CN118066182B - Intelligent monitoring method and system for working state of rotary direct-drive servo valve - Google Patents
Intelligent monitoring method and system for working state of rotary direct-drive servo valve Download PDFInfo
- Publication number
- CN118066182B CN118066182B CN202410470060.3A CN202410470060A CN118066182B CN 118066182 B CN118066182 B CN 118066182B CN 202410470060 A CN202410470060 A CN 202410470060A CN 118066182 B CN118066182 B CN 118066182B
- Authority
- CN
- China
- Prior art keywords
- servo valve
- model
- index
- teacher
- class
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F15—FLUID-PRESSURE ACTUATORS; HYDRAULICS OR PNEUMATICS IN GENERAL
- F15B—SYSTEMS ACTING BY MEANS OF FLUIDS IN GENERAL; FLUID-PRESSURE ACTUATORS, e.g. SERVOMOTORS; DETAILS OF FLUID-PRESSURE SYSTEMS, NOT OTHERWISE PROVIDED FOR
- F15B19/00—Testing; Calibrating; Fault detection or monitoring; Simulation or modelling of fluid-pressure systems or apparatus not otherwise provided for
- F15B19/007—Simulation or modelling
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F15—FLUID-PRESSURE ACTUATORS; HYDRAULICS OR PNEUMATICS IN GENERAL
- F15B—SYSTEMS ACTING BY MEANS OF FLUIDS IN GENERAL; FLUID-PRESSURE ACTUATORS, e.g. SERVOMOTORS; DETAILS OF FLUID-PRESSURE SYSTEMS, NOT OTHERWISE PROVIDED FOR
- F15B19/00—Testing; Calibrating; Fault detection or monitoring; Simulation or modelling of fluid-pressure systems or apparatus not otherwise provided for
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F15—FLUID-PRESSURE ACTUATORS; HYDRAULICS OR PNEUMATICS IN GENERAL
- F15B—SYSTEMS ACTING BY MEANS OF FLUIDS IN GENERAL; FLUID-PRESSURE ACTUATORS, e.g. SERVOMOTORS; DETAILS OF FLUID-PRESSURE SYSTEMS, NOT OTHERWISE PROVIDED FOR
- F15B20/00—Safety arrangements for fluid actuator systems; Applications of safety devices in fluid actuator systems; Emergency measures for fluid actuator systems
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Fluid Mechanics (AREA)
- Mechanical Engineering (AREA)
- General Engineering & Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The application provides an intelligent monitoring method and system for the working state of a rotary direct-drive servo valve, and relates to the technical field of intelligent monitoring of servo valves, wherein the method comprises the following steps: rendering the class A index parameters to a servo valve model to obtain a teacher model; screening the characteristic indexes to obtain a class B index; rendering the B-type index parameters to a servo valve model to obtain a student model; analyzing the teacher model and the student model to obtain model similarity; dynamically monitoring through a student model to obtain a prediction index; correcting the prediction index based on the model similarity to obtain a target prediction index; and when the target prediction index reaches a preset threshold value, activating the teacher model to perform monitoring and early warning. The application can solve the technical problems that the data processing efficiency is low due to the large monitoring data volume and redundant data, so that the abnormal working state of the servo valve cannot be found in time, can improve the monitoring data processing efficiency, and achieves the technical effect of finding the abnormal working state of the servo valve in time.
Description
Technical Field
The application relates to the technical field of intelligent monitoring of servo valves, in particular to an intelligent monitoring method and system for working states of a rotary direct-drive servo valve.
Background
The rotary direct-drive servo valve is a high-performance hydraulic control element, and has wide application prospects in various fields by virtue of the characteristics of high precision, high reliability and high performance, such as: industrial automation, aerospace, shipbuilding, and the like.
At present, when the working state of the rotary direct-drive servo valve is monitored by the existing method, a large amount of monitoring data is usually required to be collected from a plurality of monitoring dimensions for state analysis, but because the data processing calculation force is limited, and meanwhile, the data volume obtained through the monitoring of the plurality of dimensions is large and redundant data exists, the data processing efficiency is low, so that the abnormal working state of the servo valve cannot be found in time and the abnormal processing can be rapidly carried out, and the operation stability and the safety of a hydraulic system are adversely affected.
In summary, the existing method for monitoring the working state of the rotary direct-drive servo valve has low processing efficiency of the monitored data due to large monitoring data volume and redundant data, so that the abnormal working state of the servo valve cannot be found in time, and the technical problems of stability and safety of the operation of the hydraulic system are affected.
Disclosure of Invention
The application aims to provide an intelligent monitoring method and system for the working state of a rotary direct-drive servo valve, which are used for solving the technical problems that the monitoring data processing efficiency is low due to the fact that the monitoring data size is large and redundant data exists in the existing monitoring method for the working state of the rotary direct-drive servo valve, so that the abnormal working state of the servo valve cannot be found in time, and the operation stability and safety of a hydraulic system are affected.
In view of the above problems, the application provides an intelligent monitoring method and system for the working state of a rotary direct-drive servo valve.
In a first aspect, the present application provides an intelligent monitoring method for a working state of a rotary direct-drive servo valve, where the method is implemented by an intelligent monitoring system for a working state of a rotary direct-drive servo valve, and the method includes: constructing an initial servo valve model according to physical characteristic information of the rotary direct-drive servo valve acquired based on the preset physical characteristics; reading a preset multidimensional feature, and marking a plurality of feature fingers in the preset multidimensional feature as a class A index; rendering the class A index parameters dynamically monitored based on the class A index to the initial model of the servo valve to obtain a twin model of a teacher of the servo valve; screening the characteristic indexes by combining the historical servo valve working records to obtain a class B index; rendering the B-type index parameters dynamically monitored based on the B-type index to the servo valve initial model to obtain a servo valve student twin model; comparing and analyzing the servo valve teacher twin model and the servo valve student twin model to obtain model similarity; dynamically monitoring and analyzing the rotary direct-drive servo valve through the servo valve student twin model to obtain a predicted state index; performing deviation correction analysis on the predicted state index based on the model similarity to obtain a target predicted state index; and when the target prediction state index reaches a preset index threshold, activating the servo valve teacher twin model to dynamically monitor and early warn the rotary direct-drive servo valve.
