An intelligent butterfly valve digital twin system and method based on calculation and measurement fusion

Abstract

The present invention provides a digital twin system and method for intelligent butterfly valves based on calculation and measurement fusion, which belongs to the field of digital twin technology. The system includes a physical entity module, an information interaction module, a twin construction module, and a visualization module. The method is as follows: first, a high-fidelity three-dimensional geometric model of the butterfly valve is constructed, and a fluid-solid coupling simulation is performed; second, the real state information of the butterfly valve is transmitted to the intelligent butterfly valve digital twin system in real time to obtain the motion state data of the butterfly valve; third, a prediction agent model is constructed based on the structural performance data; finally, a high-fidelity butterfly valve virtual model is constructed, and the real-time motion state data of the butterfly valve is used as input to drive the butterfly valve virtual model, and the visualization of the intelligent butterfly valve digital twin system is completed. The present invention can realize accurate mapping of the operating status of the butterfly valve and full-factor monitoring, ensure the integrity, real-time and high-fidelity of the digital twin data through calculation and measurement fusion, and improve the safety and control accuracy of the pipeline network system.

Description

Technical Field

[0001] The present invention belongs to the field of digital twin technology and relates to an intelligent butterfly valve digital twin system and method based on calculation and measurement fusion. Background Art

[0002] Butterfly valves are core control components in the fluid control system of hydropower project pipelines, and their operational performance directly determines the operational safety and reliability of the pipeline system. The valve body is connected to the pipeline at both ends via flanges, forming a complete flow path. The actuator drives the valve stem, rotating the valve disc to a specific position, enabling precise control of the fluid flow within the pipeline and ensuring the functionality of the pipeline network. However, during valve design and actual operation, the hydrodynamic torque coefficient is subject to uncertainty and load variations, resulting in significant interference with the valve's actuation and control process, posing a significant safety hazard. Furthermore, as the butterfly valve's opening and closing cycles accumulate, the valve stem, as the core force transmission component, is subjected to a combination of alternating stresses, mechanical shock, and vibration loads. This can cause fatigue damage to its internal structure, leading to failures such as fracture and deformation. As a critical component in the energy transmission path, the valve stem's service reliability is directly linked to the valve system's ability to achieve safe, reliable, and precisely controlled opening and closing operations.

[0003] To address these issues, domestic and international scholars have conducted extensive research and proposed several solutions. For example, a Chinese invention patent (application number 201910700513.6) provides a device and method for monitoring the sealing performance of butterfly valves based on changes in valve stem torque. This device uses several matching valve stem torque sensors and valve opening sensors to collect valve operating parameters. The data is transmitted to a computer terminal via a switch for storage. A software system determines whether to issue a warning by determining the changing trend of the torque data. A Chinese invention patent (application number 202510099357.8) provides a method and system for real-time abnormality monitoring of water inlet butterfly valves. Sensors located within the water butterfly valve acquire information such as flow rate per unit time, inner wall water pressure, and valve opening. By comparing this data with standard specifications, the valve's operating status is determined. While existing solutions can monitor the safe and healthy operation of butterfly valves, their reliance on data from a single sensor prevents real-time monitoring and early warning of the valve's overall structural performance.

[0004] Digital twin technology constructs a digital model based on a physical entity in a virtual space and deeply integrates it with real-time data. Using virtual simulation, analysis, and prediction, it dynamically updates the digital twin's motion state, performance characteristics, reliability, and other information, enabling virtual and real-world interaction. This technology can be applied throughout all stages of a product's lifecycle, from design to manufacturing and maintenance. Digital twin technology enables real-time monitoring of the structural performance and operating status of butterfly valves, as well as reliability analysis of key components such as the valve stem, providing a reference for improving and optimizing the valve structure. Furthermore, accurate data on the valve's motion can be obtained, providing data support for further motion control. Summary of the Invention

[0005] In response to the problems existing in the existing technology, the present invention provides an intelligent butterfly valve digital twin system and method based on calculation and measurement fusion, which integrates the motion state data collected by sensors in real time with the structural performance data based on mechanisms and statistical models to construct a multi-dimensional data collaborative system, ensuring the integrity, timeliness and high fidelity of the digital twin data, thereby completing all-round monitoring and precise mapping of the structural performance and operating parameters of the butterfly valve.

