Valve health degree monitoring and maintaining method and system based on electric actuator

Abstract

The invention relates to a valve health degree monitoring and maintaining method and system based on an electric actuator, and belongs to the technical field of mechanical control and automation. The method comprises the steps that valve maintenance data of the electric actuator are obtained, a position-pressure bimodal grid is constructed based on the valve maintenance data, pressure testing is conducted on a valve, torque-opening curves in the opening and closing processes are analyzed, and valve health degree data are obtained; an electric execution monitoring model is constructed, a valve torque curve is obtained, maintenance information labels are added to grid nodes according to the valve maintenance information through a labeling binding method, and a valve health maintenance scheme is generated; for the pressure test, identifying switching time delay, hysteresis and leakage data in a valve torque profile and an operation log through a curve feature extraction method to obtain a valve scheme optimization strategy; and executing the maintenance scheme optimized by the valve scheme optimization strategy, and automatically outputting the updated valve health maintenance scheme based on a recursive execution feedback method.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of mechanical control and automation, and particularly relates to a valve health degree monitoring and maintenance method and system based on an electric actuator. BACKGROUND

[0002] As a key component in the industrial control system, the valve is widely used in the fields of petroleum chemical industry, electric power, metallurgy and water treatment, and its running state is directly related to the safety and stability of the production process. The traditional valve relies on manual inspection and periodic maintenance, but due to the high frequency of valve opening and closing, the complex working environment, the easy wear of sealing parts and other factors, it is often difficult to find potential faults in time, causing leakage, jamming or failure, and thus causing equipment downtime or safety accidents. With the popularity of electric actuators, the automation control level of the valve has been improved, but how to monitor and evaluate the health degree of the valve driven by the electric actuator in real time is still the focus and difficulty of current research. The existing technology mainly monitors the valve operating parameters through a single way such as pressure sensor, current detection or displacement acquisition, lacks joint analysis of multi-modal data, and thus the identification of key parameters such as valve switching delay, hysteresis characteristics and leakage is not accurate enough. In addition, the existing monitoring method relies on experience threshold for judgment, and it is difficult to dynamically adapt to the degradation process of the valve in the long-term operation, especially when the sealing performance of the valve gradually decays over time, there is a lack of effective prediction and verification means. At the same time, the maintenance strategy usually relies on manual experience to formulate, lacks recursive feedback and optimization mechanism, and is easy to cause over-maintenance or insufficient maintenance, increasing the operation cost and equipment risk. Therefore, it is urgent to propose a valve health degree monitoring and maintenance method based on an electric actuator, which can construct a monitoring grid combining the position and pressure dual-modal data, extract the torque curve features, identify the key operating parameters, and realize the intelligent diagnosis and dynamic maintenance of the valve running state through feedback optimization and degradation cycle verification. SUMMARY

[0003] To solve the above problems in the prior art, the present application provides a valve health degree monitoring and maintenance method based on an electric actuator, The purpose of the present application can be achieved by the following technical solutions: S1: obtaining valve maintenance data of an electric actuator, constructing a position-pressure dual-modal grid based on the valve maintenance data, performing pressure test on the valve, analyzing the torque-opening curve in the opening and closing process, and evaluating the sealing performance of the valve according to the torque-opening curve to obtain valve health degree data; S2: Construct an electric actuator monitoring model based on the valve health data, which includes valve undisturbed response judgment and torque switching data, input the valve health data as monitoring parameters into the electric actuator monitoring model, output operation logs containing valve undisturbed response judgment and torque switching data, set valve maintenance information in combination with maintenance execution indicators, attach maintenance information labels to grid nodes through label binding method, update the position-pressure dual-mode grid, and generate a valve health maintenance scheme; S3: According to the valve health maintenance scheme, analyze the valve maintenance data of the electric actuator, identify valve torque profile, switching delay, hysteresis and leakage data in operation logs through curve feature extraction method, and verify valve seal degradation period based on electrostatic discharge control method to obtain valve scheme optimization strategy; S4: According to the current valve torque health data, execute the maintenance scheme optimized by the valve scheme optimization strategy, update the valve torque performance when the electric actuator monitoring model outputs the valve torque curve that meets the maintenance execution indicators, and return the updated valve torque to the electric actuator monitoring model based on recursive execution feedback method to automatically output the updated valve health maintenance scheme.

