Digital twinborn visualization method, platform and equipment for valve monitoring
Through digital twin visualization methods and platforms, deep learning algorithms are used to monitor and analyze valve status, and the problems of inaccurate valve monitoring and difficulty in fault identification in the existing technology are solved, real-time monitoring of valve status and analysis of parameter relationships are realized, and the improvement of improvement of valve status is achieved. Monitor efficiency and productivity.
Patent Information
- Application Number
- CN202411190834.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-28
- Publication Date
- 2024-12-06
AI Technical Summary
The prior art cannot accurately and quickly monitor the operating status of the valve, and cannot establish the relationship between the valve status characterization parameters, making it difficult to identify and judge the faulty valve, especially in the complex operating conditions of high parameters, the safety and functional reliability of the valve are proposed. Higher requirements.
Using digital twin visualization methods and platforms, by collecting real production line equipment data in the factory, building a visual three-dimensional digital twin model, using deep learning algorithms to conduct simulation tests and predictions, monitoring the valve status in real time, and feedback predicted data through the data interface module to realize abnormal alarms and parameter relationship analysis.
Real-time monitoring and analysis of valve opening and pressure state parameters is realized, which improves monitoring efficiency and productivity, reduces the destructiveness in system operation, and extends the service life of the valve.
Smart Images

Figure CN119091047A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of valve monitoring technology, and more specifically to a digital twin visualization method, platform and device for valve monitoring. Background Art
[0002] With the rapid development of information technology, informatization and intelligence are constantly changing the face of industrial and agricultural production and people's social life. As the terminal actuator for controlling the movement of fluid in the pipeline, the valve operating state plays a key role in the operation of the system.
[0003] At present, the online operation status of valves cannot be monitored accurately and quickly; and the parameters characterizing the valve status, such as torque, valve position, pressure, etc., are often tested independently on different instruments, and the relationship between the parameters cannot be established, and the faulty valves cannot be identified and judged. In addition, high-parameter and complex working conditions such as high temperature, high pressure, deep cold, high vacuum, strong corrosion, radioactivity, toxicity, flammability and explosion are increasing in industrial production, which puts forward higher and stricter requirements on the safety of valve use, reliability of functions and real-time monitoring of valves.
[0004] Therefore, how to improve the safety of valve monitoring is an issue that technical personnel in this field need to solve urgently. Summary of the invention
[0005] In view of this, the present invention provides a digital twin visualization method, platform and device for valve monitoring, which are used to achieve safe monitoring of valves.
[0006] In order to achieve the above object, the present invention adopts the following technical solution:
[0007] The present invention first discloses a digital twin visualization method for valve monitoring, comprising the following steps:
[0008] Collect the real production line equipment data of the factory;
[0009] Build a visual 3D digital twin model of the factory's real production line equipment based on the factory's real production line equipment data;
[0010] Using the visualized three-dimensional digital twin model to complete a simulation production test, and predicting the results of the simulation test based on a deep learning algorithm;
[0011] The predicted data is fed back to the factory's actual production line equipment through the data interface module.
[0012] Furthermore, the above method also includes initiating a data anomaly alarm when the collected real production line equipment data is abnormal.
[0013] The present invention also discloses a digital twin visualization platform for valve monitoring, comprising:
[0014] Data interface module, digital twin visualization module and business window module,
[0015] Wherein, the data interface module is used to collect the real production line equipment data of the factory, and the real production line equipment data of the factory includes the factory valve detection equipment data and the factory valve operation equipment data;
[0016] The digital twin visualization module includes a twin model and a twin panel; the twin model is used to construct a visualized three-dimensional digital twin model corresponding to the real production line equipment of the factory based on the real production line equipment data of the factory; the twin panel is used to display the basic information of the twin model, and the basic information of the twin model includes the twin name, twin ID and twin model;
[0017] The business window module is used to complete the simulation production test using the twin model, and predict the results of the simulation test based on the deep learning algorithm, and feed back the predicted data to the factory's real production line equipment through the data interface module.
[0018] Furthermore, the data interface module uses the MQTT gateway to transmit the collected factory real production line equipment data to the digital twin visualization module.
[0019] Furthermore, the factory valve detection equipment data includes sensor parameter data for detecting valve operation, and the valve operation equipment data includes valve opening and closing state data, valve front and rear pressure data, valve temperature data, fluid flow data through the valve, valve vibration data, and valve power consumption data measured based on sensors.
