Power equipment adjusting device virtual-real cooperation platform construction method based on digital twinning

Abstract

The invention relates to the technical field of digital twinning of power equipment, in particular to a virtual-real collaboration platform construction method for a power equipment adjusting device based on digital twinning, which comprises the following steps: S1, constructing a high-fidelity physical platform; s2, establishing a high-precision digital model based on multi-field unified modeling; s3, constructing a visual digital terminal; and S4, establishing data communication. According to the design and construction method of the digital virtual-real collaborative platform, a high-fidelity physical platform, a high-precision digital model for multi-field unified modeling and a visual digital terminal are used as main bodies. Compared with the prior art, the method has the advantages that the physical state mapping capability and the twinborn model analysis capability are improved, the use and operation convenience is enhanced, and a new technical architecture is provided for fusion driving of virtual and real models of ship power equipment.

Description

Technical Field

[0002] The present invention relates to the technical field of digital twins of power equipment, and in particular to a method for constructing a virtual-reality collaborative platform of a power equipment regulating device based on digital twins. Background Art

[0003] As the core of the ship, the performance of the power system directly determines the maneuverability and vitality of the ship. Since ships are far away from land when sailing, they have the characteristics of long voyage time and high intensity, unpredictable marine environment, and difficult replenishment and guarantee. In addition, they are affected by high salt fog in the ocean, changes in sea conditions, etc., and the operating environment is harsh. In addition, the pipeline equipment of the ship power system is numerous and extremely complex. The equipment is coupled and operates, and failures occur frequently. Therefore, its operation and management requirements are much higher than similar systems. The high-energy medium control valve is the "throat" of the power system, which directly affects the power output and safe and reliable operation of the power equipment. The hydraulic system is the "nerve" for realizing precise control of the high-energy medium control valve and other key equipment, and plays a decisive role in the operation management and performance of the power equipment. However, in the current process of equipment digital transformation and construction, a set of complete and feasible digital empowerment theoretical architecture has not yet been formed for it.

[0004] The key to promoting the digital construction of power equipment is how to fully combine real-time data and equipment operation conditions on the basis of building high-fidelity digital models to establish a dynamic process of virtual-real symbiosis. Digital twin technology is an important way to integrate the physical world and the information world. By digitally modeling actual physical equipment, depicting its real characteristics in the digital space, and combining mechanism modeling with data-driven methods to establish its behavior rule model, the full-factor, multi-dimensional display of the physical space is achieved through virtual-real interaction, dynamic parameter correction, decision analysis and other methods. The ultimate goal is to establish a real-time, efficient, multi-task intelligent collaborative workspace.

[0005] The development trend of digitalization and intelligentization of equipment has put forward higher requirements for the operation and maintenance of new ship nuclear power equipment. The concept of digital twin of power equipment has been widely used in aerospace, shipbuilding, vehicles and other fields. However, how to build a high-fidelity simulation model for ship power equipment, especially the regulating device, and realize the closed-loop interaction between virtual and real has not been studied in depth.

[0006] The information disclosed in this background technology section is only intended to deepen the understanding of the overall background technology of the present invention, and should not be regarded as acknowledging or suggesting in any form that the information constitutes the prior art already known to those skilled in the art. Summary of the invention

[0007] The purpose of the present invention is to provide a method for constructing a virtual-reality collaborative platform for a power equipment regulating device based on digital twins to solve the technical problems existing in the prior art.

[0008] In order to achieve the above object, the present invention adopts the following technical solutions:

[0009] The present invention provides a method for constructing a virtual-reality collaborative platform of a power equipment regulating device based on digital twin, which comprises:

[0010] S1. Build a high-fidelity physical platform;

[0011] S2. Establish high-precision digital models based on unified modeling in multiple fields;

[0012] S3, build a visual digital terminal;

[0013] S4. Establish data communication.