[0004] Specifically, the logical process of generating the valve health data is as follows: Based on the position-pressure dual-mode grid, the valve is segmented and pressure loaded, and the corresponding pressure response curve is collected at different opening positions. The valve opening and closing instruction issuing time and the actual pressure and torque curve response starting time are recorded to obtain the undisturbed response interval in the valve switching operation process. According to the undisturbed interval, the switching delay parameter is structured and stored, and in combination with the pressure response curve in the pressure loading process, a valve health data set is formed. The valve health data set is used as the basic input data source of the electric actuator monitoring model, the pressure response curve is reconstructed by interpolation fitting, and the valve health data is obtained.

[0005] Specifically, the method for obtaining the valve maintenance data is as follows: analyze the running parameters in the valve opening and closing process, the running parameters including torque data, pressure data, and operation log data, time sequence the running parameters and modal fusion, and in combination with the opening and closing instructions and feedback signals in the operation log data, generate complete valve maintenance data.

[0006] Specifically, the operation log generation method comprising valve disturbance-free response judgment and torque switching data is: at the time point of valve opening and closing instruction issuance, recording valve displacement and conduction pressure response signals, and according to valve response judgment, determining whether there is a delay, obtaining a disturbance-free response interval, during valve action, analyzing switching time delay parameters according to the disturbance-free response interval, and simultaneously sampling displacement-pressure time sequence data in the action process, identifying hysteresis characteristics and transient fluctuations, writing leakage change curves, action priorities and abnormality judgment marks in the process into the log, and forming a traceable operation log data.

[0007] Specifically, the construction method of the electric execution monitoring model is: Extracting torque, displacement and pressure original data characteristics in the valve opening and closing process, converting unstructured valve maintenance data into extracted switching time delay, hysteresis and leakage parameter sets, and combining switching time delay associated valve health in the operation log record to generate a valve health maintenance feature matrix; Based on the valve health maintenance feature matrix, receiving the time delay, hysteresis and leakage parameter set, processing valve maintenance monitoring features, and converging resource allocation results through curve feature extraction method, generating a motor execution maintenance scheme according to the disturbance-free response and switching time delay operation log; According to the valve operation execution data and feedback results collected in real time according to the maintenance allocation scheme, under the constraint of the position-pressure dual-mode grid, dynamically correcting the parameter weight of the electric execution monitoring model for valve health monitoring through error back propagation, and taking the valve health constraint condition as an external mapping parameter to construct the electric execution monitoring model.

[0008] Specifically, the process of updating the valve torque back to the electric execution monitoring model is: Obtaining the difference data between the valve health maintenance scheme generated by the model and the torque curve in the actual execution process; and based on the valve health maintenance scheme and the corrected valve switching response time, the valve torque switching response time corrected by the executed valve scheme optimization strategy is data encoded, when the deviation exceeds the preset threshold of the valve health maintenance scheme, the model optimization process is triggered; Based on the recursive execution feedback method in the model optimization process, the updated valve torque curve is input into the electric execution monitoring model as a correction parameter, and the optimization parameter is fused with the monitoring and maintenance allocation rules of the electric execution monitoring model to form an iterative input sequence; Based on the running result of the updated electric execution monitoring model, a new valve health maintenance scheme is automatically generated, and the optimized scheme is written into the health association database of the electric execution monitoring model through cyclic feedback.

[0009] Specifically, the pressure test verifies the torque curve scheme by simulating the valve micro-motion execution process through synchronous sampling of undisturbed valve switching response, and generates a valve scheme optimization strategy according to the test results.

[0010] Specifically, the verification method of the valve seal degradation period is: obtaining the leakage amount and pressure holding time operation data of the sealing part in the valve opening and closing process, and implementing an accelerated aging test on the valve seal by an electrostatic discharge control method, determining a degradation rate index based on the changes of characteristic parameters before and after the accelerated aging, and performing curve fitting on the degradation rate index and the operation data of the valve health maintenance scheme to calculate the valve seal degradation period.

[0011] Specifically, the construction method of the position-pressure dual-mode grid is: obtaining displacement and pressure data of the electric actuator in the valve opening and closing process, pairing the position data and pressure data in dual mode, and dividing grid nodes according to the valve opening and closing stroke, recording corresponding pressure characteristic parameters and position characteristic parameters on each node, and performing interpolation and encryption processing on the grid according to the node characteristic parameters to construct the position-pressure dual-mode grid.

[0012] Specifically, the labeling binding method classifies and stores the maintenance information according to valve operations, assigns grid labels to the maintenance information, quickly accesses the health data, and stores the grid labels in the valve health maintenance database through distributed caching technology.