[0020] Furthermore, the simulation production test includes a valve opening and flow test, a valve opening and valve core position test, a valve flow and valve core pressure relationship test, and a valve positioner valve core stroke test.
[0021] Furthermore, the business window module predicts the results of the simulation test based on the deep learning algorithm, specifically including the following steps:
[0022] Collect valve operation equipment data and perform preprocessing, wherein the preprocessing includes data denoising, filling missing values, and data normalization;
[0023] Divide the preprocessed valve operation equipment data into a training set, a validation set, and a test set;
[0024] Build a convolutional neural network model and use the training set to train the convolutional neural network model;
[0025] Use the validation set and test set to evaluate the trained convolutional neural network model, and perform hyperparameter tuning or model structure adjustment based on the evaluation results to obtain a trained convolutional neural network model;
[0026] The pre-optimized valve operation equipment data is input into the trained convolutional neural network model to obtain simulation test prediction data.
[0027] Furthermore, the above-mentioned digital twin visualization platform for valve monitoring also includes a background management module.
[0028] Furthermore, the background management module includes a user management submodule, a device management submodule, a historical data management submodule, and a log management submodule;
[0029] The user management submodule is used to configure permissions for users, and different users have different permissions;
[0030] The device management submodule is used to manage the three-dimensional digital twin model in the digital twin visualization module and determine whether the three-dimensional digital twin model is displayed;
[0031] The historical data management submodule is used to view the equipment operation data within a set time period;
[0032] The log management submodule is used to view user operation logs, user execution logs and user login logs in real time.
[0033] The present invention also discloses a digital twin visualization device for valve monitoring, wherein the device includes a computer program, and when the computer program is executed, the digital twin visualization method for valve monitoring described in any one of the present invention is implemented.
[0034] It can be seen from the above technical solutions that, compared with the prior art, the present invention discloses a digital twin visualization method, platform and device for valve monitoring, which has the following beneficial effects:
[0035] 1. The present invention uses a visualized digital twin model to monitor and display in real time the valve opening, temperature before and after the valve, pressure and other status parameters in industrial production, making it convenient for operators to monitor the safety and status of the valve during the production process.
[0036] 2. The present invention can display the valve position, pressure and torque in the form of graphs under the same business view, establish connections between the state characterization parameters, and intuitively analyze the relationship between the parameters; historical data can be used to guide industrial production, and global big data summary analysis can be used to archive, analyze and warn abnormal data.
[0037] 3. The present invention can improve monitoring and measurement efficiency, increase efficiency and productivity, reduce the destructiveness of valves during system operation, and increase the service life of valves. BRIEF DESCRIPTION OF THE DRAWINGS
[0038] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.
[0039] Figure 1 A schematic diagram of the overall process of the digital twin visualization method for valve monitoring provided by the present invention;
[0040] Figure 2 A schematic diagram of the overall structure of the digital twin visualization platform for valve monitoring provided by the present invention;
[0041] Figure 3 A schematic diagram of the data interface docking process of the MQTT communication protocol provided by the present invention. DETAILED DESCRIPTION
[0042] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0043] Example 1
[0044] like Figure 1 As shown, an embodiment of the present invention discloses a digital twin visualization method for valve monitoring, comprising the following steps:
[0045] Collect the real production line equipment data of the factory;
[0046] Build a visual 3D digital twin model of the factory's real production line equipment based on the factory's real production line equipment data;
[0047] Use the constructed visualized 3D digital twin model to complete the simulation production test, and predict the results of the simulation test based on the deep learning algorithm;
[0048] The predicted data is fed back to the factory's actual production line equipment through the data interface module.
[0049] In this embodiment, the results of the simulation test are predicted based on the deep learning algorithm, which may specifically include the following steps: collecting valve operation equipment data and preprocessing it, wherein the preprocessing includes data denoising, filling missing values, and data normalization; dividing the preprocessed valve operation equipment data into a training set, a validation set, and a test set; constructing a convolutional neural network model, and training the convolutional neural network model using the training set; evaluating the trained convolutional neural network model using the validation set and the test set, and performing hyperparameter tuning or model structure adjustment based on the evaluation results to obtain a trained convolutional neural network model; inputting the pre-optimized valve operation equipment data into the trained convolutional neural network model to obtain simulation test prediction data.