In a second aspect, the present application further provides an intelligent monitoring system for a working state of a rotary direct-drive servo valve, for performing an intelligent monitoring method for a working state of a rotary direct-drive servo valve according to the first aspect, where the system includes: the servo valve initial model building module is used for building a servo valve initial model according to the physical characteristic information of the rotary direct-drive servo valve acquired based on the preset physical characteristic; the A-type index marking module is used for reading preset multidimensional features and marking a plurality of feature fingers in the preset multidimensional features as A-type indexes; the servo valve teacher twin model obtaining module is used for rendering the class A index parameters dynamically monitored based on the class A index to the servo valve initial model to obtain a servo valve teacher twin model; the B-type index obtaining module is used for screening the characteristic indexes by combining the historical servo valve working records to obtain B-type indexes; the servo valve student twin model obtaining module is used for rendering B-type index parameters dynamically monitored based on the B-type index to the servo valve initial model to obtain a servo valve student twin model; the model similarity obtaining module is used for comparing and analyzing the servo valve teacher twin model and the servo valve student twin model to obtain model similarity; the prediction state index obtaining module is used for carrying out dynamic monitoring analysis on the rotary direct-drive servo valve through the servo valve student twin model to obtain a prediction state index; the prediction state index deviation rectifying module is used for carrying out deviation rectifying analysis on the prediction state index based on the model similarity to obtain a target prediction state index; and the dynamic monitoring and early warning module is used for activating the servo valve teacher twin model to dynamically monitor and early warn the rotary direct-drive servo valve when the target prediction state index reaches a preset index threshold.
One or more technical schemes provided by the application have at least the following technical effects or advantages:
Constructing an initial servo valve model according to physical characteristic information of the rotary direct-drive servo valve acquired based on preset physical characteristics; reading a preset multidimensional feature, and marking a plurality of feature fingers in the preset multidimensional feature as a class A index; rendering the class A index parameters dynamically monitored based on the class A index to the initial model of the servo valve to obtain a twin model of a teacher of the servo valve; screening the characteristic indexes by combining the historical servo valve working records to obtain a class B index; rendering the B-type index parameters dynamically monitored based on the B-type index to the servo valve initial model to obtain a servo valve student twin model; comparing and analyzing the servo valve teacher twin model and the servo valve student twin model to obtain model similarity; dynamically monitoring and analyzing the rotary direct-drive servo valve through the servo valve student twin model to obtain a predicted state index; performing deviation correction analysis on the predicted state index based on the model similarity to obtain a target predicted state index; when the target prediction state index reaches a preset index threshold, activating the servo valve teacher twin model to dynamically monitor and early warn the rotary direct-drive servo valve, namely, firstly, constructing a complex servo valve teacher twin model based on class A indexes; then screening the class A index to obtain a simplified key characteristic index which is set as a class B index, and further constructing a simplified version of servo valve student twin model based on the class B index; and the working state of the servo valve is monitored through the servo valve student twin model, and after the abnormal state is primarily monitored, the servo valve teacher twin model is activated to dynamically monitor and early warn the servo valve, so that the processing efficiency of the servo valve state monitoring data can be improved, the abnormal working state of the servo valve can be timely found, and further effective measures can be taken to conduct targeted processing, so that the technical effect of ensuring the safe and stable operation of the hydraulic system is achieved.
The foregoing description is only an overview of the present application, and is intended to be implemented in accordance with the teachings of the present application in order that the same may be more clearly understood and to make the same and other objects, features and advantages of the present application more readily apparent. It should be understood that the description in this section is not intended to identify key or critical features of the embodiments of the application or to delineate the scope of the application. Other features of the present application will become apparent from the description that follows.
Drawings
In order to more clearly illustrate the application or the technical solutions of the prior art, the following brief description will be given of the drawings used in the description of the embodiments or the prior art, it being obvious that the drawings in the description below are only exemplary and that other drawings can be obtained from the drawings provided without the inventive effort for a person skilled in the art.
FIG. 1 is a schematic flow chart of an intelligent monitoring method for the working state of a rotary direct-drive servo valve;
FIG. 2 is a schematic flow chart of a teacher-to-teacher model of a rotary direct-drive servo valve obtained in an intelligent monitoring method of the working state of the servo valve;
FIG. 3 is a schematic diagram of a system for intelligently monitoring the working state of a rotary direct-drive servo valve.
Reference numerals illustrate:
The system comprises a servo valve initial model construction module 11, a class A index marking module 12, a servo valve teacher twin model obtaining module 13, a class B index obtaining module 14, a servo valve student twin model obtaining module 15, a model similarity obtaining module 16, a prediction state index obtaining module 17, a prediction state index deviation correcting module 18 and a dynamic monitoring and early warning module 19.
Detailed Description
The application provides an intelligent monitoring method and system for the working state of a rotary direct-drive servo valve, which solve the technical problems that the monitoring data processing efficiency is low due to the large monitoring data quantity and redundant data in the existing monitoring method for the working state of the rotary direct-drive servo valve, so that the abnormal working state of the servo valve cannot be found in time, and the operation stability and safety of a hydraulic system are affected. The processing efficiency of the servo valve state monitoring data can be improved, so that the abnormal working state of the servo valve can be found in time, and further effective measures can be taken for targeted processing, and the technical effect of ensuring the safe and stable operation of the hydraulic system is achieved.
