Lubrication analysis system for water-lubricated bearings based on digital twins and its construction method

Abstract

The present invention discloses a water-lubricated bearing lubrication analysis system based on digital twins and its construction method. The system includes a water-lubricated bearing test platform, a water-lubricated bearing digital twin model, water-lubricated bearing twin data, a bearing twin service system, and the connections between these components. The system can be divided into three levels: the underlying equipment layer, the twin model layer, and the functional application layer. The physical entity part perceives and transmits information through a series of sensors and signal receivers. The digital twin includes a model, a MATLAB lubrication simulation program, a real-time lubrication status monitoring, an operating status prediction, and a feedback control module. It can monitor and predict the bearing lubrication status in real time and provide comprehensive data support for the bearing lubrication status by mapping it with the actual bearing. The present invention can comprehensively and accurately monitor and analyze the lubrication status of the bearing, optimize the bearing design and performance, and realize intelligent lubrication analysis of water-lubricated bearings.

Description

Technical Field

[0001] The present invention relates to the technical field of water-lubricated bearing lubrication, and in particular to a water-lubricated bearing lubrication analysis system based on digital twins and a construction method thereof. Background Art

[0002] The shafting system is a crucial component of a ship's propulsion system, encompassing the main engine, transmission system, bearings, and propellers. Its safety and reliability are crucial to the ship's operation. During water-lubricated bearing testing, analyzing the stable lubrication state of the bearing and visualizing and analyzing it in real time have always been key design challenges. Traditional water-lubricated bearing state analysis typically relies on sensor data and empirical models. While these methods can provide useful information to a certain extent, they also have limitations. First, the acquisition and analysis of sensor data often requires expensive equipment and complex data processing algorithms, increasing system cost and complexity. Second, traditional empirical models often struggle to accurately describe the complex state of lubricated bearings under varying operating conditions, particularly under nonlinear and dynamically changing operating conditions. Analyzing the lubrication characteristics of water-lubricated bearings requires physical experiments in the laboratory to analyze the lubrication state. However, physical laboratories are characterized by high costs, long work hours, heavy workloads, harsh environments, and the inability to intuitively display the bearing lubrication state. The emergence of digital twin technology offers new solutions to these problems.

[0003] In recent years, the emergence of digital twin technology has provided a new approach to addressing these issues. Digital twin technology involves capturing relevant data from physical entities through sensors and data acquisition devices, and then building a real-world twin model of the physical entity using a computer. This allows for comprehensive simulation analysis of bearings, thereby enabling digital experimentation, optimization, and evaluation of bearings. By combining the physical characteristics of the physical system with a virtual simulation model, digital twins enable real-time monitoring, prediction, and optimization of the lubrication state of water-lubricated bearings. Digital twin technology can more accurately simulate the operating state of water-lubricated bearings under different operating conditions, enabling accurate analysis and prediction of the lubrication state, thereby improving system reliability and efficiency. Therefore, applying digital twin technology to lubrication state analysis of water-lubricated bearings has significant theoretical significance and practical application value. The application of digital twin technology can help better understand the lubrication state of water-lubricated bearings during operation, predict future lubrication conditions, optimize design solutions, improve experimental quality, reduce costs, and shorten design cycles. Summary of the Invention

[0004] The purpose of the present invention is to provide a water-lubricated bearing lubrication analysis system based on digital twins and a construction method thereof. Through digital twin technology, real-time monitoring of the lubrication status of bearing experiments, intelligent control and optimization of lubrication characteristics can be achieved, thereby optimizing the lubrication characteristics of water-lubricated bearings, improving the level of bearing lubrication performance experiments, and embodying scientificity, intuitiveness and accuracy.

[0005] To achieve the above objectives, according to a first aspect of the present invention, a water-lubricated bearing lubrication analysis system based on digital twin is provided. The system includes a water-lubricated bearing test platform, a water-lubricated bearing digital twin model, water-lubricated bearing twin data, a bearing twin service system, and connections between various components.

[0006] The water-lubricated bearing test platform is used to provide real physical data and experimental operating conditions for the construction and calibration of the water-lubricated bearing digital twin model;

[0007] The digital twin model of water-lubricated bearings is used to fully model and simulate water-lubricated bearings, enabling real-time monitoring and optimization of their operating process. The digital twin model digitizes and aggregates various elements of the water-lubricated bearing test bench, including the vibration of the water-lubricated bearings, bearing pressure distribution, bearing water film thickness, bearing inlet and outlet water temperature, and friction torque, to achieve comprehensive modeling and simulation of the water-lubricated bearings. The digital twin model of water-lubricated bearings also includes a lubrication program developed based on the MATLAB programming environment. Based on the initial parameters input from the water-lubricated bearing test, it performs simulation calculations of the internal fluid dynamics of the water-lubricated bearings. The lubrication program simulates and analyzes the operating process of the water-lubricated bearings in real time, including the bearing friction coefficient, bearing water film thickness, and predicts the bearing lubrication status.