[0050] As a preferred technical solution, the above method also includes, when the collected real production line equipment data is abnormal, starting a data abnormality alarm.
[0051] Example 2
[0052] like Figure 2 As shown, this embodiment discloses a digital twin visualization platform for valve monitoring, which mainly includes a data interface module, a digital twin visualization module and a business window module.
[0053] Among them, the data interface module is used to collect the equipment data of the factory's real production line. In this embodiment, the equipment data of the factory's real production line includes the factory valve detection equipment data and the factory valve operation equipment data; the digital twin visualization module includes a twin model and a twin panel; the twin model is used to construct a visual three-dimensional digital twin model corresponding to the factory's real production line equipment based on the factory's real production line equipment data; the twin panel is mainly used to display the basic information of the twin model, and the basic information of the twin model may include the twin name, twin ID and twin model; the business window module is mainly used to use the twin model to complete the simulation production test, and predict the results of the simulation test based on the deep learning algorithm, and feed back the predicted data to the factory's real production line equipment through the data interface module.
[0054] In addition, the digital twin visualization platform also includes a background management module, which includes a user management sub-module, a device management sub-module, a historical data management sub-module, and a log management sub-module; among them, the user management sub-module is used to configure user permissions, and different users have different permissions; the device management sub-module is used to manage the three-dimensional digital twin model in the digital twin visualization module and determine whether the three-dimensional digital twin model is displayed; the historical data management sub-module is used to view the equipment operation data within a set time period; the log management sub-module is used to view user operation logs, user execution logs, and user login logs in real time.
[0055] In this embodiment, the data interface module uses the MQTT gateway to transmit the collected factory real production line equipment data to the digital twin visualization module.
[0056] In this embodiment, the factory valve detection equipment data includes sensor parameter data for detecting valve operation, and the valve operation equipment data may include valve opening and closing status data based on sensor measurements, valve front and rear pressure data, valve temperature data, fluid flow data through the valve, valve vibration data, and valve power consumption data.
[0057] In this embodiment, the simulation production test includes a valve opening and flow test, a valve opening and valve core position test, a valve flow and valve core pressure relationship test, and a valve positioner valve core stroke test.
[0058] In this embodiment, the business window module predicts the results of the simulation test based on the deep learning algorithm, which specifically includes the following steps: collecting valve operation equipment data and preprocessing it, wherein the preprocessing includes data denoising, filling missing values, and data normalization; dividing the preprocessed valve operation equipment data into a training set, a validation set, and a test set; constructing a convolutional neural network model, and training the convolutional neural network model using the training set; evaluating the trained convolutional neural network model using the validation set and the test set, and performing hyperparameter tuning or model structure adjustment based on the evaluation results to obtain a trained convolutional neural network model; inputting the pre-optimized valve operation equipment data into the trained convolutional neural network model to obtain simulation test prediction data.
[0059] Furthermore, the business window module in the present invention can use algorithms or technologies such as multidimensional data analysis, machine learning, neural networks, deep learning, etc. to calculate optimization plans, prediction and warning models, emergency command strategies, etc. to complete self-optimization.
[0060] Specifically, multidimensional data analysis includes: valve physical entity data, mainly including operating status and working conditions; virtual mirror data, consisting of valve model parameters and model operation data; fusion processing data of physical data and virtual mirror data, which can be obtained through data fusion algorithms (such as neural networks, DS algorithms and Kalman filtering).
[0061] As the main control unit in the industrial fluid delivery system, the industrial control valve is a vital link to ensure industrial production safety and smooth system operation. It has the advantages of stable performance and quick response. However, due to the influence of harsh working conditions such as frequent mechanical movements and the strong corrosiveness of the conveying medium, the control valve is prone to various faults. Therefore, improving the fault diagnosis ability of the control valve is of great significance to ensure the safe and stable operation of the industrial fluid delivery system. Taking valve fault diagnosis as an example, the deep learning process is as follows:
[0062] Data collection: First, select the parameters that can represent the working state of the valve as the analysis object. The action time of the valve is very short, and it is usually a static state after reaching a given position, which is difficult to describe with state parameters such as temperature and vibration. Therefore, the result parameters of the valve action, such as the working fluid flow, valve opening, valve passing capacity, etc., are used as the analysis object, that is, the valve action result parameters are collected.