In the following, the technical solutions of the present application will be clearly and completely described with reference to the accompanying drawings, and it should be understood that the described embodiments are only some embodiments of the present application, but not all embodiments of the present application, and that the present application is not limited by the exemplary embodiments described herein. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application. It should be further noted that, for convenience of description, only some, but not all of the drawings related to the present application are shown.
Example 1
Referring to fig. 1, the application provides an intelligent monitoring method for a working state of a rotary direct-drive servo valve, wherein the method is applied to an intelligent monitoring system for the working state of the rotary direct-drive servo valve, and the method specifically comprises the following steps:
Step one: constructing an initial servo valve model according to physical characteristic information of the rotary direct-drive servo valve acquired based on the preset physical characteristics;
Specifically, the method provided by the application is used for optimizing the existing working state monitoring method of the rotary direct-drive servo valve to achieve the purpose of improving the processing efficiency of the servo valve state monitoring data, and the method is specifically implemented in an intelligent monitoring system of the working state of the rotary direct-drive servo valve.
Firstly, acquiring preset physical characteristics, wherein the preset physical characteristics are data types required by simulation modeling of a rotary direct-drive servo valve, and the data comprise specification information, structural characteristics, working principles, dynamic performance parameters, control parameters and the like; and then, based on the physical characteristics, information acquisition is carried out on the rotary direct-drive servo valve, and the physical characteristic information of the rotary direct-drive servo valve is obtained.
The digital twin technology is a novel technology combining a physical world with a virtual world based on a digital technology, and a virtual entity corresponding to the physical world is created by digitally modeling the physical entity, wherein the digital twin technology mainly comprises digital modeling, data acquisition and simulation analysis, and the simulation analysis is to utilize the digital twin model to carry out virtual experiments so as to predict and optimize the operation effect of the physical entity. Based on a digital twin technology, simulation modeling is carried out on the rotary direct-drive servo valve in a visual simulation platform according to the physical characteristic information, and a servo valve initial model is generated.
The digital twin technology-based servo valve initial model is constructed, so that support is provided for the simulation operation of the rotary direct-drive servo valve, and meanwhile, the authenticity and reliability of the simulation operation of the rotary direct-drive servo valve can be improved.
Step two: reading a preset multidimensional feature, and marking a plurality of feature fingers in the preset multidimensional feature as a class A index;
Specifically, a predetermined multidimensional feature is read, and the predetermined multidimensional feature is a plurality of dimension data types for monitoring the rotary direct-drive servo valve, and a person skilled in the art can set according to actual situations, for example: an operating parameter dimension, a vibration monitoring dimension, a noise monitoring dimension, and the like, wherein each dimension feature comprises a plurality of feature indexes; and then marking a plurality of characteristic indexes in the preset multidimensional characteristic as class A indexes, wherein the characteristic indexes comprise zero drift, response speed, flow parameters, pressure parameters, valve core opening degree, vibration amplitude, vibration frequency and the like, the class A indexes are monitoring indexes with large and comprehensive data quantity, the monitoring analysis of the working state of the servo valve is carried out based on the class A indexes, the obtained prediction result is more accurate and reliable, but meanwhile, the monitoring data quantity corresponding to the class A indexes is very large, redundant data exists, so that the data transmission speed and the calculation efficiency are lower, the timely discovery and the treatment of the abnormal working state of the servo valve are not facilitated, and the timeliness is poor.
Step three: rendering the class A index parameters dynamically monitored based on the class A index to the initial model of the servo valve to obtain a twin model of a teacher of the servo valve;
Specifically, based on the class A index, dynamically monitoring the rotary direct-drive servo valve through a plurality of sensors to obtain class A index parameters, wherein the class A index parameters comprise index data obtained through multiple monitoring; and rendering the A-type index parameters to the initial model of the servo valve to obtain a twin model of a teacher of the servo valve, wherein the twin model of the teacher of the servo valve is a model with higher monitoring accuracy and reliability but poorer data analysis efficiency. By obtaining the teacher twin model of the servo valve, support is provided for accurate and reliable monitoring of the servo valve under abnormal working conditions.
Step four: screening the characteristic indexes by combining the historical servo valve working records to obtain a class B index;
Specifically, a working log of the rotary direct-drive servo valve is read, and a historical servo valve working record is obtained based on the working log, wherein the historical servo valve working record comprises a plurality of servo valve historical working data; and then screening the plurality of characteristic indexes by combining the historical servo valve working record, wherein the screening refers to simplifying the plurality of characteristic indexes, extracting the characteristic indexes with high importance degree from the plurality of characteristic indexes, and setting the extracted characteristic indexes as B-type indexes. The B-type index is slightly worse than the A-type index in monitoring accuracy, but the data processing amount is smaller, so that the monitoring data processing efficiency can be improved, and the abnormal working state of the servo valve can be timely found and early warning can be carried out.
Step five: rendering the B-type index parameters dynamically monitored based on the B-type index to the servo valve initial model to obtain a servo valve student twin model;
Specifically, the rotary direct-drive servo valve is dynamically monitored based on the class B index to obtain class B index parameters, the class B index parameters are rendered to the initial servo valve model to obtain a servo valve student twin model, wherein the servo valve student twin model is a model with relatively poor monitoring accuracy and reliability, but extremely high data analysis efficiency, the abnormal working state of the servo valve can be rapidly found, and support is provided for abnormal monitoring of the servo valve in a conventional state.
Step six: comparing and analyzing the servo valve teacher twin model and the servo valve student twin model to obtain model similarity;
Specifically, the similarity of the servo valve teacher twin model and the servo valve student twin model is compared, and model similarity is obtained according to the comparison result. For example: the model similarity comparison can be performed by setting a model comparison index and an existing cosine similarity algorithm. And by obtaining the model similarity, support is provided for the next step of correcting the prediction state indexes obtained based on the servo valve student twin model.