[0008] The water-lubricated bearing twin data collects information and data from the water-lubricated bearing test bench through physical equipment, including water-lubricated bearing operating status information, water-lubricated bearing laboratory physical environment information, video monitoring information, and test information during the water-lubricated bearing experiment;

[0009] The bearing twin service system is used to provide a visual interface display of the bearing model, real-time human-computer interaction, experimental data processing and storage, and experimental process monitoring. It also includes a model training module. By analyzing the experimental data of water-lubricated bearings, the model is intelligently trained. Through machine learning and deep learning models, the bearing lubrication performance is optimized, and the level of lubrication status monitoring and analysis of water-lubricated bearings is improved.

[0010] The digital twin model of the water-lubricated bearing realizes data sharing and data interaction with the water-lubricated bearing test bench through digital twin technology, thereby improving data utilization and the accuracy and efficiency of data analysis; specifically, by writing protocols for communication, the collected information is accurately transmitted and quickly processed to achieve effective data transmission; the connection between the physical bearing experimental platform and the water-lubricated bearing twin model realizes the digital mapping of the physical model, and the connection between the physical model and the twin data and the connection between the water-lubricated bearing twin model and the twin data realizes data-driven real-time simulation and synchronous feedback of bearing experimental data; the connection between the bearing twin data and the bearing twin service system realizes data analysis, decision-making and optimization in the system service.

[0011] Furthermore, the operating status information of the water-lubricated bearing includes the bearing operating status, input current, voltage and bearing fault information. The bearing fault information includes time, components and fault cause; the physical environment information of the water-lubricated bearing laboratory includes the test bench location, time, room temperature and air humidity; the video monitoring information includes video data collected by the monitoring equipment in the water-lubricated bearing laboratory; the test information during the water-lubricated bearing experiment includes bearing speed, bearing pressure, bearing inlet and outlet water temperature, bearing vibration, bearing inner and outer diameter dimensions, water film thickness and friction coefficient.

[0012] Furthermore, the water-lubricated bearing test platform is also used to test and verify the performance and predictive capabilities of the water-lubricated bearing digital twin model. By comparing the digital model with relevant data of the real physical system, the advantages and limitations of the digital model are determined, and the MATLAB lubrication program is further improved and corrected.

[0013] Furthermore, the system is divided into three layers: underlying device layer, twin model layer, and functional application layer;

[0014] The underlying equipment layer is divided into physical equipment and hardware support equipment. The physical equipment includes a water-lubricated bearing test bench, video monitor, feedback controller, sensors placed on the water-lubricated bearing, torque meters, and temperature measurement devices. Real-time data acquisition and transmission from the physical equipment drives the operation of the digital twin model to complete the entire process and compare it with the results of the MATLAB lubrication program to achieve synchronous simulation. The hardware support equipment is used to support the physical equipment. The data collected by the physical equipment needs to be transmitted through the data transmission interface. At the same time, the feedback control, instruction issuance, and command execution during the human-computer interaction process are all transmitted to the physical equipment through the data transmission interface.

[0015] The twin model layer is used for model building and data processing. Based on the acquired physical test bench, water-lubricated bearing geometric dimensions, physical conditions and regular conditions, the water-lubricated bearing test bench is modeled in an all-round and multi-level manner through modeling software, model processing software and scene rendering software. The model is driven to perform real-time synchronous simulation motion through the written script program. For the data acquisition part, a protocol is written for communication to realize the communication between the water-lubricated bearing test platform and the water-lubricated bearing digital twin model, the data forwarding and transmission of the physical water-lubricated bearing test bench, and the continuous iterative update of the virtual model motion state in the information space based on the twin data. The collected data is classified and stored in the database for subsequent data exchange, feedback control, model update optimization and evaluation. In the MATLAB lubrication program simulation part, simulation is performed with the same parameters and working conditions to achieve real-time comparison and analysis of data, and compare the similarities and differences between the actual working conditions and the simulation parameters. The curves of the friction coefficient and water film thickness parameters of the water-lubricated bearing are fitted through data-driven fitting.

[0016] The functional application layer includes a visualization interface, a model training optimization and evaluation module, and a fault simulation module. The visualization interface can display the operating status and related data of water-lubricated bearings in two and three dimensions, including real-time speed, vibration conditions, and bearing inlet and outlet water temperatures. This interface can be used to control and debug bearing-related data, set the parameters required for the experiment, and conduct water-lubricated bearing experiments. The visualization interface also provides a historical data query function to view historical operating conditions over a period of time, analyze data, and predict the future operating status of water-lubricated bearings. The model training optimization and evaluation module conducts deep learning and machine learning on experimental data, performs intelligent training, and optimizes the design and performance of water-lubricated bearings. The fault simulation module pre-demonstrates possible faults that may be encountered during the water-lubricated bearing experiment to reduce experimental abnormalities.

[0017] Furthermore, a lubrication program developed based on the MATLAB programming environment performs simulation calculations of the internal fluid dynamics of water-lubricated bearings. By analyzing the fluid dynamics model and considering the bearing material properties, the thickness of the water lubrication film and the changes in the friction coefficient are dynamically simulated and predicted.

[0018] Furthermore, the lubrication program developed based on the MATLAB programming environment flexibly adjusts the input parameters to analyze and compare the lubrication status of water-lubricated bearings under different working conditions, so as to optimize the bearing design and operating parameters.