[0013] Specifically, the generation process of the valve health maintenance scheme includes: Inputting the torque, displacement and pressure operation data obtained in the position-pressure dual-mode grid and the operation log into the electric actuator monitoring model, respectively obtaining the torque curves changing with the opening degree in the opening and closing processes of the valve, and generating an initial health index according to the hysteresis area value of the difference area between the two curves; Comparing the initial health index with the maintenance execution index, re-matching the weight of the deviation parameter according to the change of the hysteresis area, and attaching the judgment result in the form of maintenance information to the dual-mode grid node to form a structured maintenance information label; Combining the iterative feedback mechanism, the maintenance information label and the node data of the dual-mode grid are fused and stored to output the valve health maintenance scheme, and a dynamic correction process is triggered when the operation parameter deviates to automatically generate the valve health maintenance scheme.

[0014] Specifically, a valve health monitoring and maintenance system based on an electric actuator includes: The health data acquisition module acquires valve maintenance data of the electric actuator, constructs a position-pressure bimodal grid based on the valve maintenance data, performs pressure testing on the valve, analyzes torque-opening curves in opening and closing processes, and evaluates valve sealing performance according to the torque-opening curves to obtain valve health degree data; The monitoring model construction module constructs an electric actuator monitoring model based on the valve health degree data, inputs the valve health degree data as monitoring parameters into the electric actuator monitoring model, outputs an operation log containing data of valve undisturbed response judgment and torque switching, sets valve maintenance information in combination with maintenance execution indexes, attaches maintenance information labels to grid nodes through a labeling binding method, updates the position-pressure bimodal grid, and generates a valve health maintenance scheme. The valve cycle verification module analyzes valve maintenance data of the electric actuator according to the valve health maintenance scheme, identifies valve torque profiles, switching time delays, hysteresis and leakage data in the operation log through a pressure test passing curve feature extraction method, verifies valve sealing degradation cycles based on an electrostatic discharge control method, and obtains a valve scheme optimization strategy. The maintenance scheme optimization module executes a maintenance scheme optimized by the valve scheme optimization strategy according to current valve torque health data, updates valve torque performance when it is monitored that valve torque curves output by the electric actuator monitoring model meet the maintenance execution indexes, returns the updated valve torque to the electric actuator monitoring model based on a recursive execution feedback method, and automatically outputs an updated valve health maintenance scheme.

[0015] The present application has the following advantages: This invention proposes a valve health monitoring and maintenance method based on electric actuators. By introducing a position-pressure dual-modal mesh, it achieves joint modeling of displacement and pressure data during valve opening and closing, effectively improving the accuracy of valve operating status monitoring and avoiding judgment bias caused by a single monitoring parameter. Through the collection and analysis of operation logs for disturbance-free response and switching delay, this invention can accurately identify valve torque curve characteristics and, combined with curve feature extraction, quantify switching hysteresis and leakage, providing highly reliable data support for valve health assessment. The introduction of error backpropagation and recursive execution feedback mechanisms in the monitoring model construction enables the electric actuator monitoring model to dynamically adjust parameter weights, ensuring the adaptability of the health monitoring process. When the valve operating status deviates, the system can iteratively update the solution through model optimization, thereby avoiding misjudgments and lag problems caused by fixed thresholds in traditional methods. Furthermore, by using electrostatic discharge control to accelerate the aging verification of sealing components, this invention can accurately obtain the valve seal degradation cycle, forming a maintenance strategy oriented towards the entire life cycle and significantly improving predictive maintenance capabilities. By attaching maintenance information tags to bimodal grid nodes, this invention enables structured storage and rapid indexing of valve maintenance information, facilitating the establishment of a valve health database and providing data support for subsequent large-scale distributed maintenance. Overall, this invention not only improves the real-time performance and accuracy of valve operating status monitoring but also reduces maintenance costs, extends valve lifespan, and enhances the safety and reliability of industrial systems. Attached Figure Description

[0016] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings.

[0017] Figure 1 This is a schematic diagram of the structure of a valve health monitoring and maintenance method and system based on an electric actuator according to the present invention.

[0018] Figure 2 This is a flowchart illustrating the valve health monitoring and maintenance method based on an electric actuator and the electric actuator monitoring model described in the system of the present invention. Detailed Implementation

[0019] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided.