Step seven: dynamically monitoring and analyzing the rotary direct-drive servo valve through the servo valve student twin model to obtain a predicted state index;
Specifically, the rotary direct-drive servo valve is dynamically monitored through the servo valve student twin model to obtain dynamic monitoring data, and the working state of the servo valve is predicted based on the dynamic monitoring data to obtain a predicted state index, wherein the greater the predicted state index is, the higher the abnormal degree of the working state of the servo valve is represented.
The method for predicting the working state of the servo valve based on the dynamic monitoring data comprises the following steps of firstly, constructing a state prediction channel based on a BP neural network, wherein the state prediction channel is a neural network model which can be subjected to iterative optimization in machine learning, and is obtained by performing supervision training through historical training data. The state prediction channel comprises an input layer, a plurality of hidden layers and an output layer, wherein the input data of the input layer are dynamic monitoring data, and the output data of the output layer are prediction state indexes. And acquiring a plurality of sample training data based on the historical servo valve working record, and performing supervised training on the state prediction channel through the plurality of sample training data to obtain a state prediction channel meeting the constraint of the preset training accuracy, and embedding the state prediction channel into the servo valve student twin model. And then inputting dynamic monitoring data obtained by dynamic monitoring into the state prediction channel to obtain the predicted state index.
The working state quality of the servo valve can be clearly and intuitively obtained by obtaining the prediction state index, and a basis is provided for judging the abnormal working state.
Step eight: performing deviation correction analysis on the predicted state index based on the model similarity to obtain a target predicted state index;
and carrying out deviation rectification analysis on the predicted state index based on the model similarity, namely carrying out weighting calculation on the predicted state index by taking the model similarity as a weight coefficient, and taking a weighting calculation result as a target predicted state index to obtain the target predicted state index. And correcting the predicted state index through the model similarity, so that the accuracy of obtaining the target predicted state index can be improved, and the accuracy of judging the abnormal state of the servo valve is improved.
Step nine: and when the target prediction state index reaches a preset index threshold, activating the servo valve teacher twin model to dynamically monitor and early warn the rotary direct-drive servo valve.
Specifically, a preset index threshold is obtained, a person skilled in the art can set the preset index threshold based on the monitoring precision requirement, wherein the higher the monitoring precision requirement is, the smaller the preset index threshold is, the target prediction state index is further judged according to the preset index threshold, when the target prediction state index is larger than the preset index threshold, the condition that the working state of the rotary direct-drive servo valve is abnormal is characterized, namely the rotary direct-drive servo valve is initially monitored to be abnormal, then the servo valve teacher twin model is activated to dynamically monitor the rotary direct-drive servo valve, namely in order to further determine the abnormality degree of the rotary direct-drive servo valve, accurate monitoring and early warning are carried out through the servo valve teacher twin model with higher monitoring accuracy and reliability.
The intelligent monitoring method for the working state of the rotary direct-drive servo valve is applied to an intelligent monitoring system for the working state of the rotary direct-drive servo valve, and can solve the technical problems that the existing monitoring method for the working state of the rotary direct-drive servo valve is low in monitoring data processing efficiency due to the fact that the monitoring data amount is large and redundant data exists, so that abnormal working state of the servo valve cannot be found in time, and the operation stability and safety of a hydraulic system are affected. Firstly, constructing an initial servo valve model according to physical characteristic information of a rotary direct-drive servo valve acquired based on preset physical characteristics; then, reading a preset multidimensional feature, and marking a plurality of feature fingers in the preset multidimensional feature as a class A index; then rendering the class A index parameters dynamically monitored based on the class A index to the initial model of the servo valve to obtain a twin model of a teacher of the servo valve; then, screening the characteristic indexes by combining the historical servo valve working records to obtain a class B index; then rendering the B-type index parameters dynamically monitored based on the B-type index to the servo valve initial model to obtain a servo valve student twin model; further, comparing and analyzing the servo valve teacher twin model and the servo valve student twin model to obtain model similarity; then, dynamically monitoring and analyzing the rotary direct-drive servo valve through the servo valve student twin model to obtain a predicted state index; further, performing deviation correction analysis on the predicted state index based on the model similarity to obtain a target predicted state index; and finally, when the target prediction state index reaches a preset index threshold, activating the servo valve teacher twin model to dynamically monitor and early warn the rotary direct-drive servo valve.
Constructing a complex teacher twin model based on the class A index; then screening the class A index to obtain a simplified key characteristic index which is set as a class B index, and further constructing a simplified version of student twin model based on the class B index; the working state of the servo valve is monitored through the student twin model, and after the abnormal state is primarily monitored, the teacher twin model is activated to dynamically monitor and early warn the servo valve, so that the processing efficiency of the monitoring data of the state of the servo valve can be improved, the abnormal working state of the servo valve can be timely found, and further effective measures can be taken to conduct targeted processing, so that the technical effect of ensuring safe and stable operation of the hydraulic system is achieved.
Further, rendering the class a index parameters dynamically monitored based on the class a index to the initial model of the servo valve to obtain a twin model of a teacher of the servo valve, as shown in fig. 2, the third step of the present application includes:
step 10: taking a first index parameter set in the randomly extracted A-type index parameters as a first model parameter set;
step 20: adjusting the initial model of the servo valve based on the first model parameter set to obtain a first teacher model;
Step 30: removing the first index parameter set from the class A index parameters to obtain a first check parameter set;
Step 40: judging whether the first teacher model reaches a preset constraint according to the first verification parameter set;
Specifically, first, a first index parameter set is randomly extracted from the class a index parameters, the first index parameter set is any one of the class a index parameters, the first index parameter set is used as a first model parameter set, and then the servo valve initial model is adjusted based on the first model parameter set, so that a first teacher model is obtained. And rejecting the first index parameter set from the class A index parameters, and taking the class A index parameters obtained after the first index parameter set is rejected as a first check parameter set. And further judging the first teacher model according to the first checking parameter set, and judging whether the first teacher model meets the preset constraint.