[0063] Data preparation: First, prepare the collected data into a format that can be used by the machine learning algorithm. Depending on the algorithm library, the algorithm can adapt to different types of input formats.
[0064] Data segmentation: divide the data into training data and test data, use the training data to train the algorithm, and use the test data to calculate the final accuracy of the generated model. The test data does not participate in the algorithm training.
[0065] Algorithm selection: Convolutional neural networks have the ability to represent learning and can perform translation-invariant classification of input information according to their hierarchical structure. A type of feedforward neural network that includes convolution calculations and has a deep structure. Here we choose convolutional neural networks.
[0066] Algorithm training: After selecting the algorithm, the model training begins. The training model is performed on the training dataset. The weights / parameters of most algorithms are randomly assigned at the beginning of training and improved in each iteration. During the algorithm training process, it is run multiple times on the training dataset to produce results.
[0067] Parameter Tuning: After choosing the right algorithm, improve its parameters to get better performance.
[0068] Model usage: Use the model to predict values for new data.
[0069] The following is a further explanation of the visualization platform:
[0070] In the embodiment of the present invention, the digital twin visualization platform for valve status monitoring consists of four parts: data interface, visualization interface, business window, and background management. The data interface is used to connect data with factory valve detection equipment and operation equipment. The visualization interface is used to display each digital twin model and its information. The business window is a different operation window made according to different operation requirements. The background management is used to manage user information, historical data, logs, alarms, and integrated operation and maintenance.
[0071] Data interface:
[0072] The data interface of the digital twin visualization platform for valve status monitoring can adapt to common integration protocols, such as HTTP(s), OPC, MQTT, etc. Figure 3As shown, this embodiment takes the MQTT communication protocol as an example to introduce the docking process of digital twin platform data with valve detection equipment and other operating equipment data.
[0073] MQTT (Message Queuing Telemetry Transport) is a "lightweight" communication protocol based on the publish / subscribe model, which is built on the TCP / IP protocol. The biggest advantage of MQTT is that it can provide real-time and reliable message services for remote connected devices with very little code and limited bandwidth. As an instant messaging protocol with low overhead and low bandwidth, it is widely used in the Internet of Things, small devices, mobile applications, etc. MQTT is a client-server based message publishing / subscription transmission protocol. The MQTT protocol is lightweight, simple, open and easy to implement, which makes it very widely applicable.
[0074] The MQTT gateway is used to actively collect device data and perform protocol analysis. After the data is standardized, the MQTT protocol is used as the uplink access protocol. After the data is obtained, the data is processed and packaged, and then sent to the database of the digital twin platform. The valve status monitoring digital twin visualization platform obtains data by subscribing to the topic of the message. When the valve status monitoring digital twin visualization platform wants to send instructions to the device or write data, the platform publishes a message to the corresponding gateway, and then the gateway transmits the data to the device, thereby realizing the connection between the valve status monitoring digital twin visualization platform and the field equipment data, that is, realizing the reporting and control of data.
[0075] The lower computers in industrial sites are mostly PLC controllers, among which Siemens PLC is the most widely used. This type of PLC data communication method uses the Profinet protocol, while the digital twin visualization platform supports communication protocols such as MQTT, OPC, HTTP, etc., but does not support the Profinet protocol. A protocol conversion gateway is required to convert the Profinet protocol into a protocol supported by the digital twin visualization platform. Here, a Profinet to MQTT protocol conversion gateway is used to convert the Profinet protocol into the MQTT protocol, which facilitates the digital twin visualization platform to collect and process industrial site data.
[0076] MQTT collects industrial field data such as equipment data and valve data. The collected equipment data includes: pressure and displacement measured by sensors, speed and power of fans, etc. The collected valve data includes: valve core displacement, current, valve leakage and other valve working data; the relationship between MQTT and valves: the data collected by sensors helps to judge the working status of valves.