[0014] Preferably, step S1 includes: aiming at the characteristics of high power density integrated design, high response and high precision of the ship regulating device, designing a scaled structure diagram that highly restores the actual equipment through the similarity criterion, selecting equipment with reference to the actual and design requirements of the ship, arranging high-precision sensing equipment at key monitoring positions, realizing all-round and refined collection of multi-dimensional data, so as to complete the construction of a high-fidelity physical platform.

[0015] Preferably, step S2 includes: based on the high-fidelity physical platform built in step S1, a system simulation model is built in the unified modeling software MWorks with Modelica multi-domain unified modeling language as the technical basis and parameter analysis is performed on it, a digital model integrating the hydraulic system and the high-energy medium regulation loop is constructed, and real-time operation data is collected based on the high-precision sensing equipment arranged in step S1 to guide the dynamic correction of the simulation model, so as to realize the construction of a dual-driven model driven by mechanism and data.

[0016] Preferably, step S2 further comprises: using the monitoring parameters of the high-precision sensing device and the rated parameters of the device in step S1 for model parameter calibration, wherein the model parameter calibration comprises the following features:

[0017] (1) Supports the association of model variables with measurement data and enables online parameter adjustment by minimizing data differences;

[0018] (2) MWorks automatically calls the solver and outputs a dynamic result comparison curve;

[0019] (3) Parameter correlation analysis, calibration sensitivity check of related parameters.

[0020] Preferably, step S3 includes: developing visualization terminal functions based on the three-dimensional modeling tool Maya and the three-dimensional development tool Unit3D, wherein the visualization terminal realizes simulation control management of the professional knowledge base and human-computer interaction.

[0021] Preferably, the professional knowledge base includes: a 3D model module, a graphic module and a video module; the 3D model module has a structure display submodule and a working principle submodule; the graphic module has an information introduction submodule and a fault list submodule; the video module has a principle animation submodule and a fault animation submodule; the human-computer interaction includes: a simulation drive module and a control instruction module; the simulation drive module has a state monitoring submodule and a state drive submodule; the control instruction module has a parameter adjustment submodule and a instruction sending submodule.

[0022] Preferably, step S4 includes: a universal physical communication module based on the Modbus protocol to realize the communication between the simulation model and the physical hardware; a three-dimensional communication external function library based on the MDD library to realize the data communication between the simulation model and the three-dimensional visualization interface and the control interface.

[0023] The present invention provides a method for constructing a virtual-reality collaborative platform for a power equipment regulating device based on digital twins, which adopts key virtual-reality fusion technologies such as cross-scale high-fidelity modeling, domain-expanding optimization joint simulation, and full-scenario intelligent control strategy. The implementation method is as follows:

[0024] Cross-scale high-fidelity modeling At the system level, the visual interface maps the macroscopic operating status of the physical world, switches to the DCS interface, and realizes rapid control of equipment in a two-dimensional simplified form.

[0025] Cross-scale high-fidelity modeling At the device level, single device virtual disassembly and assembly achieves high-precision structural cognition, and driven by one-dimensional simulation data, solves the internal fluid flow characteristics and flow parameters of a single device.

[0026] Cross-scale high-fidelity modeling supports the conversion of system-level macroscopic state display to local details at the meter to millimeter level at the part level. Based on CFD numerical results, it defines the physical field data mapping matrix conversion relationship of key position nodes.

[0027] The domain-expanding optimization joint simulation keeps Modelica's continuous discrete mixed differential algebraic equation solving process unchanged, adds fluid, mechanical, control and other computing modules to the discrete event computing process, completes the computing tasks within a specified time period, and extracts the computing boundary values ​​to define the data exchange protocol.

[0028] The full-scenario intelligent control strategy is based on the virtual-to-real strategy of reinforcement learning, with policy gradient and value evaluation as the core methods for policy learning and value function estimation.

[0029] Preferably, in the method construction module, the method for constructing the policy gradient and value evaluation includes:

[0030] ;

[0031] In the formula, It’s a strategy In the performance function J about the parameters The gradient of Indicated in strategy Downsampled state-action pairs expectations, Indicates in status Take action The logarithm of the probability with respect to the parameter The gradient of Yes Status Take action The state value function of Yes Status The state value function under .