[0019] Furthermore, the model training, optimization, and evaluation module is responsible for extracting lubrication data features from experimental data of water-lubricated bearings and using the extracted features to train and update the machine learning model to generate predictions and decisions for the digital twin system. This module consists of the following three parts:

[0020] Data preprocessing: Before training the model, the bearing lubrication data is preprocessed, including data cleaning, outlier removal, and feature extraction steps to improve the accuracy and robustness of the model;

[0021] Model training: Preprocessed data is fed into a machine learning model to learn patterns and relationships within the data. Leveraging the powerful nonlinear function fitting capabilities of the BP neural network, the complex nonlinear functional relationships between water film thickness and friction coefficient and lubrication characteristics are fitted by given input and output parameters related to water-lubricated bearings. The goal of model training is to generate a model that can accurately predict the future lubrication state of water-lubricated bearings.

[0022] Model update: As time goes by and data accumulates, the model is continuously updated to adapt to new data distribution and patterns.

[0023] Furthermore, the fault simulation module is used to simulate lubrication anomalies in actual bearing tests to evaluate the robustness and response capabilities of the digital twin model; the fault simulation module simulates a variety of lubrication conditions, including dry friction, mixed lubrication, and full lubrication, and simulates different lubrication conditions by adding or deleting certain parameters to test the response and performance of water-lubricated bearings under different lubrication conditions.

[0024] According to a second aspect of the present invention, a method for constructing a digital twin-based water-lubricated bearing lubrication analysis system is provided, which is applied to any of the digital twin-based water-lubricated bearing lubrication analysis systems described above, comprising:

[0025] Step 1: Physical entity analysis: Analyze and decompose the various elements of the water-lubricated bearing test bench, use different signal sensors to collect data from different parts, and perform data transmission and processing;

[0026] Step 2: Model construction: Based on the analysis of the physical test bench elements, the test bench is described and characterized at the physical and geometric levels. The model includes both the water-lubricated bearing test bench and the physical environment of the entire laboratory.

[0027] Step 3: Lightweight processing: Lighten the model;

[0028] Step 4: Rendering with Unity3D: Import the lightweight processed format file into Unity3D and use Unity3D for rendering;

[0029] Step 5: Design a visualization interface: Design a visualization interface in Unity3D for users to interact with and add various experimental parameters to the interface;

[0030] Step 6: Lubrication model optimization: Use the MATLAB lubrication program to simulate the lubrication characteristics of water film thickness and friction coefficient, and compare them with actual experimental data to optimize the lubrication equation.

[0031] Furthermore, the system operates as follows:

[0032] Input the initial test parameters, including: input the rotation speed of the water-lubricated bearing, input the pressure applied to the bearing, and input the water flow rate flowing through the bearing;

[0033] Run the simulation program: Start the MATLAB simulation program, run the preset model based on the input parameters, and calculate the water film thickness and friction coefficient lubrication characteristics of the water-lubricated bearing under different conditions;

[0034] Get simulation results: The simulation program will output lubrication characteristic data including water film thickness and friction coefficient, and present them in graphical or numerical form;

[0035] Perform data comparison: Collect the water film thickness and friction coefficient data obtained in the actual experiment, compare and analyze the simulation results with the actual experimental data, and determine the accuracy and error range of the simulation model;

[0036] Optimize the lubrication equation: Evaluate the performance of the simulation model based on the data comparison results. Adjust the lubrication equation parameters in the MATLAB program based on the errors and deviations found, rerun the simulation program, perform data comparison again, and continuously iterate and optimize the lubrication model until the simulation results match the experimental data.

[0037] In general, the above technical solutions conceived by the present invention can achieve the following beneficial effects compared with the prior art:

[0038] The present invention proposes a water-lubricated bearing lubrication analysis system based on digital twins. The digital twin water-lubricated bearing lubrication analysis system includes a bearing physical entity, a digital twin, and a twin service system. The bearing state detection module of the physical entity part perceives and transmits information through a series of sensors and signal receivers; the digital twin includes a model, a MATLAB lubrication simulation program, a real-time monitoring of the lubrication state, an operating state prediction and feedback control module, which can monitor and predict the bearing lubrication state in real time, and provide comprehensive data support for the bearing lubrication state through mutual mapping with the actual bearing. Through the intelligent prediction and feedback control of the digital twin system, the accuracy and efficiency of the bearing experiment are improved. The digital twin water-lubricated bearing system proposed in the present invention can comprehensively and accurately monitor and analyze the lubrication state of the bearing, optimize the design and performance of the bearing, and realize the intelligent lubrication analysis of the water-lubricated bearing. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1This is a structural diagram of the water-lubricated bearing lubrication analysis system based on digital twins;

[0040] Figure 2 This is the framework diagram of the water-lubricated bearing lubrication analysis system based on digital twin;

[0041] Figure 3 Modeling flow chart for lubrication analysis system of water-lubricated bearings based on digital twin;

[0042] Figure 4 This is a visual operation interface diagram of the water-lubricated bearing lubrication analysis system based on digital twin;

[0043] Figure 5 This is a schematic diagram of the communication code between PLC and SQL Server. DETAILED DESCRIPTION

[0044] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only intended to illustrate the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.

[0045] The present invention provides a water-lubricated bearing lubrication analysis system based on digital twin and a construction method thereof, and relates to the design and construction of a water-lubricated bearing lubrication state analysis system based on digital twin.