[0006] To achieve the above object, the technical solution adopted by the present invention is:

[0007] A digital twin system for intelligent butterfly valves based on computational and measurement fusion, comprising a physical entity module, an information interaction module, a twin construction module, and a visualization module; specifically:

[0008] The physical entity module consists of a butterfly valve physical entity, along with torque and pressure sensors deployed at key locations on the valve stem. A high-fidelity 3D geometric model of the butterfly valve is constructed using the physical entity. Fluid-structure interaction simulation is then performed on the 3D geometric model to generate structural performance data based on both mechanistic and statistical models.

[0009] The information exchange module collects real-time butterfly valve status information via torque sensors and pressure sensors deployed at key locations, such as the valve stem. This information is transmitted to the intelligent butterfly valve digital twin system via Modbus TCP and Modbus RTU communication protocols. The system then processes this information to generate the valve's motion status data.

[0010] The twin construction module consists of a flow field prediction proxy model, a structural stress prediction proxy model, a structural deformation prediction proxy model, and a valve stem hydrodynamic torque prediction proxy model. By integrating the structural data, flow field data, structural stress data, structural deformation data, and valve stem hydrodynamic torque data from the mechanism- and statistical model-based structural performance data in the physical entity module with the motion state data acquired and processed in real time, the flow field prediction proxy model, the structural stress prediction proxy model, the structural deformation prediction proxy model, and the valve stem hydrodynamic torque prediction proxy model are constructed.

[0011] The visualization module consists of a high-fidelity butterfly valve virtual model and key parameters of the intelligent butterfly valve digital twin system.

[0012] A digital twin method for intelligent butterfly valves based on computational and measurement fusion is implemented based on the above-mentioned intelligent butterfly valve digital twin system and specifically includes the following steps:

[0013] The first step is to build a high-fidelity 3D geometric model of the butterfly valve based on the physical entity of the butterfly valve, and perform fluid-solid coupling simulation on the 3D geometric model of the butterfly valve.

[0014] In the second step, torque sensors and pressure sensors are deployed at key locations such as the butterfly valve stem. The real status information of the butterfly valve collected by the sensors is transmitted to the intelligent butterfly valve digital twin system in real time through the ModbusTCP and Modbus RTU communication protocols. The real status information is then processed to obtain the motion status data of the butterfly valve.

[0015] In the third step, based on the structural performance data based on the mechanism and statistical model in step 1.2, a flow field prediction agent model, a structural stress prediction agent model, a structural deformation prediction agent model and a valve stem hydrodynamic torque prediction agent model are constructed.

[0016] In the fourth step, in the visualization module, a high-fidelity virtual model of the butterfly valve based on the physical entity of the butterfly valve is constructed by building a "point-surface-volume" fusion 3D visualization modeling technology based on the Unity 3D engine. Computer graphics technology is then used to visualize the flow field prediction proxy model, structural stress prediction proxy model, structural deformation prediction proxy model, and valve stem dynamic torque prediction proxy model established in the third step on the high-fidelity butterfly valve virtual model. Finally, the real-time butterfly valve motion state data obtained in step 2.2 is used as input to drive the high-fidelity butterfly valve virtual model. At the same time, the butterfly valve motion state data is visualized on the intelligent butterfly valve digital twin system, completing the visualization of the intelligent butterfly valve digital twin system.

[0017] The beneficial effects of the present invention are:

[0018] (1) This invention addresses the problem of existing butterfly valve safety and health detection methods being highly dependent on a single sensor and unable to monitor structural performance and provide early warnings in real time. It proposes an intelligent butterfly valve digital twin system and method based on computational fusion. This invention achieves real-time computation of butterfly valve motion state data through multi-source data fusion, establishing a digital twin system that can dynamically update and visualize structural performance data such as butterfly valve structural stress, structural deformation, and valve stem hydrodynamic torque, thus overcoming the limitations of traditional monitoring methods.