[0032] Optionally, constructing the value assessment function based on reinforcement learning includes:

[0033] ;

[0034] Here, γ is the discount factor and τ(t+1) is the reward at time t+1.

[0035] By adopting the above technical solution, the present invention has the following beneficial effects:

[0036] The present invention forms a design and construction method for a digital virtual-real collaborative platform with a high-fidelity physical platform, a high-precision digital model with unified modeling in multiple fields, and a visualized digital terminal as the main body. Compared with the prior art, the present invention improves the physical state mapping capability and the twin model analysis capability, and enhances the convenience of operation, providing a new technical architecture for the fusion drive of virtual-real models of ship power equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0038] Figure 1 A schematic diagram of a method flow chart provided by an embodiment of the present invention;

[0039] Figure 2 A system architecture provided for an embodiment of the present invention;

[0040] Figure 3 The physical model and digital model construction process provided by the embodiment of the present invention;

[0041] Figure 4 The main functions of the visual digital terminal provided by the embodiment of the present invention;

[0042] Figure 5 A visual digital terminal operation monitoring interface provided by an embodiment of the present invention;

[0043] Figure 6 Demonstration of cross-scale modeling cases provided by embodiments of the present invention;

[0044] Figure 7 A multi-domain simulation solution architecture provided for an embodiment of the present invention;

[0045] Figure 8 A digital twin control strategy based on reinforcement learning is provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0046] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. 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.

[0047] The specific implementation of the present invention is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described here is only used to illustrate and explain the present invention, and is not used to limit the present invention.

[0048] Combination Figures 1 to 8 As shown, this embodiment provides a method for constructing a virtual-reality collaborative platform of a power equipment regulating device based on digital twins, which includes:

[0049] S1. Build a high-fidelity physical platform;

[0050] The equipment selection is carried out according to the actual and design requirements of the ship. The core equipment of the device, the high-energy medium regulating valve, consists of the valve body, connecting pipelines and hydraulic control equipment. In actual use, multiple equipment linkage is required, and the motion control of the regulating valve can be finally realized with the help of control signals and feedback data. When the equipment is running, first start the hydraulic oil pump 1 to establish the system pressure, open the electromagnetic reversing valve to connect the working oil circuit, and the working oil circuit can control the operation of three units. Unit 1 adjusts the opening of the high-energy medium regulating valve through the proportional valve, and units 2 and 3 respectively adjust the high-energy medium regulating valve and the fast-closing valve through the reversing valve. Unit 4 is a spare manual reversing module, which can be connected to the hydraulic pipeline for emergency use. In the pipeline part where the high-energy medium regulating valve is located, multiple types of high-energy media such as air, water, and steam can be introduced for valve performance testing. The target system is determined to include more than 50 types of key equipment models and parameters such as hydraulic oil tank, servo motor, loading pump, oil pump motor, safety valve, overflow valve, accumulator, proportional servo valve, reversing valve, high-energy medium regulating valve, gate valve, etc., and the physical system is built according to the design drawings. Compared with the actual system of the ship, some high-precision sensing equipment listed in Table 1 are arranged for real-time monitoring of relevant physical quantities at more than 20 key monitoring locations, such as lubricating oil pump outlet flow, lubricating oil pump outlet pressure, proportional valve rear pressure, return oil pipeline flow, high-energy medium regulating valve hydraulic cylinder inlet and outlet hydraulic oil flow, high-energy medium regulating valve inlet and outlet hydraulic oil pressure, high-energy medium regulating valve cylinder inlet and outlet pressure, high-energy medium circuit flow, high-energy medium circuit temperature, high-energy medium regulating valve inlet and outlet pressure, high-energy medium regulating valve valve cavity pressure, etc. The above monitoring parameters and equipment rated parameters will be used for model parameter calibration.