[0020] Please see Figure 1 A valve health monitoring and maintenance method based on electric actuators: S1: Obtain valve maintenance data of the electric actuator, construct a position-pressure dual-mode grid based on the valve maintenance data, perform pressure testing on the valve, analyze torque-opening curves in the opening and closing processes, and evaluate valve sealing performance according to the torque-opening curves to obtain valve health data; S2: Construct an electric actuator monitoring model based on the valve health data, which includes valve undisturbed response judgment and torque switching data, input the valve health data as monitoring parameters into the electric actuator monitoring model, output operation logs containing valve undisturbed response judgment and torque switching data, set valve maintenance information in combination with maintenance execution indicators, attach maintenance information tags to grid nodes through label binding method, update the position-pressure dual-mode grid, and generate a valve health maintenance scheme; S3: According to the valve health maintenance scheme, analyze the valve maintenance data of the electric actuator, identify valve torque profile, switching delay, hysteresis and leakage data in the operation log through curve feature extraction method, and verify valve sealing degradation period based on electrostatic discharge control method to obtain valve scheme optimization strategy; S4: According to the current valve torque health data, execute the maintenance scheme optimized by the valve scheme optimization strategy, update valve torque performance when the electric actuator monitoring model outputs valve torque curve that meets the maintenance execution indicators, and return the updated valve torque to the electric actuator monitoring model based on recursive execution feedback method to automatically output the updated valve health maintenance scheme.

[0021] Specifically, the logic process of generating the valve health data is as follows: Based on the position-pressure dual-mode grid, the valve is subjected to segmented pressure loading, and the corresponding pressure response curve is collected at different opening positions. The valve opening and closing instruction issuing time and the actual pressure and torque curve response starting time are recorded to obtain the undisturbed response interval in the valve switching operation process. According to the undisturbed interval, the switching delay parameter is stored in a structured manner, and in combination with the pressure response curve in the pressure loading process, a valve health data set is formed. The valve health data set is used as the basic input data source of the electric actuator monitoring model, the pressure response curve is reconstructed by interpolation fitting, and the valve health data is obtained.

[0022] Specifically, the valve maintenance data acquisition method is to analyze the running parameters in the valve opening and closing process, including torque data, pressure data, and operation log data. The running parameters are time-sequenced and modal fused, and in combination with the opening and closing instructions and feedback signals in the operation log data, complete valve maintenance data is generated.

[0023] In this embodiment, the electric actuator monitoring model analyzes the operation log of the undisturbed response determination and switching test to define a set of operating data of the valve: where u i is the operating data sample corresponding to the i-th valve opening and closing operation.

[0024] Let the torque curve of the i-th operation be: where m ij is the torque value at time point t j .

[0025] In combination with the maintenance execution index of the valve and the bimodal grid constraint, when the operating curve of the valve deviates from the health degree target, the recursive feedback module starts the optimization algorithm to correct the health degree parameter, and the specific calculation formula is: where, is the corrected health degree value, M ref is the reference torque parameter, and λ is the dynamic correction factor.

[0026] In the valve seal degradation period verification process, let the leakage before the accelerated aging test be L pre , the leakage after the accelerated aging test be L post , and the degradation rate calculation formula be: where D is the degradation rate, Δt is the accelerated aging time interval, and when D exceeds the health degree threshold δ, the system triggers the maintenance warning mechanism and enters the optimization maintenance scheme generation process.

[0027] Specifically, the operation log generation method containing valve undisturbed response determination and torque switching data is: at the time point of issuing the valve opening and closing instruction, record the valve displacement and conduction pressure response signals, and according to the valve response determination whether there is a delay, obtain the undisturbed response interval, in the valve action process, according to the undisturbed response interval, analyze the switching delay parameter, at the same time, sample the displacement-pressure time series data in the action process, identify the hysteresis characteristic and transient fluctuation, write the leakage change curve, action priority and abnormality determination mark in this process into the log, form the operation log data which can be traced back.

[0028] Specifically, the construction method of the electric actuator monitoring model is: ​​​​The torque, displacement and pressure raw data features in the opening and closing process of the valve are extracted, the unstructured valve maintenance data is converted into a set of extracted switching delay, hysteresis and leakage parameters, and the switching delay is associated with the valve health degree in the operation log record to generate a valve health maintenance feature matrix; Based on the valve health maintenance feature matrix, the set of delay, hysteresis and leakage parameters is received, the valve maintenance monitoring features are processed, the resource allocation result is converged through the curve feature extraction method, and the motor execution maintenance scheme is generated according to the undisturbed response and the switching delay operation log; According to the valve operation execution data and feedback results collected in real time according to the maintenance allocation scheme, the parameter weight of the electric actuator monitoring model for monitoring the valve health degree is dynamically corrected through error back propagation under the constraint of the position-pressure dual-mode grid, and the valve health degree constraint condition is taken as an external mapping parameter to construct the electric actuator monitoring model.