[0046] like Figure 1 As shown, based on the digital twin five-dimensional model, the digital twin water-lubricated bearing lubrication state analysis system of the present invention includes a water-lubricated bearing test platform, a water-lubricated bearing digital twin model, water-lubricated bearing twin data, a bearing twin service system and connections between various parts.

[0047] The water-lubricated bearing test platform is the foundation of the entire digital twin system and a prerequisite for establishing digital twin models and conducting effective analysis. It provides real-world physical data and experimental operating conditions. These data and conditions are collected and recorded through sensors, monitors, and other physical devices for use in building and calibrating digital models, ensuring their accuracy and reliability.

[0048] The water-lubricated bearing test bench can also be used to test and validate the performance and predictive capabilities of digital models. By comparing the digital models with data related to real physical systems, the strengths and limitations of the digital models can be identified, allowing for further improvements and corrections to the MATLAB lubrication program.

[0049] The digital twin model of a water-lubricated bearing is a key component of the digital twin system. Its function is to fully model and simulate water-lubricated bearings, enabling real-time monitoring and optimization of their operation. Specifically, the digital twin model digitizes and aggregates various elements of a water-lubricated bearing test bench, including vibration, bearing pressure distribution, bearing water film thickness, bearing inlet and outlet water temperatures, friction torque, and other aspects, enabling comprehensive modeling and simulation of water-lubricated bearings.

[0050] Furthermore, a lubrication program developed in the MATLAB programming environment simulates the internal fluid dynamics of water-lubricated bearings based on the initial parameters input from the water-lubricated bearing test. This program can simulate and analyze the operation of water-lubricated bearings in real time, including the bearing friction coefficient, bearing water film thickness, and predict the bearing lubrication status. This provides targeted optimization solutions, thereby improving the analysis of the lubrication characteristics of water-lubricated bearings. Furthermore, the virtual model can share and interact with physical test benches through digital twin technology, improving data utilization and the accuracy and efficiency of data analysis.

[0051] The twin data of water-lubricated bearings needs to collect information and data from the bearing test bench through sensors, RFID tags, etc. The collection of multi-source heterogeneous data should include the following aspects: (1) Water-lubricated bearing operating status information: bearing operating status, input current, voltage, bearing fault information (time, components, fault cause); (2) Water-lubricated bearing laboratory physical environment information: test bench location, time, room temperature, air humidity; (3) Video monitoring information: video data collected by monitoring equipment in the water-lubricated bearing laboratory, including laboratory physical equipment, laboratory staff on duty, laboratory visitors, experimental equipment replacement process, experimental engineering change process, etc.; (4) Test information during the water-lubricated bearing experiment: bearing speed, bearing pressure, bearing inlet and outlet water temperature, bearing vibration, bearing inner and outer diameter dimensions, water film thickness, friction coefficient, etc.

[0052] The bearing twin service system provides functions such as visual display of bearing models, real-time human-computer interaction, experimental data processing and storage, and experimental process monitoring. This component should also include a model training module, which analyzes water-lubricated bearing experimental data to intelligently train the model. This module uses machine learning and deep learning models to optimize bearing lubrication performance and improve the monitoring and analysis of water-lubricated bearing lubrication conditions.

[0053] The present invention uses the S7 Communication via TCP / IP module in the TIA Portal V15.1 version to write a protocol for communication, accurately transmits and quickly processes the information collected by the data collection module, and realizes effective data transmission. The connection between the physical bearing experimental platform and the water-lubricated bearing model can realize the digital mapping of the physical model. The connection between the physical model and the twin data and the connection between the bearing twin model and the twin data can realize data-driven real-time simulation and synchronous feedback of bearing experimental data. The connection between the bearing twin data and the system service can realize data analysis, decision-making and optimization in the system service.

[0054] In particular, based on these five parts, the present invention can be divided into three levels, namely: the underlying device layer, the twin model layer and the functional application layer, such as Figure 2 shown.

[0055] In the underlying device layer, such as Figure 2 As shown, it is divided into physical equipment and hardware support equipment. Physical equipment is the basis of the present invention. All data sources, model basis, and numerical dimensions of the present invention come from this section. Physical equipment mainly includes water-lubricated bearing test benches, video monitors, feedback controllers, sensors placed on water-lubricated bearings, torque meters, temperature measuring devices, etc. Through the real-time data acquisition and transmission of these physical equipment, the digital twin model is driven to operate in order to complete the whole process, compare with the results of the MATLAB lubrication equation, and realize synchronous simulation; hardware support equipment is the basic condition for realizing the twin model. The data collected by the physical equipment needs to be transmitted through the data transmission interface. At the same time, the feedback control, instruction issuance, and command execution in the human-computer interaction process also need to be transmitted to the physical equipment through the data transmission interface, so as to perform corresponding reaction operations. For the construction of the twin model, based on a specific platform, the present invention is based on the Unity 3D 2022 platform, uses the Windows 10 operating system, and uses development tools such as Visual Studio 2021 and MATLAB to implement it in development languages ​​such as C# and C++.