[0019] (2) The present invention uses a steady-state approximation computational fluid dynamics steady-state calculation model in fluid-solid coupling simulation to analyze the changes in the fluid domain of the butterfly valve system and the structural performance of the valve body under different working conditions. At the same time, it ensures that the node construction order under different working conditions remains unchanged during meshing, providing a data basis for using structural data to construct a high-fidelity butterfly valve virtual model based on the physical entity of the butterfly valve. The real-time motion state data of the butterfly valve obtained by the torque sensor and pressure sensor at the key parts such as the valve stem of the butterfly valve drives the established flow field prediction agent model, structural stress prediction agent model, structural deformation prediction agent model and valve stem dynamic water torque prediction agent model to be integrated. Not only can the structural performance data of the entire valve be calculated and visualized in real time, but the data obtained by the sensor can also be used to predict the reliability analysis of the valve and monitor and warn the overall performance of the valve. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 This is the overall flow chart of the intelligent butterfly valve digital twin system provided by the present invention.

[0021] Figure 2 This is a system framework diagram of the intelligent butterfly valve digital twin system provided by the present invention. DETAILED DESCRIPTION

[0022] The specific implementation of the present invention is further described below in conjunction with the accompanying drawings and specific technical solutions.

[0023] The present invention provides an intelligent butterfly valve digital twin system and method based on calculation and measurement fusion, and its construction method is as follows: Figure 1 As shown in the figure, the motion state data collected in real time by sensors is integrated with the calculation data based on mechanism and statistical models to build a multi-dimensional data collaboration system. This ensures the integrity, timeliness, and high fidelity of the digital twin data, thereby completing the comprehensive monitoring and precise mapping of the butterfly valve's structural performance and operating parameters.

[0024] This embodiment provides an intelligent butterfly valve digital twin system based on calculation and measurement fusion, such as Figure 2 The intelligent butterfly valve digital twin system includes a physical entity module, an information interaction module, a twin construction module, and a visualization module; specifically:

[0025] The physical entity module consists of a butterfly valve physical entity, along with torque and pressure sensors deployed at key locations such as the valve stem. A high-fidelity 3D geometric model of the butterfly valve is constructed using the physical entity. Fluid-structure interaction simulation is then performed on the 3D geometric model to generate structural performance data based on both mechanistic and statistical models.

[0026] The information exchange module collects real-time butterfly valve status information via torque sensors and pressure sensors deployed at key locations, such as the valve stem. This information is transmitted to the intelligent butterfly valve digital twin system via Modbus TCP and Modbus RTU communication protocols. The system then processes this information to generate the valve's motion status data.

[0027] The twin construction module consists of a flow field prediction proxy model, a structural stress prediction proxy model, a structural deformation prediction proxy model, and a valve stem hydrodynamic torque prediction proxy model. By integrating the structural data, flow field data, structural stress data, structural deformation data, and valve stem hydrodynamic torque data from the mechanism- and statistical model-based structural performance data in the physical entity module with the motion state data acquired and processed in real time, the flow field prediction proxy model, the structural stress prediction proxy model, the structural deformation prediction proxy model, and the valve stem hydrodynamic torque prediction proxy model are constructed.

[0028] The visualization module consists of a high-fidelity butterfly valve virtual model and key parameters of the intelligent butterfly valve digital twin system. A high-fidelity butterfly valve virtual model based on the physical entity of the butterfly valve is constructed using a "point-surface-volume" 3D visualization modeling technique built on the Unity 3D engine. Computer graphics technology is then used to visualize the established flow field prediction proxy models, structural stress prediction proxy models, structural deformation prediction proxy models, and valve stem hydrodynamic torque prediction proxy models on the high-fidelity butterfly valve virtual model. Finally, the butterfly valve's motion state data is used as input to drive the high-fidelity butterfly valve virtual model, and the butterfly valve's motion state data is visualized on the intelligent butterfly valve digital twin system.