Further, in the step 40, the present application further includes the following steps:
extracting a first check parameter in the first check parameter set;
acquiring a first space distance between the first verification parameter and the first teacher model;
Marking the first check parameter as an inner point if the first spatial distance meets a preset distance threshold value, and marking the first check parameter as an outer point if the first spatial distance does not meet the preset distance threshold value;
Obtaining an internal-external point ratio by the number of the internal points and the number of the external points, wherein the internal-external point ratio is used for representing the reasonable degree of taking the first model parameter group as the parameters of the teacher model;
And when the internal and external point ratio is not in the preset ratio range, the first teacher model does not reach the preset constraint.
Specifically, first, a first calibration parameter in the first calibration parameter set is extracted, the first calibration parameter is any one calibration parameter in the first calibration parameter set, a first spatial distance between the first calibration parameter and the first teacher model is obtained, the first spatial distance is obtained by calculating parameter deviations of the first model parameter set and the first calibration parameter in the first teacher model, and weighting calculation is performed on a plurality of parameter deviations, so that the overall deviation degree of the first calibration parameter and the first teacher model is represented, wherein the greater the first spatial distance is, the greater the overall deviation degree between the first calibration parameter and the first teacher model is represented.
Acquiring a preset distance threshold, wherein the preset distance threshold can be set by a person skilled in the art based on actual conditions, judging the first space distance according to the preset distance threshold, and marking the first verification parameter as an inner point if the first space distance is larger than the preset distance threshold; and if the first space distance is smaller than or equal to the preset distance threshold, marking the first verification parameter as an outer point, counting to obtain the number of inner points and outer points, and calculating the ratio of the number of the inner points to the number of the outer points to obtain an inner point-outer point ratio, wherein the inner point-outer point ratio is used for representing the reasonable degree of taking the first model parameter set as the parameter of the teacher model, and the larger the inner point-outer point ratio is, the higher the reasonable degree of representing the first model parameter set as the parameter of the teacher model is.
And judging the internal and external point ratio according to a preset ratio range, wherein the preset ratio range can be set based on the reasonable degree requirement of the teacher model, the higher the reasonable degree requirement is, the larger the preset ratio range is, when the internal and external point ratio is larger than or equal to the preset ratio range, the first teacher model reaches the preset constraint, and when the internal and external point ratio is smaller than the preset ratio range, the first teacher model does not reach the preset constraint.
Step 50: if not, repeating steps 10-40 until the predetermined constraint is reached, and taking the teacher model obtained therefrom as the servo valve teacher twin model.
Specifically, if the first teacher model does not reach the predetermined constraint, repeating steps 10 to 40 until the first teacher model reaches the predetermined constraint, and taking the teacher model obtained at this time as the servo valve teacher twin model.
The teacher model is judged by setting the preset constraint, and the servo valve teacher twin model reaching the preset constraint is obtained, so that the accuracy and rationality of setting the servo valve teacher twin model can be improved, and the accuracy of monitoring the working state of the servo valve is improved.
Further, the step four of the present application includes the steps of:
extracting a first history record in the history servo valve working record, wherein the first history record comprises a first history servo valve characteristic parameter set and a first history servo valve state index;
Taking a first historical parameter obtained by traversing the plurality of characteristic indexes in the first historical servo valve characteristic parameter set as an independent variable and taking the first historical servo valve state index as an independent variable;
And screening to obtain the class B index according to a correlation result obtained by performing correlation analysis on the independent variable and the dependent variable.
Specifically, a first history record in the history servo valve working records is extracted, wherein the first history record is any one history working record in the history servo valve working records, and comprises a first history servo valve characteristic parameter set and a first history servo valve state index.
And extracting index parameters of the first historical servo valve characteristic parameter set based on the characteristic indexes to obtain a first historical parameter, wherein the first historical parameter is used as an independent variable, and the first historical servo valve state index is used as a dependent variable. And further carrying out correlation analysis on the independent variable and the dependent variable to obtain a correlation analysis result, and screening the characteristic indexes based on the correlation analysis result to obtain a class B index.
Further, the class B index is obtained by screening according to a correlation result obtained by performing correlation analysis on the independent variable and the dependent variable, and the application further comprises the following steps:
Drawing a scatter diagram of each variable in the independent variables and the dependent variables to form a scatter diagram set;
sequentially analyzing all the scatter diagrams in the scatter diagram set to obtain a plurality of maximum information coefficients;
descending the order of the plurality of maximum information coefficients and reversely matching to obtain an independent variable sequence;
And extracting an independent variable of a preset ranking threshold value in the independent variable sequence to form the class B index.
Specifically, first, a scatter diagram of each variable in the independent variables and the dependent variable is drawn to form a scatter diagram set, wherein the scatter diagram is a graphical representation method for showing the relationship between the two variables, in the scatter diagram, each point represents an observed value, the horizontal axis generally represents one variable, the vertical axis represents the other variable, and whether the relationship, the positive correlation or the negative correlation exists between the two variables and the strength of the relationship can be primarily judged through the distribution mode and the trend of the observed points.