[0077] The MQTT control message consists of three parts, namely a fixed header, a variable header and a payload (the collected valve data, device data, etc. are stored in the payload part of the message), of which the fixed header is included in all control messages, and the variable header and payload are included in some control messages. After data collection, a protocol parsing program is written on the digital twin visualization platform according to the MQTT standard message format to perform message parsing. After message parsing, valid data (data after removing the command, address, data length and other information in the message), that is, standardized data, can be extracted. In this embodiment, the message parsing part is performed on the digital twin visualization platform and parsed by a script program. The collected real data of the equipment and the valve data are input into the digital twin visualization platform. On the one hand, the equipment information can be displayed on the visualization interface, which is convenient for operators to view; on the other hand, the monitoring data can be stored in the database of the visualization platform for data analysis.
[0078] Visual interface:
[0079] The visualization interface consists of a twin model and a twin panel. Using three-dimensional graphics as the main visualization method, it helps to intuitively and clearly understand and perceive data, and realize monitoring and analysis from multiple dimensions of time and space.
[0080] A twin (Digital Twin) is a mirror image of a physical object in the digital world. It is interconnected with the physical object, reflecting the data and status of the physical world in real time, and can reversely control the physical entity through the twin. The twin model is a digital model of an existing or future physical object. It can sense, diagnose, and predict the status of the physical object in real time through actual measurement, simulation, and data analysis, and regulate the behavior of the physical object through optimization and instructions. It can evolve itself through mutual learning between related digital models, and improve the decision-making of stakeholders during the life cycle of the physical object. Through the digital twin model, it is possible to fully monitor the key parameters of the system and analyze the various performances of the system under unconventional conditions, such as polluted, noisy, and harsh working environments. Using the digital twin model for virtual testing shortens the time and cost of testing and analysis, and can optimize the test parameters based on the virtual test results.
[0081] Specifically, the performance items that need to be monitored include the relationship between valve opening and flow and pressure, the relationship between valve positioner valve core stroke, position and valve opening, etc. The digital twin model evolves itself to improve the life cycle of physical entity objects, such as using genetic algorithms to process and analyze historical monitoring data (electrical signals, flow signals, pressure signals), analyze and predict valve status, thereby realizing predictive maintenance of valve industrial sites, and then realizing the management of the entire life cycle of physical entities.
[0082] The twin panel is a two-dimensional screen used to display data such as basic twin information, monitoring alarm information, and related information. Select a twin with the mouse to pop up the twin panel. The basic twin information includes the twin name, twin ID, twin model, etc. The twin name is the display name of the twin instance, such as the device name or instrument name. The twin ID is the unique identification field of the twin and cannot be repeated with other twins. The system displays the model corresponding to the twin through the mapping relationship between the twin model and the model.
[0083] The twin model consists of a virtual three-dimensional model and is a real-time mirror of the physical entity layer in the virtual information space.
[0084] The twin panel is mainly composed of the valve management module. The valve management module monitors the key parameters of the valves in the entire workshop, performs real-time analysis of the valve operation data, and intuitively displays the valves with problems through data visualization. At the same time, it provides fault warnings for the valves to detect potential hazards.
[0085] Refineries, steel mills, power plants, etc. are equipped with a large number of valves for measuring and controlling liquids, air or gas. The twin panel mainly consists of key parameters for valve monitoring, such as leakage, valve core displacement, relative opening, and pressure before and after the valve.
[0086] Business Window:
[0087] Multiple windows can be set up in the business window according to the specific production process, such as the valve positioner detection business. This business consists of tests such as the air source pressure change impact test, the ambient temperature change impact test, the mechanical vibration impact test, and the electrostatic discharge anti-interference test. In this business window, you can view the equipment data during these tests.
[0088] Data display uses a variety of graphics, images, and charts. In addition to statistical graphics that display statistical analysis data, it also has the ability to display relational data and hierarchical data charts, with a variety of data display capabilities.
[0089] Self-optimization is achieved through deep learning. Relying on continuously aggregated data and the digital twin model platform, multi-dimensional data analysis, machine learning, neural networks, deep learning and other algorithms or technologies are used to calculate optimization plans, prediction and warning models, emergency command strategies, etc. to achieve self-optimization.
[0090] Taking the common failure forms of industrial valves as an example, the vibration signal of the valve body is monitored by a vibration sensor (faults such as valve body wear and cracks will gradually accumulate and become the excitation source of vibration. The valve action cycle of these excitation sources is periodic and manifested in the vibration signal), the pressure sensor monitors the pressure before and after the valve, the flow sensor monitors the flow before and after the valve, and the displacement sensor monitors the valve stem position; the acquired signal is mathematically processed, and the monitoring data is processed and analyzed using discrete and continuous wavelet transforms combined with genetic algorithms to obtain algorithm models of industrial valves under normal working conditions and fault conditions, respectively. The development trend of the valve status is predicted based on the obtained model, so as to make relevant operations, controls and other temporary maintenance or shutdown repair decisions.