[0056] In the twin model layer, such as Figure 2As shown in the figure, this section primarily handles model construction and data processing. Based on the physical test bench, water-lubricated bearing geometry, physical conditions, and regulatory requirements, a comprehensive, multi-layered model of the water-lubricated bearing test bench was constructed using professional modeling software, model processing software, and scene rendering software. A C# script was then used to drive the model's real-time, synchronized simulation motion. For data acquisition, the S7 Communication via TCP / IP module in TIA Portal V15.1 was used to implement communication between the Siemens PLC S7-1200 and Unity 3D. This protocol forwarded and transmitted data from the physical water-lubricated bearing test bench, enabling continuous iterative updates of the virtual model's motion state in the information space based on twin data. The collected data was categorized and stored in a MySQL 8.0 database for subsequent data exchange, feedback control, model update optimization, and evaluation. In the MATLAB lubrication simulation, simulations were performed using the same parameters and operating conditions, enabling real-time data comparison and analysis, comparing the differences between the actual operating conditions and the simulated parameters. Data-driven curve fitting was performed for parameters such as the friction coefficient and water film thickness of the water-lubricated bearing.

[0057] In the functional application layer, such as Figure 2 As shown, the system primarily comprises a visualization interface, a model training, optimization, and evaluation module, and a fault simulation module. The visualization interface displays the operating status of water-lubricated bearings in both two and three dimensions, including relevant data such as real-time speed, vibration, and bearing inlet and outlet water temperatures. This interface allows users to control and debug bearing data, set experimental parameters, and conduct water-lubricated bearing experiments. This interface also provides a historical data query function, allowing users to view historical operating conditions over a period of time, analyze data, and predict future operating conditions for water-lubricated bearings. The model training, optimization, and evaluation modules utilize Python to perform deep learning and machine learning on experimental data, enabling intelligent training. For example, Python machine learning libraries can be used to train models to optimize the design and performance of water-lubricated bearings. The fault simulation module, based on this data, can simulate potential faults encountered during water-lubricated bearing experiments, thereby minimizing experimental anomalies. This improves experimental efficiency, reduces the loss of experimental equipment and resources, and wastes human resources, ultimately achieving intelligent laboratory operation.

[0058] Among them, the bearing test platform has a digital twin model and a real physical system. It models the water-lubricated bearing test bench in four dimensions: geometry, physics, behavior, and rules, and constructs a height mapping from the physical test bench to the virtual test bench. It can also transmit the data of the physical bearing system into the digital twin model to verify the accuracy of the digital twin model, and use the digital twin model to simulate and optimize the performance of the physical bearing system; data acquisition equipment, including eddy current sensors, pressure sensors, torque meters, etc., are used to collect data from the physical bearing system. The data that needs to be collected by the acquisition system include: the temperature of the water inlet and outlet of the bearing during the experiment, the noise of the bearing vibration, the speed of the bearing, the water film thickness of the water-lubricated bearing, and the real-time pressure of the bearing, and these parameters are input into the MATLAB lubrication program. Through data analysis, optimization and decision-making, the data of the digital twin model and the actual bearing system are processed to provide analysis results on the bearing performance;

[0059] The system also includes a control system for controlling the operation of the bearing test platform, data acquisition equipment, data processing equipment, and data storage equipment. For the torque, vibration, and water film thickness data collected by the Siemens PLC S7-1200, the communication is carried out through the S7 Communication via TCP / IP module in TIA Portal V15.1, such as Figure 5 As shown, the real-time data is then transmitted to the digital twin model on the Unity 3D client, where multi-source heterogeneous data is processed and analyzed to achieve real-time simulation of the digital twin bearing lubrication system. The data transmission module can also be remotely accessed and controlled over the network, allowing remote users to access and control the physical bearing test conditions in the digital twin water-lubricated bearing test rig and obtain real-time data and analysis results.

[0060] like Figure 4 As shown in the figure, the visual interface provides real-time monitoring capabilities, displaying real-time data and analysis results from the digital twin model and the physical bearing system. This data and analysis results can include information such as the bearing system's temperature, pressure, speed, and vibration acceleration, as well as the differences and similarities between the digital twin model and the physical bearing system. The interface also provides interactive features, allowing users to interact with the digital twin model and the physical bearing system. Users can modify the parameters of the digital twin model through the visual interface to control the operating status of the actual bearing system and conduct virtual-realistic fusion simulation experiments.

[0061] A lubrication program, developed in the MATLAB programming environment, simulates the internal fluid dynamics of water-lubricated bearings based on the input parameters. By analyzing the fluid dynamics model and considering the bearing material properties, the program dynamically simulates and predicts changes in water lubrication film thickness and friction coefficient. The program can flexibly adjust input parameters to analyze and compare the lubrication state of water-lubricated bearings under different operating conditions, thereby optimizing bearing design and operating parameters. This module also includes a model training and evaluation module, which extracts lubrication data features from experimental data of water-lubricated bearings and uses these features to train and update a machine learning model to generate predictions and decisions for the digital twin system. This module consists of three parts: Data preprocessing: Before model training, the bearing lubrication data is preprocessed, including data cleaning, outlier removal, and feature extraction, to improve model accuracy and robustness. Model training: The preprocessed data is input into the machine learning model to learn patterns and relationships within the data. Leveraging the powerful nonlinear function fitting capabilities of the BP neural network, the complex nonlinear functional relationship between water film thickness and friction coefficient and lubrication characteristics is fitted by given input and output parameters related to water-lubricated bearings. The purpose of model training is to generate a model that can accurately predict the lubrication status of water-lubricated bearings in the future. Model update: As time goes by and data accumulates, the model needs to be continuously updated to adapt to new data distribution and patterns.