[0029] In this embodiment, the electric actuator monitoring model is deployed in the micro-service architecture of the edge computing gateway and the cloud platform, as shown in Figure 2 The system first continuously collects the torque, displacement and pressure operation logs in the opening and closing process of the valve through the industrial bus protocol (such as Modbus, OPC UA), and performs real-time processing through the streaming computing framework (such as Flink), including raw data cleaning, time series alignment and noise removal; then, the model calculates the torque curve features of the valve at different opening degrees based on the curve feature extraction method, forming health parameters such as switching delay, hysteresis and leakage; on this basis, the model uses the rule engine (such as Drools) to weight match the valve maintenance execution index, health degree constraint condition and operation priority, dynamically generates the valve health maintenance scheme, and writes the maintenance information into the position-pressure dual-mode grid node through the label binding method, realizes the coupled storage with the structured data; when it is detected that the actual operation curve of the valve deviates significantly from the maintenance scheme, the model automatically triggers the scheme correction logic, re-inputs the updated valve torque into the model based on the recursive execution feedback method, and dynamically adjusts the parameter weight, thereby forming an iterative optimization monitoring process; finally, all monitoring results and maintenance schemes are stored in a relational database (such as PostgreSQL), and part of the high-frequency node data and health degree labels are cached to Redis to ensure real-time analysis and fast retrieval. Through this embodiment, the electric actuator monitoring model can realize high-precision monitoring, self-adaptive optimization and dynamic output of the whole life cycle maintenance strategy of the valve running state without complex hardware modification, and improve the reliability and intelligent level of system operation.

[0030] Specifically, the process of returning the updated valve torque to the electric actuator monitoring model is as follows: Obtain the difference data between the valve health maintenance scheme generated by the model and the torque curve in the actual execution process; and based on the valve health maintenance scheme and the corrected valve switching response time, the valve torque switching response time corrected by the executed valve scheme optimization strategy is data encoded, and when it is detected that the deviation exceeds the preset threshold of the valve health maintenance scheme, the model optimization process is triggered; Based on the recursive execution feedback method in the model optimization process, the updated valve torque curve is input into the electric actuator monitoring model as a correction parameter, and the optimization parameter is fused with the monitoring and maintenance distribution rule of the electric actuator monitoring model to form an iterative input sequence. Based on the running result of the updated electric actuator monitoring model, a new valve health maintenance scheme is automatically generated, and the optimized scheme is written into the health association database of the electric actuator monitoring model through cyclic feedback.

[0031] The embodiment provides a valve health monitoring and maintenance system based on an electric actuator, which mainly comprises a position-pressure dual-mode grid construction module, an electric actuator monitoring model module, a valve feature recognition and degradation cycle verification module, and a recursive feedback and scheme iterative optimization module. Each module can interact through an industrial bus protocol (such as Modbus, OPC UA) or a microservice interface, and run on an edge computing gateway and a cloud distributed server cluster on the enterprise side.

[0032] The specific technical stack is: a distributed microservice architecture is constructed by using Spring Cloud + Kubernetes; a database layer uses PostgreSQL to store valve maintenance data and health information; Redis is used for high-speed caching of dual-mode grid node data and maintenance information tags; a valve torque curve and a feature extraction model are built by using Python + TensorFlow; and a scheme optimization strategy and an execution rule are realized by a Java rule engine (Drools).

[0033] The specific implementation process is as follows: For example Figure 1As shown, the position-pressure bimodal grid construction module collects position and pressure data of the electric actuator during the valve opening and closing process, and constructs a bimodal grid, performs pressure loading and unloading tests, and generates health data; then, the electric actuator monitoring model module takes the undisturbed response and switching time delay operation log as input, outputs the valve torque curve, and generates a preliminary health maintenance scheme combined with the maintenance execution index; the valve feature recognition and degradation cycle verification module identifies the switching time delay, hysteresis and leakage by the curve feature extraction method, and verifies the degradation cycle by the electrostatic discharge control method to implement accelerated aging test on the valve seal, and outputs the scheme optimization strategy; the recursive feedback and scheme iterative optimization module updates the valve torque curve according to the real-time feedback after executing the optimized scheme, and returns the updated results to the electric actuator monitoring model through the recursive feedback method, forming an automatic iterative optimization of the health maintenance scheme, so as to realize long-term stable monitoring and maintenance of the valve operating state.