[0091] Relying on the deep learning conclusions such as optimization plans, early warning models, and command strategies obtained from the above-mentioned digital space, operations such as remote control of sensors and optimization and correction of information system functions are realized in the digital twin model platform, which acts reversely on the equipment operating in the physical world to complete dynamic decision-making, rapid response, and intelligent control.
[0092] The pressure difference before and after the valve is often determined by the fluid channel formed by the valve opening (the displacement L of the valve core, which is related to the valve position). The smaller the opening, the smaller the relative opening, and the greater the pressure difference before and after the valve; the larger the opening, the greater the relative opening, and the smaller the pressure difference before and after the valve. For a working regulating valve, when the valve opening changes, not only the flow rate changes, but also the pressure difference before and after the valve changes. The flow rate through the regulating valve is not only related to the valve opening, but also to the pressure difference before and after the valve.
[0093] The formula for calculating valve torque is: half the square of the valve diameter × 3.14 to get the area of the valve plate, multiply it by the pressure (i.e. the valve working pressure) to get the static pressure borne by the shaft, multiply it by the friction coefficient (check the table, the friction coefficient of general steel is 0.1, and the friction coefficient of steel to rubber is 0.15), multiply it by the shaft diameter and divide it by 1000 to get the valve torque number, the unit is Newton meter, the reference safety value of electric devices and pneumatic actuators is 1.5 times the valve torque.
[0094] Backstage management:
[0095] Backend management consists of user management, historical data management, log management, alarm and integrated operation and maintenance management. User management includes authority management and equipment management. Authority management implements the configuration of authority for roles, and different roles have different authorities. Equipment management is used to manage valve detection equipment and other equipment, and determine whether the equipment is displayed in the visual interface. Historical data management can view the equipment operation data within a fixed time period and export the data. Log management can view user operation logs, execution logs, login logs and other management in real time. Alarm and integrated operation and maintenance management sets alarm strategies to the business engine through the Internet of Things. The system monitors the operating status in real time and sends abnormal alarms to the corresponding alarm receivers, and generates work orders through alarm event linkage to notify the operation and maintenance personnel to deal with them in time. A full closed-loop management of real-time monitoring of abnormalities-initiating and handling of alarm work orders-continuous monitoring is formed.
[0096] Taking valve status monitoring as an example, the first thing that digital twins need to do is valve synchronization, that is, based on the real valve application industrial site, a virtual production line is built, and 3D modeling is performed on each device on the real production line and the virtual scene. The built model is placed in the virtual scene to achieve a one-to-one correspondence between the real production line and the virtual production line, and then data synchronization is performed. The real valve is driven by PLC to enable the valve to perform some predetermined actions. By collecting PLC data to drive the corresponding valve model in the virtual environment to perform the same predetermined actions, the real valve and the virtual valve are linked in real time. Through these, real-time monitoring of valves in the industrial site can be realized. The monitoring personnel only need to sit in front of the computer in the monitoring room to understand the working status of the industrial site in real time, thereby improving the monitoring and measurement efficiency, and then improving productivity.
[0097] Taking the valve opening control system as an example, the control system is mainly composed of a position transmitter, an emitter follower, an A / D converter, a controller, a motor driver, a motor, a reducer and an adjustment circuit. Assume that the controller here is a PLC.
[0098] First, add the MQTT industrial gateway to the PLC, and connect the PLC to the industrial gateway through the network port or serial port. Then the MQTT gateway will parse the PLC data and perform protocol analysis. After the data is standardized, the MQTT protocol is used as the uplink access protocol. After the data is obtained, it is processed and packaged, and then sent to the database of the digital twin platform.
[0099] Through the twin models of valves, motors, etc. in the visualization interface, the valve opening, motor speed, etc. are observed. Data such as torque, valve position, and pressure are continuously collected, and algorithms or technologies such as multidimensional data analysis, machine learning, neural networks, and deep learning are used to calculate and obtain optimization solutions. Relying on the optimization solution obtained in the above digital space, remote control of sensors and motors is realized in the digital twin model platform, which acts in reverse on the equipment operating in the physical world to complete dynamic decision-making, rapid response, and intelligent control.