And further sequentially analyzing each scatter diagram in the scatter diagram set to obtain a plurality of maximum information coefficients, wherein the maximum information coefficients are used for representing the association degree of the independent variables and the dependent variables, and the larger the maximum information coefficients are, the larger the association degree is represented. And then arranging the plurality of maximum information coefficients from small to large according to the coefficient size to obtain a maximum information coefficient sequence, and further matching to obtain an independent variable sequence corresponding to the maximum information coefficient sequence. And then extracting independent variables of a preset ranking threshold value in the independent variable sequence to form the class B index, wherein the preset ranking threshold value can be set on the basis of actual conditions, for example: and setting a preset ranking threshold to be 3, namely extracting the first 3 independent variables in the independent variable sequence.
By carrying out correlation analysis on the characteristic index parameters and the state indexes of the servo valve and selecting a plurality of indexes with the highest degree of correlation with the state indexes of the servo valve as B-type indexes, the accuracy of setting the B-type indexes can be improved, and therefore the accuracy and rationality of construction of the student twin model of the servo valve are improved.
In summary, the intelligent monitoring method for the working state of the rotary direct-drive servo valve provided by the application has the following technical effects:
1. Constructing a complex teacher twin model based on the class A index; then screening the class A index to obtain a simplified key characteristic index which is set as a class B index, and further constructing a simplified version of student twin model based on the class B index; the working state of the servo valve is monitored through the student twin model, and after the abnormal state is primarily monitored, the teacher twin model is activated to dynamically monitor and early warn the servo valve, so that the processing efficiency of the monitoring data of the state of the servo valve can be improved, the abnormal working state of the servo valve can be timely found, and further effective measures can be taken to conduct targeted processing, so that the technical effect of ensuring safe and stable operation of the hydraulic system is achieved.
2. The teacher model is judged by setting the preset constraint, and the servo valve teacher twin model reaching the preset constraint is obtained, so that the accuracy and rationality of setting the servo valve teacher twin model can be improved, and the accuracy of monitoring the working state of the servo valve is improved.
3. By carrying out correlation analysis on the characteristic index parameters and the state indexes of the servo valve and selecting a plurality of indexes with the highest degree of correlation with the state indexes of the servo valve as B-type indexes, the accuracy of setting the B-type indexes can be improved, and therefore the accuracy and rationality of construction of the student twin model of the servo valve are improved.
Example two
Based on the same inventive concept as the intelligent monitoring method of the working state of the rotary direct-drive servo valve in the foregoing embodiment, the application also provides an intelligent monitoring system of the working state of the rotary direct-drive servo valve, referring to fig. 3, the system includes:
The servo valve initial model building module 11 is used for building a servo valve initial model according to the physical characteristic information of the rotary direct-drive servo valve acquired based on the preset physical characteristics;
A class a index marking module 12, wherein the class a index marking module 12 is configured to read a predetermined multidimensional feature, and mark a plurality of feature fingers in the predetermined multidimensional feature as class a indexes;
The servo valve teacher twin model obtaining module 13 is used for rendering the class A index parameters dynamically monitored based on the class A index to the servo valve initial model to obtain a servo valve teacher twin model;
The class B index obtaining module 14 is used for screening the characteristic indexes by combining the historical servo valve working records to obtain class B indexes;
The servo valve student twin model obtaining module 15 is used for rendering the B-type index parameters dynamically monitored based on the B-type index to the servo valve initial model to obtain a servo valve student twin model;
The model similarity obtaining module 16 is used for comparing and analyzing the servo valve teacher twin model and the servo valve student twin model to obtain model similarity;
The predicted state index obtaining module 17 is used for dynamically monitoring and analyzing the rotary direct-drive servo valve through the servo valve student twin model to obtain a predicted state index;
The prediction state index correction module 18, where the prediction state index correction module 18 is configured to perform correction analysis on the prediction state index based on the model similarity to obtain a target prediction state index;
And the dynamic monitoring and early warning module 19 is used for activating the servo valve teacher twin model to dynamically monitor and early warn the rotary direct-drive servo valve when the target prediction state index reaches a preset index threshold value.
Further, the servo valve teacher twinning model obtaining module 13 in the system is further configured to: step 10: taking a first index parameter set in the randomly extracted A-type index parameters as a first model parameter set; step 20: adjusting the initial model of the servo valve based on the first model parameter set to obtain a first teacher model; step 30: removing the first index parameter set from the class A index parameters to obtain a first check parameter set; step 40: judging whether the first teacher model reaches a preset constraint according to the first verification parameter set; step 50: if not, repeating steps 10-40 until the predetermined constraint is reached, and taking the teacher model obtained therefrom as the servo valve teacher twin model.
Further, the servo valve teacher twinning model obtaining module 13 in the system is further configured to: extracting a first check parameter in the first check parameter set; acquiring a first space distance between the first verification parameter and the first teacher model; marking the first check parameter as an inner point if the first spatial distance meets a preset distance threshold value, and marking the first check parameter as an outer point if the first spatial distance does not meet the preset distance threshold value; obtaining an internal-external point ratio by the number of the internal points and the number of the external points, wherein the internal-external point ratio is used for representing the reasonable degree of taking the first model parameter group as the parameters of the teacher model; and when the internal and external point ratio is not in the preset ratio range, the first teacher model does not reach the preset constraint.
Further, the class B indicator obtaining module 14 in the system is further configured to: extracting a first history record in the history servo valve working record, wherein the first history record comprises a first history servo valve characteristic parameter set and a first history servo valve state index; taking a first historical parameter obtained by traversing the plurality of characteristic indexes in the first historical servo valve characteristic parameter set as an independent variable and taking the first historical servo valve state index as an independent variable; and screening to obtain the class B index according to a correlation result obtained by performing correlation analysis on the independent variable and the dependent variable.
Further, the class B indicator obtaining module 14 in the system is further configured to: drawing a scatter diagram of each variable in the independent variables and the dependent variables to form a scatter diagram set; sequentially analyzing all the scatter diagrams in the scatter diagram set to obtain a plurality of maximum information coefficients; descending the order of the plurality of maximum information coefficients and reversely matching to obtain an independent variable sequence; and extracting an independent variable of a preset ranking threshold value in the independent variable sequence to form the class B index.