[0034] Specifically, the pressure test simulates the valve micro-motion execution process by synchronously sampling the undisturbed valve switching response, verifies the torque curve scheme, and generates a valve scheme optimization strategy according to the test results.

[0035] Specifically, the verification method of the valve seal degradation cycle is: obtaining the leakage amount, pressure holding time operation data of the sealing part during the valve opening and closing process, and implementing accelerated aging test on the valve seal by the electrostatic discharge control method, determining the degradation rate index based on the characteristic parameter changes before and after the accelerated aging, and calculating the valve seal degradation cycle by curve fitting the degradation rate index and the operation data of the valve health maintenance scheme.

[0036] Specifically, the construction method of the position-pressure bimodal grid is: obtaining the displacement and pressure data of the electric actuator during the valve opening and closing process, pairing the position data and pressure data in a bimodal manner, and dividing the grid nodes according to the valve opening and closing stroke, recording the corresponding pressure characteristic parameters and position characteristic parameters on each node, and constructing the position-pressure bimodal grid by interpolating and encrypting the grid according to the node characteristic parameters.

[0037] Specifically, the labeling binding method classifies and stores the maintenance information according to the valve operation, allocates grid labels to the maintenance information, quickly accesses the health data, and stores the grid labels in the valve health maintenance database through the distributed cache technology.

[0038] Specifically, the specific process of generating the valve health maintenance scheme includes: The torque, displacement and pressure operation data obtained in the position-pressure dual-mode grid are input into the electric actuator monitoring model, the torque curves changing with the opening degree during the opening and closing processes of the valve are obtained respectively, and the initial health index is generated according to the hysteresis area value of the difference region between the two curves; The health initial index is compared with the maintenance execution index, the weight of the deviation parameter is re-matched according to the hysteresis area change, and the judgment result is attached to the dual-mode grid node in the form of maintenance information to form a structured maintenance information tag; The maintenance information tag is stored by combining with the iterative feedback mechanism, the valve health maintenance scheme is output, and the dynamic correction process is triggered when the operation parameter deviates, and the valve health maintenance scheme is automatically generated.

[0039] Specifically, a valve health monitoring and maintenance system based on an electric actuator, characterized in that it comprises: A health data acquisition module: acquires valve maintenance data of an electric actuator, constructs a position-pressure dual-mode grid based on the valve maintenance data, performs pressure testing on the valve, analyzes the torque-opening degree curves during the opening and closing processes, and evaluates the valve sealing performance according to the torque-opening degree curves to obtain valve health data; A monitoring model construction module: constructs an electric actuator monitoring model based on the valve health data, the valve health data includes valve undisturbed response judgment and torque switching data, inputs the valve health data as monitoring parameters into the electric actuator monitoring model, outputs operation logs containing valve undisturbed response judgment and torque switching data, sets valve maintenance information combined with maintenance execution index, attaches maintenance information tags to grid nodes through label binding method, updates the position-pressure dual-mode grid, and generates valve health maintenance scheme; A valve cycle verification module: analyzes the valve maintenance data of the electric actuator according to the valve health maintenance scheme, identifies valve torque profile, switching delay, hysteresis and leakage data in the operation log through curve feature extraction method, verifies valve sealing degradation cycle based on electrostatic discharge control method, and obtains valve scheme optimization strategy; A maintenance scheme optimization module: executes the maintenance scheme optimized by the valve scheme optimization strategy according to the current valve torque health data, updates the valve torque performance when the valve torque curve output by the electric actuator monitoring model meets the maintenance execution index, and returns the updated valve torque to the electric actuator monitoring model based on recursive execution feedback method, and automatically outputs the updated valve health maintenance scheme.

[0040] The above merely describes the preferred embodiments of the present application, and is not intended to limit the present application in any form. Although the present application has been disclosed with the preferred embodiments as above, it is not intended to limit the present application. Any person skilled in the art can make some changes or modifications to the above disclosed technical content to obtain equivalent embodiments with equivalent changes, as long as the changes or modifications do not deviate from the technical solution of the present application. Any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present application still belongs to the scope of the technical solution of the present application.