[0029] This embodiment provides a digital twin method for an intelligent butterfly valve based on calculation and measurement fusion, which is implemented based on the above-mentioned intelligent butterfly valve digital twin system. The specific steps are as follows:

[0030] The first step is to build a high-fidelity 3D geometric model of the butterfly valve based on its physical entity and perform fluid-structure coupling simulation on the 3D geometric model. Specifically:

[0031] Step 1.1: Construct a high-fidelity 3D geometric model of the butterfly valve based on its physical entity. Divide the flow path into three sections: upstream, valve, and downstream. Mesh the 3D geometric model of the butterfly valve using a steady-state computational fluid dynamics (CFD) model.

[0032] Step 1.2: Using a spatial sampling method, select valve opening and pressure across the disc as key design variables. A sample space is constructed from multiple operating conditions. Based on this sample space, a parametric simulation method is used to perform a fluid-structure interaction simulation on the meshed 3D butterfly valve model constructed in Step 1.1. This method yields structural performance data calculated based on both mechanistic and statistical models.

[0033] In the fluid-structure coupling simulation solution process, the turbulence model selection model, selecting the standard wall function; after setting the boundary conditions, the butterfly valve opening and the pressure before and after the disc in the sample space are used as input parameters for parametric simulation to obtain the structural performance data based on the mechanism and statistical model, specifically the structural data of the valve geometric model, flow field data, structural stress, structural deformation and valve stem hydrodynamic torque data.

[0034] The second step is to deploy torque sensors and pressure sensors at key locations such as the butterfly valve stem, and transmit the real status information of the butterfly valve collected by the sensors to the intelligent butterfly valve digital twin system in real time through ModbusTCP and Modbus RTU communication protocols. The real status information is then processed to obtain the motion status data of the butterfly valve.

[0035] Step 2.1: Deploy a grating torque sensor along the valve stem axis and a grating strain sensor in the bolt stress concentration area to achieve real-time calculation of valve opening. Deploy pressure sensors in the upstream and downstream areas of the pipeline to establish a differential pressure measurement loop and obtain the pressure before and after the valve disc.

[0036] Step 2.2: The valve opening and disc pressure data acquired in Step 2.1 are transmitted in real time to the intelligent butterfly valve digital twin system via Modbus TCP and Modbus RTU communication protocols. The Modbus TCP and Modbus RTU communication decoding programs process the actual status information (valve opening and disc pressure data) transmitted to the system to generate real-time butterfly valve motion status data.

[0037] The third step is to build a flow field prediction agent model, a structural stress prediction agent model, a structural deformation prediction agent model, and a valve stem hydrodynamic torque prediction agent model based on the structural performance data based on the mechanism and statistical model in step 1.2. Specifically:

[0038] Step 3.1: First, extract flow field node data from the structural performance data based on the mechanism and statistical model in Step 1.2. This data is then converted into a basis coefficient matrix and eigenvalues ​​using the Proper Orthogonal Decomposition (POD) order reduction method. Next, a flow field prediction proxy model is constructed using the eigenvalues ​​and basis coefficients prediction proxy models, using the butterfly valve opening and pressure before and after the disc in the sample space as input parameters and the flow field modal coefficients as output.

[0039] The specific form of the base coefficient prediction agent model is as follows:

[0040] (6)

[0041] in, Representative Basis function centers; Represents the basis function. In this embodiment, the Gaussian basis ; Represents the Euclidean distance between the predicted point and the center of the basis function; Represents weight; Represents input, represent The predicted value at .

[0042] In this embodiment, the order reduction method of the Proper Orthogonal Decomposition (POD) is specifically as follows:

[0043] First, construct a snapshot matrix using the flow field data obtained in step 1.2 fluid-structure coupling simulation:

[0044] (1)

[0045] in, is the snapshot matrix; N is the number of grid nodes; M is the number of samples in the sample space; is the flow field data of the i-th sample.

[0046] Secondly, the covariance matrix and eigenvalue decomposition of the input parameters in the sample space are calculated:

[0047] (2)

[0048] (3)

[0049] Where C is the covariance matrix of the snapshot matrix; The eigenvalues ​​are arranged in descending order, representing the energy contribution of each mode to the flow field; is the feature vector.

