash as a real-time physical state feedback of the transmission mechanism, which is used to correct the prediction drift of the long-term wear prediction backlash value by the long short-term memory network model, and improve the response speed of the health monitoring and compensation module to sudden transmission state changes when fusing to generate the final compensation backlash value.

9. The hydraulic valve actuation control system based on a stepper motor of claim 1, wherein, The drive execution module generates and outputs the final drive signal to the stepper motor, which converts the final drive signal into a step-by-step rotary motion and outputs it to the transmission mechanism, which converts the step-by-step rotary motion into a linear or angular displacement of the hydraulic valve spool to control the hydraulic valve.

10. A method of hydraulic valve actuation control based on a stepper motor, characterized by The method comprises the following steps: S1, injecting a perturbation excitation signal into the stepper motor, collecting the response signal of the stepper motor, and identifying the real-time physical parameters of the transmission system, including the main resonance frequency, the real-time damping ratio, and the real-time dynamic backlash; S2, according to the real-time dynamic backlash identified in the S1 step, fusing with the long-term wear prediction backlash value output by the long short-term memory network model to generate a final compensation backlash value; S3, according to the real-time damping ratio identified in the S1 step, dynamically reconstructing the membership function in the fuzzy control model to generate a target drive current for constituting a main drive signal; S4, according to the main resonance frequency identified in the S1 step, generating an active suppression signal, and superimposing the active suppression signal on the main drive signal combined with the final compensation backlash value generated in the S2 step and the target drive current generated in the S3 step to drive the stepper motor.

Images