[0062] The fault simulation system is used to simulate lubrication anomalies in actual bearing tests to evaluate the robustness and response capabilities of the digital twin model. The system includes an automatic testing module for performing automated testing of the bearing test platform and recording the test results. The intelligent diagnosis module is used to diagnose the lubrication status of the actual bearing system using the digital twin model and provide fault diagnosis reports and repair recommendations. The fault simulation system can simulate a variety of lubrication conditions, such as dry friction, mixed lubrication, and full lubrication. The system can simulate different lubrication conditions by adding or deleting certain parameters to test the response and performance of water-lubricated bearings under different lubrication conditions. In addition, the system can also record and analyze fault data for fault diagnosis and improvement of the design and performance of bearing lubrication systems.

[0063] The following combination Figure 3 The steps of building the invention are described as follows:

[0064] Step 1: Physical Analysis: Analyze and decompose the various components of the water-lubricated bearing test bench. Use various signal sensors, information collectors, and RFID tags to collect data on different components, including bearing speed, applied pressure, inlet and outlet water temperatures, bearing vibration frequency, bearing inner and outer diameters, and friction between components. Connectivity and Data Transmission Processing: Connect the various devices and process the data. This process involves connecting sensors, data acquisition cards, and computers. Protocols are used to define the data format and transmission method to ensure the accuracy and reliability of data transmission.

[0065] Step 2: Model Construction: Based on the analysis of the physical test bench elements, the test bench is described and characterized at the physical and geometric levels. The model includes both the water-lubricated bearing test bench and the entire physical laboratory environment. Using SolidWorks modeling software, the shape, position, dimensions, and connections of the water-lubricated bearing are modeled on a 1:1 scale. During model design, details and textures are added to enhance the model's realism. For example, textures and wear effects are added to the surface of the water-lubricated bearing to simulate actual use. After the design is completed, the model needs to be checked and revised to ensure its accuracy.

[0066] Step 3: Lightweighting: Import the model created in Step 2 into 3ds Max for lightweighting. Digital twin simulations typically require extensive computation and calculations on complex 3D models. Overly complex models can reduce computational efficiency and rendering speed. Therefore, lightweighting is essential. Model complexity and storage space can be reduced by removing details and simplifying geometry.

[0067] Step 4: Rendering in Unity3D: Import the lightweight .FPX file from 3ds Max into Unity3D for rendering. In Unity3D, you can set scene, lighting, material, and texture properties to achieve optimal rendering results. Use C# scripts to control the behavior of the water-lubricated bearing twins, including rotation and acceleration. You can create and modify C# scripts using the Unity3D editor, or you can use other editors.

[0068] Step 5: Design the Visualization Interface: Design a visualization interface in Unity3D for user interaction. Use Unity3D's UI tools to create user interface elements such as buttons, text boxes, and drop-down lists, and connect them to functions in the C# script. Add various experimental parameters, such as speed, temperature, and pressure, to the interface. Users can select experimental parameters, start the experiment, and observe the results.

[0069] Step 6: Lubrication model optimization. In this step, a pre-programmed lubrication simulation program in MATLAB is used. This program simulates lubrication characteristics such as water film thickness and friction coefficient by inputting the parameters of the initial water-lubricated bearing test system. This is then compared with actual experimental data to optimize the lubrication equation.

[0070] The specific steps are as follows:

[0071] 1. Input the initial test parameters:

[0072] Speed: Enter the rotational speed of the water-lubricated bearing (in revolutions per minute, RPM);

[0073] Pressure: Enter the pressure applied to the bearing (in Pascals, Pa);

[0074] Flow rate: Enter the water flow rate through the bearing (in liters per minute, L / min).

[0075] 2. Run the simulation program:

[0076] Start the MATLAB simulation program and enter the above parameters. The program runs the preset lubrication model based on the input parameters and calculates the lubrication characteristics of the water-lubricated bearing, such as water film thickness and friction coefficient, under different conditions.

[0077] 3. Get simulation results:

[0078] The simulation program outputs a variety of lubrication characteristic data, including water film thickness and friction coefficient. These data are presented in graphical or numerical form and compared with actual experimental data.

[0079] 4. Compare data:

[0080] Collect data such as water film thickness and friction coefficient obtained from actual experiments. Compare and analyze the simulation results with the actual experimental data to determine the accuracy and error range of the simulation model.

[0081] 5. Optimize the lubrication equation:

[0082] Evaluate the performance of the simulation model based on the data comparison results. Adjust the lubrication equation parameters in the MATLAB program based on any errors and deviations found. Rerun the simulation program and perform another data comparison, iterating and optimizing the lubrication model until the simulation results closely match the experimental data.

[0083] By following these steps, the MATLAB simulation program can be used to effectively optimize the lubrication equations and improve the accuracy and reliability of the water-lubricated bearing simulation model. This method not only accelerates the design process but also reduces experimental costs, providing strong technical support for the development and application of lubrication state analysis for water-lubricated bearings.

[0084] It should be noted that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0085] It should be pointed out that, according to the needs of implementation, the various steps / components described in this application can be split into more steps / components, or two or more steps / components or partial operations of steps / components can be combined into new steps / components to achieve the purpose of the present invention.