[0100] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.
[0101] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.
Claims
1. A digital twin visualization method for valve monitoring, characterized in that: The following steps are involved: Collect the real production line equipment data of the factory; Build a visual 3D digital twin model of the factory's real production line equipment based on the factory's real production line equipment data; Using the visualized three-dimensional digital twin model to complete a simulation production test, and predicting the results of the simulation test based on a deep learning algorithm; The predicted data is fed back to the factory's actual production line equipment through the data interface module.
2. The digital twin visualization method for valve monitoring according to claim 1, characterized in that: It also includes initiating a data anomaly alarm when the collected real production line equipment data is abnormal.
3. A digital twin visualization platform for valve monitoring, characterized in that: include: Data interface module, digital twin visualization module and business window module, Wherein, the data interface module is used to collect the real production line equipment data of the factory, and the real production line equipment data of the factory includes the factory valve detection equipment data and the factory valve operation equipment data; The digital twin visualization module includes a twin model and a twin panel; the twin model is used to construct a visualized three-dimensional digital twin model corresponding to the real production line equipment of the factory based on the real production line equipment data of the factory; the twin panel is used to display the basic information of the twin model, and the basic information of the twin model includes the twin name, twin ID and twin model; The business window module is used to complete the simulation production test using the twin model, and predict the results of the simulation test based on the deep learning algorithm, and feed back the predicted data to the factory's real production line equipment through the data interface module.
4. The digital twin visualization platform for valve monitoring according to claim 3, characterized in that: The data interface module uses the MQTT gateway to transmit the collected factory real production line equipment data to the digital twin visualization module.
5. The digital twin visualization platform for valve monitoring according to claim 3, characterized in that: The factory valve detection equipment data includes sensor parameter data for detecting valve operation, and the valve operation equipment data includes valve opening and closing state data, valve front and rear pressure data, valve temperature data, fluid flow data through the valve, valve vibration data, and valve power consumption data measured based on sensors.
6. The digital twin visualization platform for valve monitoring according to claim 3, characterized in that: The simulation production test includes a valve opening and flow test, a valve opening and valve core position test, a valve flow and valve core pressure relationship test, and a valve positioner valve core stroke test.
7. The digital twin visualization platform for valve monitoring according to claim 3, characterized in that: The business window module predicts the results of the simulation test based on the deep learning algorithm, specifically including the following steps: Collect valve operation equipment data and perform preprocessing, wherein the preprocessing includes data denoising, filling missing values, and data normalization; Divide the preprocessed valve operation equipment data into a training set, a validation set, and a test set; Build a convolutional neural network model and use the training set to train the convolutional neural network model; Use the validation set and test set to evaluate the trained convolutional neural network model, and perform hyperparameter tuning or model structure adjustment based on the evaluation results to obtain a trained convolutional neural network model; The pre-optimized valve operation equipment data is input into the trained convolutional neural network model to obtain simulation test prediction data.
8. The digital twin visualization platform for valve monitoring according to claim 3, characterized in that: It is characterized in that It also includes a background management module.
9. The digital twin visualization platform for valve monitoring according to claim 8, characterized in that: The backend management module includes a user management submodule, a device management submodule, a historical data management submodule, and a log management submodule; The user management submodule is used to configure permissions for users, and different users have different permissions; The device management submodule is used to manage the three-dimensional digital twin model in the digital twin visualization module and determine whether the three-dimensional digital twin model is displayed; The historical data management submodule is used to view the equipment operation data within a set time period; The log management submodule is used to view user operation logs, user execution logs and user login logs in real time.
10. A digital twin visualization device for valve monitoring, characterized in that: The device includes a computer program, which, when executed, implements the digital twin visualization method for valve monitoring described in any one of claims 1-2.
Citation Information
Cited By
Method and system for collecting and transmitting oil casing pressure data of oil field water injection well
CN120007222A
A method and system for collecting and transmitting casing pressure and tubing pressure data of an oilfield water injection well
CN120007222B
Method for dynamically optimizing and adjusting control parameters of intelligent valve positioner
CN120595692A
Valve vibration impact test method and system based on digital twinning
CN120778322A