Claims

1. A method of health monitoring maintenance of an electric actuator based valve, characterized in that, The method comprises the following steps: S1: obtaining valve maintenance data of the electric actuator, constructing a position-pressure bimodal grid based on the valve maintenance data, performing a pressure test on the valve, analyzing torque-opening curves in the opening and closing processes, and evaluating the valve sealing performance according to the torque-opening curves to obtain valve health data; S2: constructing an electric actuator monitoring model based on the valve health data, the valve health data including valve undisturbed response judgment and torque switching data, inputting the valve health data as monitoring parameters into the electric actuator monitoring model, outputting an operation log containing valve undisturbed response judgment and torque switching data, setting valve maintenance information in combination with maintenance execution indexes, attaching maintenance information tags to grid nodes through label binding, updating the position-pressure bimodal grid, and generating a valve health maintenance scheme; S3: analyzing the valve maintenance data of the electric actuator according to the valve health maintenance scheme, identifying valve torque profiles, switching time delays, hysteresis and leakage data in the operation log through curve feature extraction, and verifying valve sealing degradation periods based on electrostatic discharge control to obtain a valve scheme optimization strategy; S4: executing the maintenance scheme optimized by the valve scheme optimization strategy according to the current valve torque health data, updating valve torque performance when the electric actuator monitoring model outputs a valve torque curve that meets the maintenance execution indexes, returning the updated valve torque to the electric actuator monitoring model based on recursive execution feedback, and automatically outputting an updated valve health maintenance scheme.

2. The method of claim 1, wherein, The logical process for generating the valve health data is as follows: Based on the position-pressure bimodal grid, the valve is subjected to segmented pressure loading, and corresponding pressure response curves are collected at different opening positions. The valve health data set is obtained by recording the valve opening and closing instruction issuing time and the actual pressure and torque curve response starting time in the valve switching operation process. According to the undisturbed interval, the switching time delay parameters are structured and stored, and in combination with the pressure response curve in the pressure loading process, the valve health data set is formed. The valve health data set is used as the basic input data source of the electric actuator monitoring model, the pressure response curve is reconstructed by interpolation fitting, and the valve health data is obtained.

3. The method of claim 1, wherein, The valve maintenance data acquisition method is to analyze the running parameters in the valve opening and closing process, including torque data, pressure data, and operation log data. The running parameters are time-sequenced and modal fused, and in combination with the opening and closing instructions and feedback signals in the operation log data, complete valve maintenance data is generated.

4. The method of claim 1, wherein, The operation log generation method comprising valve disturbance-free response judgment and torque switching data is as follows: at the time point of valve opening and closing instruction issuance, the valve displacement and conduction pressure response signals are recorded, and according to the valve response, it is judged whether there is a delay, and the disturbance-free response interval is obtained; in the valve action process, the switching time delay parameter is analyzed according to the disturbance-free response interval, and the displacement-pressure time sequence data in the action process is sampled to identify the hysteresis characteristic and transient fluctuation; the leakage change curve in the process, the action priority and the abnormality judgment mark are written into the log together to form a traceable operation log data.

5. The method of claim 2, wherein, The construction method of the electric execution monitoring model is as follows: The torque, displacement and pressure original data characteristics in the valve opening and closing process are extracted, the unstructured valve maintenance data is converted into the switching time delay, hysteresis and leakage parameter set, and the switching time delay associated valve health degree in the operation log record is combined to generate the valve health maintenance characteristic matrix; Based on the valve health maintenance characteristic matrix, the switching time delay, hysteresis and leakage parameter set are received, the valve maintenance monitoring characteristics are processed, the resource allocation result is converged through the curve characteristic extraction method, the motor execution maintenance scheme is generated according to the disturbance-free response and switching time delay operation log, and the valve health degree is monitored. According to the valve operation execution data and feedback result collected in real time according to the maintenance allocation scheme, the parameter weight of the electric execution monitoring model for valve health degree monitoring is dynamically corrected through error back propagation under the constraint of the position-pressure dual-mode grid, and the valve health degree constraint condition is taken as an external mapping parameter to construct the electric execution monitoring model.