[0051] In this embodiment, the target object may be all power equipment auxiliary systems including but not limited to power equipment regulating devices, such as fuel systems, lubricating oil systems, control systems, etc., or the entire power system.

[0052] S2. Establish high-precision digital models based on unified modeling in multiple fields;

[0053] Based on the physical test platform built in step S1, a system simulation model was built in the unified modeling software MWorks with Modelica multi-domain unified modeling language as the technical foundation and parameter analysis was performed on it. A digital model integrating the hydraulic system and high-energy medium regulation loop was constructed. The model architecture is based on equipment parameters, structural composition, working logic, operating parameters, etc., and the physical entity behavior is comprehensively constructed through system modeling, multi-domain simulation, parameter calibration, etc. Through the principle of modular modeling, the system is decomposed from top to bottom into individual thermal equipment such as oil pumps, motors, overflow valves, reversing valves, pipelines, and regulating valves. Their digital mechanism models are established in turn according to physical equations, and physical quantity transfer is completed between multiple physical fields through different coupling mechanisms. Finally, the subsystem model is integrated into the system-level digital twin model through the model interface to realize bottom-up model construction. Based on the real-time operation data collected by the sensor equipment, the data is used as identifiable parameters of the simulation model after signal conversion processing, guiding the simulation model to be dynamically corrected according to the state of the physical world.

[0054] form Figure 2 The digital model construction method shown on the right includes basic-level models (interface model, medium physical property model, sensor model, boundary model) and equipment-level models (oil tank model, pipeline model, hydraulic valve model, oil pump model, regulating valve model, auxiliary equipment model, and mechanical component model).

[0055] In addition, based on a large number of online and offline parameters in step S1, the parameter values ​​are automatically adjusted within a certain range and simulated to minimize the error between the simulation results and the measured data. The main features of parameter calibration are as follows: (1) It supports the association of model variables with measured data and realizes online parameter adjustment by minimizing data differences; (2) MWorks automatically calls the solver and outputs a dynamic result comparison curve; (3) Parameter correlation analysis and calibration sensitivity check of associated parameters.

[0056] S3, build a visual digital terminal;

[0057] Based on step S1 and step S2, a real-time driving demonstration model of the construction module is constructed to realize the real-time demonstration of the structure, working state and operating parameters of the power equipment regulating device. The main contents are as follows:

[0058] Based on the physical entity established in step S1, through Maya related actual model modeling, 1:1 high-fidelity real prototype 3D modeling simulation based on industrial models can be achieved, and based on Unity3D, high-quality film-level 3D software interactive development can be achieved.

[0059] Through on-site surveying and mapping, and consulting hydraulic-related books and manuals, the dimensions of hydraulic components and hydraulic test benches were obtained. Then, a detailed 3D model was designed in a 1:1 ratio using Solidworks. The model was imported into 3DMax in STEP format to remove redundant surfaces and adjust the axis of the hydraulic components. Finally, the model was imported into Unity3D in FBX format.

[0060] In the visual digital terminal, the system's professional knowledge base and display modes include three modules: "3D model", "picture and text", and "video". The "3D model" can rotate, scale, translate, and disassemble the system and equipment to display dynamic behaviors such as mechanical movement and fluid flow. The "picture and text" module contains various types of graphic information such as operation manuals, technical documents, product specifications, principle pictures, etc. The "video" module provides video content of operation procedures and technical demonstrations, including troubleshooting and maintenance guidance videos. The knowledge base model includes but is not limited to the main technical parameters of the hydraulic system, the composition and working principle of the hydraulic system, and fault records.

[0061] The system backend provides the functions of editing, modifying and updating the text, image and video content of the knowledge base of different devices to maintain the validity and correctness of the knowledge base.

[0062] The human-computer interaction module is based on the framework of 3D models and 2D schematic diagrams, combined with MWorks simulation and controller communication, to monitor the status of physical objects, display drive, and control and manage physical objects through parameter panels.