[0086] It will be easily understood by those skilled in the art that the above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A water-lubricated bearing lubrication analysis system based on digital twin, characterized in that: The system includes a water-lubricated bearing test platform, a water-lubricated bearing digital twin model, water-lubricated bearing twin data, a bearing twin service system, and the connections between various parts; The water-lubricated bearing test platform is used to provide real physical data and experimental operating conditions for the construction and calibration of the water-lubricated bearing digital twin model; The digital twin model of water-lubricated bearings is used to fully model and simulate water-lubricated bearings, enabling real-time monitoring and optimization of their operating process. The digital twin model digitizes and aggregates various elements of the water-lubricated bearing test bench, including the vibration of the water-lubricated bearings, bearing pressure distribution, bearing water film thickness, bearing inlet and outlet water temperature, and friction torque, to achieve comprehensive modeling and simulation of the water-lubricated bearings. The digital twin model of water-lubricated bearings also includes a lubrication program developed based on the MATLAB programming environment. Based on the initial parameters input from the water-lubricated bearing test, it performs simulation calculations of the internal fluid dynamics of the water-lubricated bearings. The lubrication program simulates and analyzes the operating process of the water-lubricated bearings in real time, including the bearing friction coefficient, bearing water film thickness, and predicts the bearing lubrication status. The water-lubricated bearing twin data collects information and data from the water-lubricated bearing test bench through physical equipment, including water-lubricated bearing operating status information, water-lubricated bearing laboratory physical environment information, video monitoring information, and test information during the water-lubricated bearing experiment; The bearing twin service system is used to provide a visual interface display of the bearing model, real-time human-computer interaction, experimental data processing and storage, and experimental process monitoring. It also includes a model training module. By analyzing the experimental data of water-lubricated bearings, the model is intelligently trained. Through machine learning and deep learning models, the bearing lubrication performance is optimized, and the level of lubrication status monitoring and analysis of water-lubricated bearings is improved. The digital twin model of the water-lubricated bearing realizes data sharing and data interaction with the water-lubricated bearing test bench through digital twin technology, thereby improving data utilization and the accuracy and efficiency of data analysis; specifically, by writing protocols for communication, the collected information is accurately transmitted and quickly processed to achieve effective data transmission; the connection between the physical bearing experimental platform and the water-lubricated bearing twin model realizes the digital mapping of the physical model, and the connection between the physical model and the twin data and the connection between the water-lubricated bearing twin model and the twin data realizes data-driven real-time simulation and synchronous feedback of bearing experimental data; the connection between the bearing twin data and the bearing twin service system realizes data analysis, decision-making and optimization in the system service.

2. The water-lubricated bearing lubrication analysis system based on digital twin according to claim 1 is characterized in that: The operating status information of water-lubricated bearings includes the bearing operating status, input current, voltage and bearing fault information. The bearing fault information includes the time, components and fault cause; the physical environment information of the water-lubricated bearing laboratory includes the test bench location, time, room temperature and air humidity; the video monitoring information includes the video data collected by the monitoring equipment in the water-lubricated bearing laboratory; the test information during the water-lubricated bearing experiment includes the bearing speed, bearing pressure, bearing inlet and outlet water temperature, bearing vibration, bearing inner and outer diameter dimensions, water film thickness and friction coefficient.

3. The water-lubricated bearing lubrication analysis system based on digital twin according to claim 1 is characterized in that: The water-lubricated bearing test platform is also used to test and verify the performance and predictive capabilities of the water-lubricated bearing digital twin model. By comparing the digital model with relevant data from the real physical system, the advantages and limitations of the digital model are determined, and the MATLAB lubrication program is further improved and corrected.

4. The water-lubricated bearing lubrication analysis system based on digital twin according to claim 1 is characterized in that: The system is divided into three layers: underlying device layer, twin model layer and functional application layer; The underlying equipment layer is divided into physical equipment and hardware support equipment. The physical equipment includes a water-lubricated bearing test bench, video monitor, feedback controller, sensors placed on the water-lubricated bearing, torque meters, and temperature measurement devices. Real-time data acquisition and transmission from the physical equipment drives the operation of the digital twin model to complete the entire process and compare it with the results of the MATLAB lubrication program to achieve synchronous simulation. The hardware support equipment is used to support the physical equipment. The data collected by the physical equipment needs to be transmitted through the data transmission interface. At the same time, the feedback control, instruction issuance, and command execution during the human-computer interaction process are all transmitted to the physical equipment through the data transmission interface. The twin model layer is used for model building and data processing. Based on the acquired physical test bench, water-lubricated bearing geometric dimensions, physical conditions and regular conditions, the water-lubricated bearing test bench is modeled in an all-round and multi-level manner through modeling software, model processing software and scene rendering software. The model is driven to perform real-time synchronous simulation motion through the written script program. For the data acquisition part, a protocol is written for communication to realize the communication between the water-lubricated bearing test platform and the water-lubricated bearing digital twin model, the data of the physical water-lubricated bearing test bench is forwarded and transmitted, and the motion state of the virtual model in the information space is continuously iterated and updated based on the twin data. The collected data is classified and stored in the database for subsequent data exchange, feedback control, model update optimization and evaluation. In the simulation part of the MATLAB lubrication program, the same parameters and working conditions are used for simulation to achieve real-time comparison and analysis of data, and to compare the similarities and differences between the actual working conditions and the simulation parameters. The friction coefficient and water film thickness parameter curves of water-lubricated bearings are fitted by data-driven fitting; The functional application layer includes a visualization interface, a model training optimization and evaluation module, and a fault simulation module. The visualization interface can display the operating status and related data of water-lubricated bearings in two and three dimensions, including real-time speed, vibration, and bearing inlet and outlet water temperatures. This interface allows control and debugging of bearing-related data, setting experimental parameters, and conducting water-lubricated bearing experiments. The visualization interface also provides a historical data query function, allowing users to view historical operating conditions over a period of time, analyze data, and predict the future operating status of water-lubricated bearings. The model training optimization and evaluation module conducts deep learning and machine learning on experimental data, performs intelligent training, and optimizes the design and performance of water-lubricated bearings; the fault simulation module pre-demonstrates the faults that may be encountered during the water-lubricated bearing experiment to reduce experimental anomalies.