[0050] Finally, according to the energy proportion criterion, the first k dominant POD mode basis functions are selected :

[0051] (4)

[0052] in, represents the kth modal basis function; represents the kth eigenvalue; represents the kth eigenvector. The energy proportion criterion is:

[0053] (5)

[0054] Among them, represents the percentage of the energy corresponding to the first k POD modes to the total energy; k represents the kth POD mode; represents the i-th eigenvalue.

[0055] Step 3.2: Using the valve opening and the differential pressure across the disc in the sample space of step 1.2 as input, and the structural stress data from the structural performance data based on the mechanism and statistical model as output, a structural stress prediction proxy model is constructed. This proxy model can calculate the structural stress of the valve geometric model for any input (valve opening, differential pressure across the disc within a set range).

[0056] Step 3.3: Take the valve opening and the pressure difference before and after the valve disc from step 1.2 as input, and the structural deformation data in the structural performance data based on the mechanism and statistical model as output, and construct a structural deformation prediction agent model. The structural deformation of the valve geometric model under arbitrary input can be calculated through the structural stress prediction agent model.

[0057] Step 3.4: Take the valve opening and the pressure difference before and after the valve disc from step 1.2 as input, and the valve stem dynamic torque data in the structural performance data based on the mechanism and statistical model as output, and construct a valve stem dynamic torque prediction agent model. The valve stem dynamic torque prediction agent model can calculate the valve stem dynamic torque of the valve under any input.

[0058] In the fourth step, a high-fidelity virtual model of the butterfly valve based on the physical entity of the butterfly valve is constructed using a 3D visualization modeling technique that integrates "point-surface-volume" technology based on the Unity 3D engine. Computer graphics technology is then used to visualize the predictive agent model established in the third step on the high-fidelity virtual model of the butterfly valve. Finally, the visualization of the intelligent butterfly valve digital twin system is completed by driving the high-fidelity virtual model with the real-time butterfly valve motion data obtained in step 2.2. Specifically:

[0059] Step 4.1: By building a "point-surface-body" fusion 3D visualization modeling technology based on the Unity 3D engine, a high-fidelity butterfly valve virtual model based on the physical entity of the butterfly valve is constructed using the structural data in the structural performance data based on the mechanism and statistical model in step 1.2.

[0060] Step 4.2: Use computer graphics technology to visualize the established flow field prediction proxy model, structural stress prediction proxy model, structural deformation prediction proxy model, and valve stem hydrodynamic torque prediction proxy model on the high-fidelity butterfly valve virtual model.

[0061] Step 4.3: By using the real-time butterfly valve motion state data obtained in step 2.2 as the input of the flow field prediction agent model, structural stress prediction agent model, structural deformation prediction agent model and valve stem hydrodynamic torque prediction agent model, the high-fidelity butterfly valve virtual model is driven in real time, thereby realizing the state mapping between the butterfly valve physical entity and the high-fidelity butterfly valve virtual model, and completing the construction of the intelligent butterfly valve digital twin system.

[0062] The above-described embodiments merely express the implementation methods of the present invention, but should not be understood as limiting the scope of the present invention. It should be pointed out that those skilled in the art can make several modifications and improvements without departing from the concept of the present invention, which all fall within the scope of protection of the present invention.