[0063] S4, establish data communication;

[0064] In this embodiment, a three-dimensional communication external function library based on the MDD library is encapsulated, and the three-dimensional communication external function library belongs to the middle layer of the digital twin system, so as to realize data communication between the simulation model and the three-dimensional visualization interface and the control interface. The library adopts a model-driven development method, and automatically generates an interface for interaction between the three-dimensional interface and the control interface by defining the structure, attributes and behavior of the model hierarchy. The transmitted data may be real-time status, control parameters or simulation results, etc.

[0065] In this embodiment, a universal physical communication module based on the Modbus protocol is developed. The simulation model and the physical hardware data interface are arranged in an integrated data acquisition box. The actual equipment operation data and sensor equipment data are transmitted to the two-dimensional simulation and three-dimensional demonstration interface through the Modbus protocol. At the same time, the digital virtual-reality collaboration platform signal is comprehensively monitored and controlled in real time based on this module to achieve deep interaction of virtual-reality collaboration.

[0066] Furthermore, through UDP data communication, the mechanism model simulation data of MWorks is transmitted to the 3D visual end to drive the real-time status drive display of the hydraulic 3D scene.

[0067] Through UDP data communication, the visual digital terminal sends new status parameters to modify the device parameters in the digital space.

[0068] This embodiment provides a method for constructing a virtual-reality collaborative platform for a power equipment regulating device based on digital twins, which adopts key virtual-reality fusion technologies such as cross-scale high-fidelity modeling, domain-expanding optimization joint simulation, and full-scenario intelligent control strategy. The implementation method is as follows:

[0069] Cross-scale high-fidelity modeling At the system level, the visual interface maps the macroscopic operating status of the physical world, switches to the DCS interface, and realizes rapid control of equipment in a two-dimensional simplified form.

[0070] Cross-scale high-fidelity modeling At the device level, single device virtual disassembly and assembly achieves high-precision structural cognition, and driven by one-dimensional simulation data, solves the internal fluid flow characteristics and flow parameters of a single device.

[0071] Cross-scale high-fidelity modeling supports the conversion of system-level macroscopic state display to local details at the meter to millimeter level at the part level. Based on CFD numerical results, it defines the physical field data mapping matrix conversion relationship of key position nodes.

[0072] The domain-expanding optimization joint simulation keeps Modelica's continuous discrete mixed differential algebraic equation solving process unchanged, adds fluid, mechanical, control and other computing modules to the discrete event computing process, completes the computing tasks within a specified time period, and extracts the computing boundary values ​​to define the data exchange protocol.

[0073] The full-scenario intelligent control strategy is based on the virtual-to-real strategy of reinforcement learning, with policy gradient and value evaluation as the core methods for policy learning and value function estimation.

[0074] Preferably, in the method construction module, the method for constructing the policy gradient and value evaluation includes:

[0075] ;

[0076] In the formula, It’s a strategy In the performance function J about the parameters The gradient of Indicated in strategy Downsampled state-action pairs expectations, Indicates in status Take action The logarithm of the probability with respect to the parameter The gradient of Yes Status Take action The state value function of Yes Status The state value function under .

[0077] Optionally, constructing the value assessment function based on reinforcement learning includes:

[0078] ;

[0079] Here, γ is the discount factor and τ(t+1) is the reward at time t+1.

[0080] Preferably, by writing the above-mentioned intelligent algorithm logic structure into the Mworks simulation model, encapsulating it as an intelligent control module, integrating it into the digital virtual-reality collaborative platform, and developing a physical communication module based on the Modbus protocol, control decisions for different mission profiles can be realized, thereby effectively improving the adaptability and robustness of the system.