5. The digital twin-based water-lubricated bearing lubrication analysis system according to claim 4 is characterized in that: The lubrication program developed based on the MATLAB programming environment performs simulation calculations of the internal fluid dynamics of water-lubricated bearings. By analyzing the fluid dynamics model and considering the bearing material characteristics, it dynamically simulates and predicts the thickness of the water lubrication film and the change in the friction coefficient.

6. The water-lubricated bearing lubrication analysis system based on digital twin according to claim 5 is characterized in that: The lubrication program developed based on the MATLAB programming environment flexibly adjusts input parameters to analyze and compare the lubrication status of water-lubricated bearings under different operating conditions, so as to optimize bearing design and operating parameters.

7. The water-lubricated bearing lubrication analysis system based on digital twin according to claim 4 is characterized in that: The model training, optimization, and evaluation module is responsible for extracting lubrication data features from experimental data of water-lubricated bearings and using the extracted features to train and update the machine learning model to generate predictions and decisions for the digital twin system. This module consists of the following three parts: Data preprocessing: Before training the model, the bearing lubrication data is preprocessed, including data cleaning, outlier removal, and feature extraction steps to improve the accuracy and robustness of the model; Model training: feeding preprocessed data into a machine learning model to learn patterns and relationships in the data; Leveraging the powerful nonlinear function fitting capabilities of the BP neural network, the complex nonlinear functional relationship between water film thickness and friction coefficient and lubrication characteristics is fitted by given input and output parameters related to water-lubricated bearings. The purpose of model training is to generate a model that can accurately predict the future lubrication status of water-lubricated bearings. Model update: As time goes by and data accumulates, the model is continuously updated to adapt to new data distribution and patterns.

8. The water-lubricated bearing lubrication analysis system based on digital twin according to claim 4 is characterized in that: The fault simulation module is used to simulate lubrication anomalies in actual bearing tests to evaluate the robustness and response capabilities of the digital twin model. The fault simulation module simulates a variety of lubrication conditions, including dry friction, mixed lubrication, and full lubrication. By adding or deleting certain parameters to simulate different lubrication conditions, the module tests the response and performance of water-lubricated bearings under different lubrication conditions.

9. A method for building a water-lubricated bearing lubrication analysis system based on digital twins, characterized in that: The digital twin-based water-lubricated bearing lubrication analysis system applied to any one of claims 1 to 8 comprises: Step 1: Physical entity analysis: Analyze and decompose the various elements of the water-lubricated bearing test bench, use different signal sensors to collect data from different parts, and perform data transmission and processing; Step 2: Model construction: Based on the analysis of the physical test bench elements, the test bench is described and characterized at the physical and geometric levels. The model includes both the water-lubricated bearing test bench and the physical environment of the entire laboratory. Step 3: Lightweight processing: Lighten the model; Step 4: Rendering with Unity3D: Import the lightweight processed format file into Unity3D and use Unity3D for rendering; Step 5: Design a visualization interface: Design a visualization interface in Unity3D for users to interact with and add various experimental parameters to the interface; Step 6: Lubrication model optimization: Use the MATLAB lubrication program to simulate the lubrication characteristics of water film thickness and friction coefficient, and compare them with actual experimental data to optimize the lubrication equation.

10. The method for constructing a water-lubricated bearing lubrication analysis system based on digital twin according to claim 9, characterized in that: The system operates as follows: Input the initial test parameters, including: input the rotation speed of the water-lubricated bearing, input the pressure applied to the bearing, and input the water flow rate flowing through the bearing; Run the simulation program: Start the MATLAB simulation program, run the preset model based on the input parameters, and calculate the water film thickness and friction coefficient lubrication characteristics of the water-lubricated bearing under different conditions; Get simulation results: The simulation program will output lubrication characteristic data including water film thickness and friction coefficient, and present them in graphical or numerical form; Perform data comparison: Collect the water film thickness and friction coefficient data obtained in the actual experiment, compare and analyze the simulation results with the actual experimental data, and determine the accuracy and error range of the simulation model; Optimize the lubrication equation: Evaluate the performance of the simulation model based on the data comparison results. Adjust the lubrication equation parameters in the MATLAB program based on the errors and deviations found, rerun the simulation program, perform data comparison again, and continuously iterate and optimize the lubrication model until the simulation results match the experimental data.

Images