Claims

1. A digital twin method for intelligent butterfly valves based on calculation and measurement fusion, characterized in that: The intelligent butterfly valve digital twin method The following steps are involved: The first step is to build a high-fidelity 3D geometric model of the butterfly valve based on its physical entity and perform fluid-structure coupling simulation on the 3D geometric model. Specifically: Step 1.1: Construct a high-fidelity 3D geometric model of the butterfly valve based on its physical entity. Divide the flow path into three sections: upstream, valve, and downstream. Mesh the 3D geometric model using a steady-state computational fluid dynamics (CFD) model. Step 1.2: Using a spatial sampling method, select valve opening and pressure before and after the disc as key design variables, and construct a sample space from multiple operating conditions. Based on this sample space, use a parametric simulation method to perform fluid-structure interaction simulation on the meshed 3D butterfly valve model constructed in Step 1.1, obtaining structural performance data calculated based on the mechanism and statistical model. The second step is to deploy torque sensors and pressure sensors at key locations of the butterfly valve. The real status information of the butterfly valve collected by the sensors is transmitted to the intelligent butterfly valve digital twin system in real time. The real status information is then processed to obtain the motion status data of the butterfly valve. Specifically: Step 2.1: The key part of the butterfly valve is the valve stem. By deploying a grating torque sensor along the valve stem axis and a grating strain sensor in the bolt stress concentration area, real-time calculation of the valve opening is achieved. Pressure sensors are deployed in the upstream and downstream areas of the pipeline to establish a pressure differential measurement loop to obtain the pressure before and after the valve disc. Step 2.2: The valve opening and the pressure data before and after the valve disc obtained in step 2.1 are transmitted to the intelligent butterfly valve digital twin system in real time, and the data is processed to obtain the real-time motion state data of the butterfly valve; The third step is to build a flow field prediction agent model, a structural stress prediction agent model, a structural deformation prediction agent model, and a valve stem hydrodynamic moment prediction agent model based on the structural performance data based on the mechanism and statistical model in step 1.2; Fourth, in the visualization module, a high-fidelity butterfly valve virtual model based on the physical entity of the butterfly valve is constructed by building a "point-surface-volume" fusion three-dimensional visualization modeling technology based on the Unity 3D engine. Computer graphics technology is used to visualize the flow field prediction agent model, structural stress prediction agent model, structural deformation prediction agent model, and valve stem dynamic torque prediction agent model established in the third step on the high-fidelity butterfly valve virtual model. Finally, the real-time butterfly valve motion state data obtained in step 2.2 is used as input to drive the high-fidelity butterfly valve virtual model, and the butterfly valve motion state data is visualized on the intelligent butterfly valve digital twin system.

2. The intelligent butterfly valve digital twin method based on calculation and measurement fusion according to claim 1 is characterized in that: In step 1.2, during the fluid-solid coupling simulation solution, the turbulence model selection model, select the standard wall function; after setting the boundary conditions, use the butterfly valve opening and the pressure before and after the valve disc in the sample space as input parameters for parametric simulation to obtain the structural performance data based on the mechanism and statistical model.

3. The intelligent butterfly valve digital twin method based on calculation and measurement fusion according to claim 2 is characterized in that: The structural performance data includes the structural data of the valve geometric model, flow field data, structural stress, structural deformation and valve stem hydrodynamic torque data.

4. The intelligent butterfly valve digital twin method based on calculation and measurement fusion according to claim 3 is characterized in that: The third step is specifically as follows: Step 3.1: First, extract the flow field node data from the structural performance data in step 1.

2. This data is then converted into a basis coefficient matrix and eigenvalues ​​using the proper orthogonal decomposition (POD) order reduction method. Secondly, using the butterfly valve opening and the pressure before and after the disc in the sample space as input parameters and the flow field modal coefficients as output, a flow field prediction proxy model is constructed using the eigenvalues ​​and basis coefficients prediction proxy models. Step 3.2: Using the valve opening and the pressure difference across the disc in the sample space of step 1.2 as input, and the structural stress data from the structural performance data based on the mechanism and statistical model as output, a structural stress prediction proxy model is constructed. This proxy model can calculate the structural stress of the valve geometry model for any input. Step 3.3: Using the valve opening and the pressure difference across the disc from step 1.2 as inputs and the structural deformation data from the structural performance data based on the mechanism and statistical model as outputs, a structural deformation prediction proxy model is constructed. This structural stress prediction proxy model can calculate the structural deformation of the valve geometry model for any input. Step 3.4: Take the valve opening and the pressure difference before and after the valve disc from step 1.2 as input, and the valve stem dynamic torque data in the structural performance data based on the mechanism and statistical model as output, and construct a valve stem dynamic torque prediction agent model. The valve stem dynamic torque prediction agent model can calculate the valve stem dynamic torque of the valve under any input.