[0081] It can be seen that the present invention builds a highly integrated physical test platform for high-energy medium circuits and hydraulic control systems based on the similarity criterion, realizes high-fidelity restoration of actual equipment, and can meet the simulation test requirements under various working conditions. Through the MWorks construction module, the real-time driving demonstration model is developed, and a general communication transceiver module is developed to realize the virtual-real mapping and virtual-real control of the equipment structure, working status, and operating parameters of the regulating device, achieving the construction of a full range of digital twin models from the equipment level to the system level. In view of the difficulties in realizing the digital twin of the high-energy medium regulating valve, cross-scale high-fidelity modeling is proposed, and multi-scale coupling modeling and multi-scale operation demonstration at the system level, equipment level, and part level are realized. A joint simulation for domain expansion and optimization is proposed, combining the advantages of one-dimensional system characteristic simulation and high-precision simulation of three-dimensional CAE software to realize multi-domain joint simulation solution. A full-scenario intelligent control strategy is proposed to form intelligent control and operation and maintenance technologies for different task profiles.

[0082] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for constructing a virtual-reality collaborative platform for a power equipment regulating device based on digital twins, characterized in that: include: S1. Build a high-fidelity physical platform; S2. Establish high-precision digital models based on unified modeling in multiple fields; S3, build a visual digital terminal; S4. Establish data communication.

2. The method for constructing a virtual-reality collaborative platform for a power equipment regulating device based on digital twin according to claim 1 is characterized in that: Step S1 includes: In view of the high power density integrated design, high response and high precision of the ship's regulating device, the scaled structure diagram of the actual equipment is highly restored through the similarity criterion design, the equipment is selected with reference to the actual and design requirements of the ship, and high-precision sensing equipment is arranged at key monitoring positions to achieve all-round and refined collection of multi-dimensional data, so as to complete the construction of a high-fidelity physical platform.

3. The method for constructing a virtual-reality collaborative platform for a power equipment regulating device based on digital twin according to claim 1 is characterized in that: Step S2 includes: According to the high-fidelity physical platform built in step S1, a system simulation model was built and parameter analysis was performed in the unified modeling software MWorks with the Modelica multi-domain unified modeling language as the technical foundation, and a digital model integrating the hydraulic system and high-energy medium regulation loop was constructed. The real-time operation data was collected based on the high-precision sensing equipment arranged in step S1 to guide the dynamic correction of the simulation model, thus realizing the construction of a model driven by both mechanism and data.

4. The method for constructing a virtual-reality collaborative platform for a power equipment regulating device based on digital twin according to claim 3 is characterized in that: Step S2 also includes: The monitoring parameters of the high-precision sensing device and the rated parameters of the device in step S1 are used for model parameter calibration, and the model parameter calibration includes the following features: (1) Supports the association of model variables with measurement data and enables online parameter adjustment by minimizing data differences; (2) MWorks automatically calls the solver and outputs a dynamic result comparison curve; (3) Parameter correlation analysis, calibration sensitivity check of related parameters.

5. The method for constructing a virtual-reality collaborative platform for a power equipment regulating device based on digital twin according to claim 1 is characterized in that: Step S3 includes: developing visualization terminal functions based on the 3D modeling tool Maya and the 3D development tool Unit3D, wherein the visualization terminal realizes simulation control management of the professional knowledge base and human-computer interaction.

6. The method for constructing a virtual-reality collaborative platform for a power equipment regulating device based on digital twin according to claim 5 is characterized in that: The professional knowledge base includes: a 3D model module, a graphic module and a video module; the 3D model module has a structure display submodule and a working principle submodule; the graphic module has an information introduction submodule and a fault list submodule; the video module has a principle animation submodule and a fault animation submodule; the human-computer interaction includes: a simulation drive module and a control instruction module; the simulation drive module has a state monitoring submodule and a state drive submodule; the control instruction module has a parameter adjustment submodule and an instruction sending submodule.

7. The method for constructing a virtual-reality collaborative platform for a power equipment regulating device based on digital twin according to claim 1 is characterized in that: Step S4 includes: Universal physical communication module based on Modbus protocol to realize communication between simulation model and physical hardware; The 3D communication external function library based on the MDD library realizes data communication between the simulation model and the 3D visualization interface and control interface.

Images