In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the difference from other embodiments, so that the foregoing method and specific example for intelligently monitoring the operating state of a rotary direct-drive servo valve in the first embodiment are equally applicable to the foregoing intelligent monitoring system for the operating state of a rotary direct-drive servo valve in the first embodiment, and by the foregoing detailed description of the foregoing method for intelligently monitoring the operating state of a rotary direct-drive servo valve, those skilled in the art can clearly know the foregoing intelligent monitoring system for the operating state of a rotary direct-drive servo valve in the first embodiment, so that the description is omitted herein for brevity. For the system disclosed in the embodiment, since the system corresponds to the method disclosed in the embodiment, the description is simpler, and the relevant points refer to the description of the method section.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
It will be apparent to those skilled in the art that various modifications and variations can be made to the present application without departing from the spirit or scope of the application. Thus, if such modifications and variations of the present application fall within the scope of the present application and the equivalent techniques thereof, the present application is also intended to include such modifications and variations.
Claims (2)
1. An intelligent monitoring method for the working state of a rotary direct-drive servo valve is characterized by comprising the following steps:
Constructing an initial servo valve model according to physical characteristic information of the rotary direct-drive servo valve acquired based on the preset physical characteristics;
Reading a preset multidimensional feature, and marking a plurality of feature fingers in the preset multidimensional feature as a class A index;
rendering the class A index parameters dynamically monitored based on the class A index to the initial model of the servo valve to obtain a twin model of a teacher of the servo valve;
screening the characteristic indexes by combining the historical servo valve working records to obtain a class B index;
Rendering the B-type index parameters dynamically monitored based on the B-type index to the servo valve initial model to obtain a servo valve student twin model;
comparing and analyzing the servo valve teacher twin model and the servo valve student twin model to obtain model similarity;
dynamically monitoring and analyzing the rotary direct-drive servo valve through the servo valve student twin model to obtain a predicted state index;
performing deviation correction analysis on the predicted state index based on the model similarity to obtain a target predicted state index;
When the target prediction state index reaches a preset index threshold, activating the servo valve teacher twin model to dynamically monitor and early warn the rotary direct-drive servo valve;
Rendering the class A index parameters dynamically monitored based on the class A index to the initial model of the servo valve to obtain a twin model of a teacher of the servo valve, wherein the method comprises the following steps:
step 10: taking a first index parameter set in the randomly extracted A-type index parameters as a first model parameter set;
step 20: adjusting the initial model of the servo valve based on the first model parameter set to obtain a first teacher model;
Step 30: removing the first index parameter set from the class A index parameters to obtain a first check parameter set;
Step 40: judging whether the first teacher model reaches a preset constraint according to the first verification parameter set;
Step 50: if not, repeating the steps 10 to 40 until the predetermined constraint is reached, and taking the teacher model obtained therefrom as the servo valve teacher twin model;
The step 40 includes:
extracting a first check parameter in the first check parameter set;
acquiring a first space distance between the first verification parameter and the first teacher model;
Marking the first check parameter as an inner point if the first spatial distance meets a preset distance threshold value, and marking the first check parameter as an outer point if the first spatial distance does not meet the preset distance threshold value;
Obtaining an internal-external point ratio by the number of the internal points and the number of the external points, wherein the internal-external point ratio is used for representing the reasonable degree of taking the first model parameter group as the parameters of the teacher model;
When the internal-external point ratio is in a preset ratio range, the first teacher model achieves the preset constraint, and when the internal-external point ratio is not in the preset ratio range, the first teacher model does not achieve the preset constraint;
The step of screening the characteristic indexes by combining the historical servo valve working records to obtain a class B index comprises the following steps:
extracting a first history record in the history servo valve working record, wherein the first history record comprises a first history servo valve characteristic parameter set and a first history servo valve state index;
Taking a first historical parameter obtained by traversing the plurality of characteristic indexes in the first historical servo valve characteristic parameter set as an independent variable and taking the first historical servo valve state index as an independent variable;
screening to obtain the class B index according to a correlation result obtained by performing correlation analysis on the independent variable and the dependent variable;
the step of screening the class B index according to a correlation result obtained by performing correlation analysis on the independent variable and the dependent variable comprises the following steps:
Drawing a scatter diagram of each variable in the independent variables and the dependent variables to form a scatter diagram set;
sequentially analyzing all the scatter diagrams in the scatter diagram set to obtain a plurality of maximum information coefficients;
descending the order of the plurality of maximum information coefficients and reversely matching to obtain an independent variable sequence;
And extracting an independent variable of a preset ranking threshold value in the independent variable sequence to form the class B index.