5. The intelligent butterfly valve digital twin method based on calculation and measurement fusion according to claim 4 is characterized in that: In step 3.1, the order reduction method of the proper orthogonal decomposition POD is specifically as follows: First, construct a snapshot matrix using the flow field data obtained in step 1.2 fluid-structure coupling simulation: （1）； in, is the snapshot matrix; N is the number of grid nodes; M is the number of samples in the sample space; is the flow field data of the i-th sample; Secondly, the covariance matrix and eigenvalue decomposition of the input parameters in the sample space are calculated: （2）； （3）； Where C is the covariance matrix of the snapshot matrix; The eigenvalues ​​are arranged in descending order, representing the energy contribution of each mode to the flow field; is the eigenvector; Finally, according to the energy proportion criterion, the first k dominant POD mode basis functions are selected : （4）； in, represents the kth modal basis function; represents the kth eigenvalue; represents the kth eigenvector.

6. The intelligent butterfly valve digital twin method based on calculation and measurement fusion according to claim 5 is characterized in that: The energy proportion criterion is: （5）； in, Indicates the percentage of energy corresponding to the first k POD modes to the total energy; k represents the kth POD mode; represents the i-th eigenvalue.

7. The intelligent butterfly valve digital twin method based on calculation and measurement fusion according to claim 4 is characterized in that: The fourth step is specifically as follows: Step 4.1: Using the Unity 3D engine to build a "point-surface-volume" fusion 3D visualization modeling technology, a high-fidelity butterfly valve virtual model based on the physical entity of the butterfly valve is constructed using the structural data from the structural performance data based on the mechanism and statistical model in step 1.

2. Step 4.2: Use computer graphics technology to visualize the established flow field prediction proxy model, structural stress prediction proxy model, structural deformation prediction proxy model, and valve stem hydrodynamic torque prediction proxy model on a high-fidelity butterfly valve virtual model; Step 4.3: By using the real-time butterfly valve motion state data obtained in step 2.2 as the input of the flow field prediction agent model, structural stress prediction agent model, structural deformation prediction agent model and valve stem hydrodynamic torque prediction agent model, the high-fidelity butterfly valve virtual model is driven in real time, thereby realizing the state mapping between the butterfly valve physical entity and the high-fidelity butterfly valve virtual model, and completing the construction of the intelligent butterfly valve digital twin system.

8. An intelligent butterfly valve digital twin system based on calculation and measurement fusion, characterized in that: The intelligent butterfly valve digital twin method based on calculation and measurement fusion according to any one of claims 1 to 7 is implemented by the intelligent butterfly valve digital twin system, and the intelligent butterfly valve digital twin system includes a physical entity module, an information interaction module, a twin construction module, and a visualization module; specifically: The physical entity module consists of a butterfly valve physical entity, torque sensors and pressure sensors deployed at key locations of the butterfly valve. A high-fidelity three-dimensional geometric model of the butterfly valve is constructed using the butterfly valve physical entity, and a fluid-structure coupling simulation is performed on the three-dimensional geometric model of the butterfly valve to obtain structural performance data based on mechanism and statistical models. The information interaction module collects the real state information of the butterfly valve in real time through torque sensors and pressure sensors deployed at key parts of the butterfly valve; transmits the collected real state information of the butterfly valve to the intelligent butterfly valve digital twin system in real time, and processes the real state information to obtain the motion state data of the butterfly valve; The twin construction module consists of a flow field prediction agent model, a structural stress prediction agent model, a structural deformation prediction agent model and a valve stem hydrodynamic torque prediction agent model; by fusing the structural data, flow field data, structural stress, structural deformation and valve stem hydrodynamic torque data in the structural performance data based on the mechanism and statistical model in the physical entity module with the motion state data obtained by real-time collection and processing, the flow field prediction agent model, the structural stress prediction agent model, the structural deformation prediction agent model and the valve stem hydrodynamic torque prediction agent model are constructed respectively; The visualization module consists of a high-fidelity butterfly valve virtual model and key parameters of the intelligent butterfly valve digital twin system.

Images