6. The method of claim 5, wherein, The process of updating the valve torque back to the electric execution monitoring model is as follows: The difference data between the valve health maintenance scheme generated by the model and the torque curve in the actual execution process is obtained; and based on the valve health maintenance scheme and the corrected valve switching response time, the valve torque switching response time corrected by the executed valve scheme optimization strategy is data encoded; when it is detected that the deviation exceeds the preset threshold of the valve health maintenance scheme, the model optimization process is triggered; Based on the recursive execution feedback method in the model optimization process, the updated valve torque curve is input into the electric execution monitoring model as a correction parameter, and the optimization parameter is fused with the monitoring and maintenance allocation rule of the electric execution monitoring model to form an iterative input sequence; Based on the running result of the updated electric execution monitoring model, a new valve health maintenance scheme is automatically generated, and the optimized scheme is written into the health association database of the electric execution monitoring model through cyclic feedback.

7. The method of claim 4, wherein, The pressure test verifies the torque curve scheme by synchronously sampling the disturbance-free valve switching response simulation monitoring valve micro-motion execution process, and generates a valve scheme optimization strategy according to the test result.

8. The method of claim 2, wherein, The verification method of the valve seal degradation period is: obtaining the leakage amount and pressure holding time operation data of the sealing part in the valve opening and closing process, and implementing accelerated aging test on the valve seal by electrostatic discharge control method, determining the degradation rate index based on the change of characteristic parameters before and after acceleration aging, and performing curve fitting on the degradation rate index and the operation data of the valve health maintenance scheme to calculate the valve seal degradation period.

9. The method of claim 4, wherein, The construction method of the position-pressure dual-mode grid is: obtaining the displacement and pressure data of the electric actuator in the valve opening and closing process, pairing the position data with the pressure data in dual-mode, dividing the grid nodes according to the valve opening and closing stroke, recording the corresponding pressure characteristic parameters and position characteristic parameters on each node, and performing interpolation and encryption processing on the grid according to the node characteristic parameters to construct the position-pressure dual-mode grid.

10. The method of claim 4, wherein, The labeling binding method classifies and stores the maintenance information according to valve operation, assigns grid labels to the maintenance information, quickly accesses the health data, and stores the grid labels in the valve health maintenance database through distributed cache technology.

11. The method of claim 7, wherein, The specific process of generating the valve health maintenance scheme includes: Inputting the torque, displacement and pressure operation data obtained in the position-pressure dual-mode grid and the operation log into the electric actuator monitoring model to obtain the torque change curve with opening degree in the opening and closing process of the valve respectively, and generating the initial health index according to the hysteresis area value of the difference area between the two curves; Comparing the initial health index with the maintenance execution index, re-matching the weight of the deviation parameter according to the change of the hysteresis area, and attaching the judgment result in the form of maintenance information to the dual-mode grid node to form a structured maintenance information label; Combining the iterative feedback mechanism, the maintenance information label and the node data of the dual-mode grid are fused and stored to output the valve health maintenance scheme, and the dynamic correction process is triggered when the operation parameter deviates to automatically generate the valve health maintenance scheme.

12. An electric actuator based valve health monitoring maintenance system, characterized by, It includes: The health data acquisition module: obtains the valve maintenance data of the electric actuator, constructs a position-pressure dual-mode grid based on the valve maintenance data, performs pressure test on the valve, analyzes the torque-opening degree curve in the opening and closing process, and evaluates the valve sealing performance according to the torque-opening degree curve to obtain the valve health data; The monitoring model construction module: constructs an electric actuator monitoring model based on the valve health data, which includes valve undisturbed response judgment and torque switching data, inputs the valve health data as monitoring parameters into the electric actuator monitoring model, outputs the operation log containing valve undisturbed response judgment and torque switching data, sets valve maintenance information combined with maintenance execution index, attaches maintenance information label to the grid node through the labeling binding method, updates the position-pressure dual-mode grid, and generates the valve health maintenance scheme; The valve cycle verification module: according to the valve health maintenance scheme, the valve maintenance data of the electric actuator is analyzed, the valve torque profile and the switching delay, hysteresis and leakage data in the operation log are identified by the pressure test passing curve feature extraction method, and the valve sealing degradation cycle is verified based on the electrostatic discharge control method, and the valve scheme optimization strategy is obtained; The maintenance scheme optimization module: according to the current valve torque health data, the maintenance scheme optimized by the valve scheme optimization strategy is executed, the valve torque performance is updated when it is monitored that the valve torque curve output by the electric actuator monitoring model meets the maintenance execution index, and the updated valve torque is returned to the electric actuator monitoring model based on the recursive execution feedback method, and the updated valve health maintenance scheme is automatically output.

Images