2. An intelligent monitoring system for the operating condition of a rotary direct drive servo valve, comprising the steps of:
the servo valve initial model building module is used for building a servo valve initial model according to the physical characteristic information of the rotary direct-drive servo valve acquired based on the preset physical characteristic;
The A-type index marking module is used for reading preset multidimensional features and marking a plurality of feature fingers in the preset multidimensional features as A-type indexes;
the servo valve teacher twin model obtaining module is used for rendering the class A index parameters dynamically monitored based on the class A index to the servo valve initial model to obtain a servo valve teacher twin model;
the B-type index obtaining module is used for screening the characteristic indexes by combining the historical servo valve working records to obtain B-type indexes;
The servo valve student twin model obtaining module is used for rendering B-type index parameters dynamically monitored based on the B-type index to the servo valve initial model to obtain a servo valve student twin model;
The model similarity obtaining module is used for comparing and analyzing the servo valve teacher twin model and the servo valve student twin model to obtain model similarity;
The prediction state index obtaining module is used for carrying out dynamic monitoring analysis on the rotary direct-drive servo valve through the servo valve student twin model to obtain a prediction state index;
The prediction state index deviation rectifying module is used for carrying out deviation rectifying analysis on the prediction state index based on the model similarity to obtain a target prediction state index;
And the dynamic monitoring and early warning module is used for activating the servo valve teacher twin model to dynamically monitor and early warn the rotary direct-drive servo valve when the target prediction state index reaches a preset index threshold.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410470060.3A CN118066182B (en) | 2024-04-18 | 2024-04-18 | Intelligent monitoring method and system for working state of rotary direct-drive servo valve |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410470060.3A CN118066182B (en) | 2024-04-18 | 2024-04-18 | Intelligent monitoring method and system for working state of rotary direct-drive servo valve |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118066182A CN118066182A (en) | 2024-05-24 |
| CN118066182B true CN118066182B (en) | 2024-06-21 |
Family
ID=91107995
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410470060.3A Active CN118066182B (en) | 2024-04-18 | 2024-04-18 | Intelligent monitoring method and system for working state of rotary direct-drive servo valve |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118066182B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110045608A (en) * | 2019-04-02 | 2019-07-23 | 太原理工大学 | Based on the twin mechanical equipment component structural dynamic state of parameters optimization method of number |
| CN115026835A (en) * | 2022-07-18 | 2022-09-09 | 上海飒智智能科技有限公司 | Method for optimizing overall performance of robot mechanical arm servo system |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| FR1146775A (en) * | 1955-02-09 | 1957-11-14 | British Messier Ltd | Improvements to servo motor systems |
| FR1488803A (en) * | 1965-08-06 | 1967-07-13 | Hydrosteer Ltd | System applicable to mechanical vehicle steering or other servo-control devices |
| CN109571476A (en) * | 2018-12-14 | 2019-04-05 | 南京理工大学 | The twin real time job control of industrial robot number, monitoring and precision compensation method |
| CN110605709B (en) * | 2019-09-25 | 2022-11-08 | 西南交通大学 | Digital twin and precise filtering drive robot integration system and use method thereof |
| CN115246125A (en) * | 2022-01-13 | 2022-10-28 | 聊城大学 | Robot Vision Servo Control Method and System Based on Hybrid Feedback |
| CN114952840B (en) * | 2022-05-26 | 2024-12-06 | 中国第一汽车股份有限公司 | Pneumatic servo flexible gripper using ultrasonic obstacle avoidance device and obstacle avoidance method thereof |
| CN117021066A (en) * | 2023-05-26 | 2023-11-10 | 浙江大学 | Robot vision servo motion control method based on deep reinforcement learning |
| CN116985134A (en) * | 2023-08-11 | 2023-11-03 | 南京工程学院 | Servo control method and system based on robot image processing |
| CN117075476A (en) * | 2023-08-21 | 2023-11-17 | 华中科技大学 | Digital twin method and system of servo press based on force-energy-displacement conversion |
-
2024
- 2024-04-18 CN CN202410470060.3A patent/CN118066182B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110045608A (en) * | 2019-04-02 | 2019-07-23 | 太原理工大学 | Based on the twin mechanical equipment component structural dynamic state of parameters optimization method of number |
| CN115026835A (en) * | 2022-07-18 | 2022-09-09 | 上海飒智智能科技有限公司 | Method for optimizing overall performance of robot mechanical arm servo system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118066182A (en) | 2024-05-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113807570B (en) | XGBoost-based reservoir dam risk level assessment method and system | |
| CN116237817B (en) | Five-axis linkage numerical control machine intelligent monitoring system based on Internet of things | |
| CN110636066B (en) | Network security threat situation assessment method based on unsupervised generative reasoning | |
| CN110530650B (en) | Method for monitoring performance state of heavy-duty gas turbine based on generalized regression neural network and box diagram analysis | |
| CN115618732B (en) | Nuclear reactor digital twin key parameter autonomous optimization data inversion method | |
| CN118568647B (en) | Industrial equipment fault intelligent detection method and system based on digital twin | |
| CN114548494B (en) | Visual cost data prediction intelligent analysis system | |
| CN113065223B (en) | Multi-level probability correction method for digital twin model of tower mast cluster | |
| CN117391310B (en) | Power grid equipment operation state prediction and optimization method based on digital twin technology | |
| CN117560300B (en) | Intelligent internet of things flow prediction and optimization system | |
| CN114266289A (en) | Complex equipment health state assessment method | |
| CN116861765A (en) | Rolling bearing fault diagnosis method and system based on digital twin and DBN | |
| CN111260082A (en) | Spatial object motion trajectory curve prediction method based on neural network | |
| CN118066182B (en) | Intelligent monitoring method and system for working state of rotary direct-drive servo valve | |
| CN118503894B (en) | Lithium battery quality detection system based on process index data analysis | |
| CN118133892B (en) | A fault detection method and system based on self-attention and neural differential equation | |
| CN119130087A (en) | Edge power distribution optimization method and system for power grid equipment | |
| CN119045448A (en) | Method and system for sensing trusted state of DCS controller | |
| CN112688836B (en) | Energy routing equipment online dynamic sensing method based on deep self-coding network | |
| Vélez Sánchez et al. | Data collection: Use and transformation in predictive maintenance models | |
| CN117407757A (en) | Intelligent fault diagnosis method and system for production line lifting appliance based on PCA-SVDD | |
| CN108490912A (en) | A kind of multi-modal process modal identification method based on pivot similarity analysis | |
| CN114076680A (en) | Engine assembly detection method, system, storage medium and electronic device | |
| CN114510518B (en) | Self-adaptive aggregation method and system for massive structured data and electronic equipment | |
| CN118656772B (en) | Real-time oil reservoir data stream processing and storage optimizing method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |