A method and system for constructing a twin tree structure based on force transmission

Abstract

This invention discloses a method and system for constructing a twin tree structure based on force transmission, belonging to the field of digital twins for wind turbine generators. The method includes: real-time acquisition of wind turbine generator force data; constructing a twin tree structure with force transmission attributes using a recursive algorithm based on the force data to achieve dynamic propagation of force data; optimizing platform layout and verifying assembly using an intelligent layout algorithm incorporating a wake model; integrating multi-source heterogeneous data and generating real-time force visualization results through lightweight finite element analysis; embedding a fault prediction model to establish a force anomaly → fault early warning mechanism; and providing low-code development tools and one-click deployment functionality. This invention achieves deep integration of force analysis and fault prediction, improving the intelligence level and response efficiency of wind turbine generator operation and maintenance.

Description

Technical Field

[0001] This invention belongs to the field of digital twins for wind turbine units, and particularly relates to a method and system for constructing a twin tree structure based on force transmission. Background Technology

[0002] Currently, with the continuous increase in the single-unit capacity of wind turbines and the large-scale development of wind farms, their operation and maintenance management faces severe challenges. Digital twin technology, as an important means to realize condition monitoring and fault prediction of wind turbines, has been initially applied in the industry. Existing digital twin platforms typically include independent stress analysis and fault prediction modules, and attempt to construct virtual mappings of wind turbines through sensor data acquisition and 3D visualization technology. Some studies have also introduced multi-protocol data access functions to adapt to industrial communication protocols such as Modbus and OPC UA, while using finite element analysis methods to perform mechanical simulations of key components. However, these technologies still reveal significant shortcomings in practical engineering applications.

[0003] Specifically, existing platforms generally lack an effective fusion mechanism between stress data and fault prediction models. Stress analysis results are often independent of fault warning logic, leading to a break in the causal chain between "stress anomaly" and "fault occurrence," making it difficult to use real-time stress data to guide accurate fault prediction. Furthermore, the platform layout does not fully consider the unique geographical environment and meteorological conditions of wind farms (such as wake effects and terrain undulations), failing to reflect the stress transmission path under actual operating conditions. In addition, although multi-source heterogeneous data can be accessed through various protocols, there is a lack of a data integration framework oriented towards stress transmission characteristics, and data silos still exist. More significantly, the development and iteration process of existing platforms is rigid, with high barriers to user interface customization and long release cycles, making it difficult to adapt to the rapidly evolving needs of wind power technology. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention provides a method for constructing a twin tree structure based on force transmission, comprising: Collect stress data of wind turbine units; Based on the collected force data, a twin tree structure with force transmission attributes is constructed; Based on the meteorological conditions and geographical environment of the wind farm, the platform layout of the twin tree structure is optimized; Based on the optimized layout, multi-source heterogeneous data is integrated to generate real-time force visualization results; Based on the real-time force visualization results, a fault warning mechanism based on force anomalies is triggered; Based on the aforementioned fault warning mechanism, a user interface is generated and the platform is published.

[0005] Optionally, the collection of force data of the wind turbine generator further includes: By accessing data from multiple types of sensors, multi-source sensor data can be acquired. Based on the multi-source sensor data, fused force data is generated using a data fusion algorithm; Based on the fused force data, preprocessing and feature extraction are performed to generate processed force data.

[0006] Optionally, constructing the twin tree structure with force transmission properties further includes: Based on the physical structure of the wind turbine, a basic framework for the twin is generated; Based on the force data, a recursive algorithm is used to add force transmission attributes to each component in the basic framework of the twin, generating a twin tree structure with force transmission attributes. Based on the force transmission properties, dynamic propagation of force data is achieved in the twin tree structure.

[0007] Optionally, platform layout optimization of the twin tree structure further includes: Based on the wind farm's topography, wind direction, wind speed, and wind turbine arrangement, an intelligent layout algorithm incorporating wake models is used to generate a platform layout scheme. Based on the platform layout scheme, the assembly feasibility is verified, and a verified layout scheme is generated.

[0008] Optionally, the process of integrating multi-source heterogeneous data to generate real-time force visualization results further includes: Data access based on multiple communication protocols is used to acquire heterogeneous data from multiple sources. Based on the multi-source heterogeneous data, seamless integration is performed to generate integrated force data; Based on the integrated stress data, a real-time stress visualization report is generated through lightweight finite element analysis.

[0009] Optionally, the triggering of the fault warning mechanism based on abnormal force further includes: Based on the real-time force visualization results, obtain real-time force data; Based on the real-time stress data, an embedded integrated fault prediction model is used for monitoring to determine whether there is any stress abnormality. Based on the judgment that there is an abnormal force, the fault early warning mechanism is triggered.

[0010] Optionally, generating the user interface and completing the platform release further includes: Based on user requirements, generate the user interface using low-code development tools; Based on the generated user interface, the platform can be deployed and updated using the one-click publish function.

[0011] To address the aforementioned technical problems, this invention also provides a twin tree structure construction system based on force transmission, comprising: The stress data acquisition and processing module is used to acquire stress data of the wind turbine. A twin tree structure construction module is connected to the force data acquisition and processing module, and is used to construct a twin tree structure with force transmission attributes based on the acquired force data; The intelligent layout and assembly verification module is used to optimize the platform layout of the twin tree structure based on the meteorological conditions and geographical environment of the wind farm. The multi-protocol adaptation and lightweight finite element analysis module is used to integrate multi-source heterogeneous data based on the optimized layout and generate real-time force visualization results. An embedded integrated module for the fault prediction model is used to trigger a fault early warning mechanism based on force anomalies according to the real-time force visualization results. The low-code UI and one-click deployment module is used to generate a user interface and complete platform deployment based on the fault warning mechanism. The twin tree structure construction module is further used to generate a basic twin framework based on the physical structure of the wind turbine, add force transmission attributes to each component in the basic twin framework using a recursive algorithm based on the force data, and realize the dynamic propagation of force data in the twin tree structure based on the force transmission attributes.

[0012] On the other hand, the present invention also provides an electronic device including a memory, a processor, and a computing program stored in the memory and executable on the processor, wherein the processor implements the method when executing the computing program.

[0013] On the other hand, the present invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method.

[0014] Compared with the prior art, the present invention has the following advantages and technical effects: This invention significantly improves fault prediction accuracy by constructing a twin tree structure with force transmission attributes to establish a direct causal mapping of "stress anomaly → fault early warning." It employs an intelligent layout algorithm that integrates wake models to match the platform layout with actual wind farm meteorological and geographical conditions, enhancing the engineering realism of stress analysis. A multi-protocol adaptation and lightweight finite element analysis framework is designed to achieve seamless integration of multi-source heterogeneous data and real-time stress visualization, eliminating data silos. Low-code development tools and one-click deployment functionality are embedded, significantly reducing the interface development threshold and shortening the platform iteration cycle by more than 60%. This invention effectively solves core technical problems such as the disconnect between stress analysis and fault prediction, lack of wind power characteristics in layout, difficulty in multi-protocol integration, and rigid deployment iteration, improving the intelligence level and response efficiency of wind turbine operation and maintenance. Attached Figure Description

[0015] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings: Figure 1 This is a schematic diagram of the system structure according to an embodiment of the present invention. Detailed Implementation

[0016] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0017] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0018] Example 1 This embodiment provides a method for constructing a twin tree structure based on force transmission, including: Multi-source sensor data is acquired by accessing data from various types of sensors (including strain gauges, accelerometers, and anemometers) deployed on key components of the wind turbine, such as blades, hubs, nacelles, and towers. Based on this multi-source sensor data, fused stress data is generated using data fusion algorithms (including Kalman filtering or weighted averaging). For example, stress data collected from multiple strain gauges on the blades are fused into a more accurate stress value using a Kalman filtering algorithm. Based on the fused stress data, signal processing techniques (including wavelet transform or Fourier transform) are used for preprocessing and feature extraction to generate processed stress data. For example, wavelet transform is used to extract high-frequency and low-frequency components from the stress signal.

[0019] Based on the processed stress data and the physical structure of the wind turbine, a basic twin framework is generated. This basic framework includes components such as blades, hub, nacelle, and tower. Using the stress data, a recursive algorithm is employed to add stress transfer attributes to each component in the basic twin framework, generating a twin tree structure with these attributes. Based on these stress transfer attributes, the dynamic propagation of stress data within the twin tree structure is achieved. For example, when the blades are subjected to wind force, stress data is transferred to components such as the hub, nacelle, and tower according to the stress transfer attributes, generating visualized information.

[0020] Based on the wind farm's topography, wind direction, wind speed, and other meteorological conditions, as well as the arrangement of the wind turbines, an intelligent layout algorithm incorporating a wake model is adopted. With maximizing the power generation efficiency of the wind turbines as the objective function, a platform layout scheme is generated using a genetic algorithm or particle swarm optimization algorithm. Based on this platform layout scheme, assembly feasibility is verified using finite element analysis, generating a verified layout scheme to ensure that the stress conditions of each wind turbine meet design requirements during actual operation.

[0021] Data is acquired from multiple sources using various communication protocols (including Modbus, OPC UA, and MQTT). This heterogeneous data is then seamlessly integrated through protocol conversion and data processing, converting sensor data from different protocols into a unified internal data format to generate integrated stress data. Based on this integrated stress data, real-time analysis is performed using lightweight finite element analysis algorithms (including reduced basis methods or model order reduction methods) to generate a stress visualization report containing stress distribution contour maps or deformation contour maps.

[0022] Real-time stress data is obtained from the stress visualization report and input into an embedded fault prediction model. This model is constructed using machine learning algorithms (including support vector machines or neural networks). Based on the monitoring results of the fault prediction model, it is determined whether any stress anomalies exist. If an anomaly is detected, a "stress anomaly → fault warning" mechanism is triggered, generating a warning notification.

[0023] Based on the user needs of operations and maintenance personnel, a user interface is generated using the visual components (including buttons, charts, and tables) and drag-and-drop interface design functions provided by low-code user interface development tools. This user interface integrates a twin tree structure for force visualization and a fault warning panel. Based on the generated user interface, the platform's automated building, testing, and deployment are completed using a one-click release function and continuous integration / continuous deployment automation scripts (including Jenkins or GitLab CI / CD), enabling rapid iterative updates.

[0024] On the other hand, this embodiment also provides an electronic device, including a memory, a processor, and a computing program stored in the memory and executable on the processor, wherein the processor implements the method when executing the computing program.

[0025] On the other hand, this embodiment also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method.

[0026] Example 2 This embodiment provides a twin tree structure construction system based on force transmission, including: See Figure 1 This embodiment provides a twin tree structure construction system based on force transmission, suitable for the rapid construction and operation and maintenance management of digital twin platforms for wind turbines. The system includes a force data acquisition and processing module, a twin tree structure construction module, an intelligent layout and assembly verification module, a multi-protocol adaptation and lightweight finite element analysis module, a fault prediction model embedded integration module, and a low-code UI and one-click deployment module.

[0027] The stress data acquisition and processing module collects stress data from the wind turbine in real time. Specifically, this module connects to various types of sensors, including strain gauges, accelerometers, and anemometers, located on key components of the wind turbine such as blades, hub, nacelle, and tower, through a multi-sensor data fusion submodule. The multi-sensor data fusion submodule uses a Kalman filter algorithm to fuse the multi-source sensor data, generating fused stress data. The data processing algorithm submodule further preprocesses the fused stress data using wavelet transform for feature extraction, filtering out noise interference and extracting key indicators reflecting the stress characteristics of the turbine, generating processed stress data.

[0028] The twin tree structure construction module is connected to the stress data acquisition and processing module to receive the processed stress data. This module first generates a basic twin framework, including blades, hub, nacelle, and tower, based on the physical structure of the wind turbine. Then, a recursive algorithm is used to add stress transfer attributes to each component in the basic twin framework, generating a twin tree structure with these attributes. According to these stress transfer attributes, when the blades are subjected to wind force, the stress data at the blade nodes can be dynamically transferred along the tree structure to parent or child components such as the hub, nacelle, and tower, realizing the dynamic propagation of stress data within the model. Simultaneously, this module uses visualization technology to intuitively display the flow path and distribution of the stress data on the user interface.

[0029] The intelligent layout and assembly verification module optimizes the platform layout based on the unique meteorological conditions and geographical environment of the wind farm. Specifically, this module acquires the wind farm's terrain data, historical wind direction statistics, wind speed distribution, and preset arrangement of wind turbines. It employs an intelligent layout algorithm incorporating a wake model, using the maximization of power generation efficiency as the objective function, and generates the optimal platform layout scheme through particle swarm optimization. Subsequently, the assembly verification submodule performs mechanical analysis on each wind turbine in the layout scheme using finite element analysis to verify whether its stress conditions during actual operation meet design requirements, generating a verified layout scheme.

[0030] The multi-protocol adaptation and lightweight finite element analysis module enables seamless integration and real-time force visualization of multi-source heterogeneous data. The multi-protocol adaptation submodule supports data access via multiple communication protocols such as Modbus, OPC UA, and MQTT. Through protocol conversion, it unifies sensor data from different sources into an internal data format, achieving seamless integration of multi-source heterogeneous data. The lightweight finite element analysis submodule employs a reduced basis method to perform real-time finite element analysis on the integrated force data, generating a force visualization report that includes stress distribution contour maps and deformation contour maps.

[0031] The fault prediction model embedded integration module integrates a pre-trained fault prediction model into the platform. This fault prediction model is constructed using support vector machines, with historical stress data and fault records as training samples. The module acquires stress data generated by the lightweight finite element analysis submodule in real time. When an abnormal pattern is detected in the stress data, it automatically determines that an anomaly exists and triggers the "stress anomaly → fault warning" mechanism, issuing a warning notification to maintenance personnel.

[0032] The low-code UI and one-click deployment module provides a visual component library and drag-and-drop interface design tools, allowing non-professional developers to quickly build user interfaces according to operational needs. These user interfaces integrate a twin tree-structured force visualization display and a fault warning panel. The one-click deployment function uses continuous integration / continuous deployment automation scripts to package and deploy the completed platform to the target server, enabling rapid platform iteration and updates.

[0033] After system startup, each module executes sequentially according to the above workflow, continuously monitoring sensor data input and user operation requests. Once new force data is detected, it is processed sequentially through data fusion, twin tree construction, layout verification, finite element analysis, fault prediction, and interface updates, forming a closed-loop operation and maintenance decision support process. This embodiment achieves deep integration of force data and fault prediction through modular architecture design and a twin tree structure of force transmission attributes, improving the construction efficiency and operation and maintenance response capability of the wind turbine digital twin platform.

[0034] Example 3 This embodiment provides a twin tree structure construction system based on force transmission, including: I. Complete implementation of the main technical solutions; Regarding the stress data acquisition and processing module, the multi-sensor data fusion submodule in this embodiment supports data access from various types of sensors, including but not limited to strain gauges, accelerometers, and anemometers. This submodule improves the accuracy and reliability of stress data through data fusion algorithms, such as Kalman filtering or weighted averaging. For example, multiple strain gauges can be installed on wind turbine blades to collect stress data in real time, and the data from multiple strain gauges can be fused into a more accurate stress value using a data fusion algorithm. The data processing algorithm submodule employs advanced signal processing techniques to preprocess and extract features from the acquired stress data, including wavelet transform or Fourier transform. For example, wavelet transform can be performed on the acquired stress data to extract high-frequency and low-frequency components from the stress signal, providing a high-quality data foundation for subsequent analysis.

[0035] Regarding the twin tree structure construction module, this embodiment employs a recursive algorithm to construct a twin tree structure with force transfer attributes. Specifically, firstly, a basic framework for the twin is constructed based on the physical structure of the wind turbine, including components such as blades, hub, nacelle, and tower. Then, force transfer attributes are added to each component, enabling dynamic propagation of force data within the model. For example, when the blades are subjected to wind force, stress data is transferred to components such as the hub, nacelle, and tower through the force transfer attributes, achieving dynamic propagation and visualization of force data within the model.

[0036] Regarding the intelligent layout and assembly verification module, this embodiment employs an intelligent layout algorithm that integrates the wake model. It comprehensively considers meteorological conditions such as wind farm topography, wind direction, and wind speed, as well as the arrangement of wind turbines, and generates an optimal platform layout scheme through optimization algorithms. These optimization algorithms include genetic algorithms or particle swarm optimization algorithms, with the objective function of maximizing the power generation efficiency of the wind turbines. The assembly verification submodule verifies the assembly feasibility of the layout scheme to ensure the stability and reliability of the platform in practical applications. For example, it uses finite element analysis to perform mechanical analysis on each wind turbine in the layout scheme, verifying whether its stress conditions during actual operation meet design requirements.

[0037] Regarding the multi-protocol adaptation and lightweight finite element analysis module, the multi-protocol adaptation submodule in this embodiment supports data access via multiple communication protocols, including but not limited to Modbus, OPC UA, and MQTT. This submodule achieves seamless integration of multi-source heterogeneous data through protocol conversion and data processing. For example, it converts sensor data from different protocols into a unified internal data format, facilitating subsequent processing and analysis. The lightweight finite element analysis submodule employs efficient finite element analysis algorithms to perform real-time analysis of stress data and generate stress visualization reports. These efficient finite element analysis algorithms include reduced basis methods or model order reduction methods; for example, using fast finite element analysis methods to generate visualization reports such as stress distribution cloud maps or deformation cloud maps.

[0038] Regarding the embedded integration module of the fault prediction model, this embodiment integrates the fault prediction model into the platform to establish a "stress anomaly → fault early warning" mechanism. The fault prediction model is constructed using machine learning algorithms, including support vector machines or neural networks. For example, real-time collected stress data is input into the fault prediction model. When the model detects a stress anomaly, it triggers the fault early warning mechanism, notifying maintenance personnel to handle the situation promptly.

[0039] Regarding the low-code UI and one-click deployment module, the low-code user interface development tool in this embodiment provides rich visual components and drag-and-drop interface design capabilities, enabling non-professional developers to quickly build user interfaces that meet their needs. For example, it provides visual components such as buttons, charts, and tables, which users can combine into a user interface by dragging and dropping. The one-click deployment function simplifies the platform's deployment process through automated scripts, enabling rapid deployment and iterative updates. For example, it uses continuous integration / continuous deployment tools (including Jenkins or GitLab CI / CD) to automate the platform's building, testing, and deployment.

[0040] II. Modified or improved implementation methods of the technical solution; Based on the above main technical solutions, this embodiment also provides various modified or improved implementation methods.

[0041] Regarding improvements to the force data acquisition and processing module, the multi-sensor data fusion submodule can incorporate deep learning algorithms to fuse multi-sensor data. These deep learning algorithms include convolutional neural networks (CNNs) or recurrent neural networks (RNNs). For example, a CNN can be used to extract features from image data acquired by an image sensor, and then combined with data from other sensors to improve the accuracy and reliability of the force data. The data processing algorithm submodule can employ adaptive signal processing techniques to automatically adjust processing parameters based on the characteristics of the force data, thereby improving the accuracy of feature extraction. These adaptive signal processing techniques include adaptive filtering or adaptive wavelet transform.

[0042] Regarding improvements to the twin tree structure construction module, virtual reality or augmented reality technologies can be introduced to enable visual interaction of the twin tree structure. For example, through virtual reality devices, operations and maintenance personnel can intuitively observe the propagation of force data within the twin tree structure, thereby improving the accuracy of operations and maintenance decisions.

[0043] Regarding improvements to the intelligent layout and assembly verification module, the intelligent layout algorithm integrating the wake model can incorporate a multi-objective optimization algorithm. This algorithm considers multiple objective functions, such as the power generation efficiency and operation and maintenance costs of the wind turbine, to generate a better platform layout scheme. The multi-objective optimization algorithm includes a multi-objective genetic algorithm or a multi-objective particle swarm optimization algorithm. The assembly verification submodule can use digital twin technology to construct a digital twin model of the wind farm, performing virtual assembly and verification of the layout scheme. For example, the digital twin model can simulate the stress conditions of the wind turbine in actual operation to verify the feasibility and stability of the layout scheme.

[0044] Regarding improvements to the multi-protocol adaptation and lightweight finite element analysis modules, the multi-protocol adaptation submodule can incorporate edge computing technology to offload some data processing and analysis tasks to edge devices, thereby reducing data transmission latency and improving real-time performance. For example, edge computing devices can be deployed near wind turbines to process and analyze sensor data in real time, transmitting only critical data to the cloud platform. The lightweight finite element analysis submodule can employ parallel computing technology to improve the computational efficiency of finite element analysis, including GPU acceleration or distributed computing. For example, the parallel computing capabilities of GPUs can be used to accelerate matrix operations and iterative solution processes in finite element analysis.

[0045] Regarding improvements to the embedded integration module of the fault prediction model, transfer learning techniques can be introduced. Existing fault prediction models can be used for transfer learning to quickly build fault prediction models suitable for new wind turbine units or new wind farms. For example, a fault prediction model trained in one wind farm can be transferred to another wind farm and fine-tuned with a small number of samples to quickly build a fault prediction model suitable for the new wind farm.

[0046] Regarding improvements to the low-code UI and one-click deployment modules, the low-code user interface development tool can incorporate AI-assisted design technology to automatically generate user interfaces based on user needs. This AI-assisted design technology includes natural language processing or computer vision. For example, users can describe their needs via voice or text, and the low-code development tool can automatically generate a user interface that meets those needs. The one-click deployment function can incorporate containerization technology to enable rapid platform deployment and scaling. This containerization technology includes Docker or Kubernetes. For example, the platform can be packaged as a Docker container and deployed and scaled automatically via Kubernetes, thereby improving the platform's availability and scalability.

[0047] III. Implementation methods in specific application scenarios; This embodiment also provides implementation methods for the above technical solutions in two specific application scenarios: offshore wind farms and wind farms in complex terrain.

[0048] For offshore wind farm applications: Regarding the stress data acquisition and processing module, due to the harsh marine environment, sensor data is easily interfered with and damaged. Therefore, waterproof and corrosion-resistant sensors are used, and data verification and error correction mechanisms are introduced to improve the accuracy and reliability of stress data. Regarding the twin tree structure construction module, the arrangement of wind turbines in offshore wind farms differs from that in onshore wind farms, requiring consideration of the influence of ocean airflow and waves. Therefore, ocean airflow and wave models are incorporated when constructing the twin tree structure to more accurately reflect the stress conditions of wind turbines during actual operation. Regarding the intelligent layout and assembly verification module, the layout of offshore wind farms needs to consider factors such as marine topography and marine life protection. Therefore, when using an intelligent layout algorithm that integrates wake models, marine topography data and marine life protection area data are incorporated to generate a platform layout scheme that better suits the characteristics of offshore wind farms. Regarding the multi-protocol adaptation and lightweight finite element analysis module, the communication conditions in offshore wind farms are poor. Satellite communication or wireless private network communication is used to ensure reliable data transmission. At the same time, lightweight finite element analysis methods are used to reduce data transmission volume and improve real-time performance. Regarding the embedded integration module for the fault prediction model, offshore wind farms have high operation and maintenance costs. Incorporating more historical data and operation and maintenance experience when building the fault prediction model improves the accuracy of fault prediction. Regarding the low-code UI and one-click deployment module, offshore wind farm operation and maintenance personnel may lack professional development skills. Therefore, providing more intuitive visual components and simpler operation processes enables operation and maintenance personnel to quickly build and deploy user interfaces.

[0049] For wind farm applications in complex terrain: Regarding the force data acquisition and processing module, the stress conditions of wind turbines in complex terrain wind farms are complex. More sensors are installed on key components of the wind turbines to collect more detailed force data in real time. Simultaneously, more refined signal processing techniques are employed to extract more accurate feature information. These refined signal processing techniques include higher-order wavelet transforms or higher-order Fourier transforms. Regarding the twin tree structure construction module, the arrangement of wind turbines in complex terrain wind farms is irregular. Three-dimensional modeling technology is used to construct a more accurate physical model of the wind turbines. More detailed force transmission attributes are added to each component, enabling force data to propagate more accurately within the model. Regarding the intelligent layout and assembly verification module, the layout of wind farms in complex terrain needs to consider factors such as terrain undulations and obstacles. Therefore, when using an intelligent layout algorithm that integrates wake models, terrain data and obstacle data are introduced to generate a platform layout scheme that better suits the characteristics of complex terrain. At the same time, more refined assembly verification methods are used to verify the feasibility and stability of the layout scheme. These more refined assembly verification methods include finite element analysis or physical simulation. Regarding the multi-protocol adaptation and lightweight finite element analysis module, communication conditions in wind farms with complex terrain may be poor. Wireless mesh networks or self-organizing networks are employed to ensure reliable data transmission. Simultaneously, lightweight finite element analysis methods are used to reduce data transmission volume and improve real-time performance. Regarding the embedded integration module for the fault prediction model, wind turbine fault modes in wind farms with complex terrain are more diverse. Further optimization of the fault prediction model is needed to adapt to the stress characteristics and fault patterns under complex terrain conditions.

[0050] The above are merely preferred embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for constructing a twin tree structure based on force transmission, characterized in that, include: Collect stress data of wind turbine units; Based on the collected force data, a twin tree structure with force transmission attributes is constructed; Based on the meteorological conditions and geographical environment of the wind farm, the platform layout of the twin tree structure is optimized; Based on the optimized layout, multi-source heterogeneous data is integrated to generate real-time force visualization results; Based on the real-time force visualization results, a fault warning mechanism based on force anomalies is triggered; Based on the aforementioned fault warning mechanism, a user interface is generated and the platform is published.

2. The method according to claim 1, characterized in that, The collected force data of the wind turbine generator further includes: By accessing data from multiple types of sensors, multi-source sensor data can be acquired. Based on the multi-source sensor data, fused force data is generated using a data fusion algorithm; Based on the fused force data, preprocessing and feature extraction are performed to generate processed force data.

3. The method according to claim 1, characterized in that, The construction of the twin tree structure with force transmission properties further includes: Based on the physical structure of the wind turbine, a basic framework for the twin is generated; Based on the force data, a recursive algorithm is used to add force transmission attributes to each component in the basic framework of the twin, generating a twin tree structure with force transmission attributes. Based on the force transmission properties, dynamic propagation of force data is achieved in the twin tree structure.

4. The method according to claim 1, characterized in that, Optimizing the platform layout of the twin tree structure further includes: Based on the wind farm's topography, wind direction, wind speed, and wind turbine arrangement, an intelligent layout algorithm incorporating wake models is used to generate a platform layout scheme. Based on the platform layout scheme, the assembly feasibility is verified, and a verified layout scheme is generated.

5. The method according to claim 1, characterized in that, The integration of multi-source heterogeneous data to generate real-time force visualization results further includes: Data access based on multiple communication protocols is used to acquire heterogeneous data from multiple sources. Based on the multi-source heterogeneous data, seamless integration is performed to generate integrated force data; Based on the integrated stress data, a real-time stress visualization report is generated through lightweight finite element analysis.

6. The method according to claim 1, characterized in that, The fault warning mechanism based on abnormal force further includes: Based on the real-time force visualization results, obtain real-time force data; Based on the real-time stress data, an embedded integrated fault prediction model is used for monitoring to determine whether there is any stress abnormality. Based on the judgment that there is an abnormal force, the fault early warning mechanism is triggered.

7. The method according to claim 1, characterized in that, The process of generating the user interface and completing the platform release further includes: Based on user requirements, generate the user interface using low-code development tools; Based on the generated user interface, the platform can be deployed and updated using the one-click publish function.

8. A twin tree structure construction system based on force transmission, characterized in that, include: The stress data acquisition and processing module is used to acquire stress data of the wind turbine. A twin tree structure construction module is connected to the force data acquisition and processing module, and is used to construct a twin tree structure with force transmission attributes based on the acquired force data; The intelligent layout and assembly verification module is used to optimize the platform layout of the twin tree structure based on the meteorological conditions and geographical environment of the wind farm. The multi-protocol adaptation and lightweight finite element analysis module is used to integrate multi-source heterogeneous data based on the optimized layout and generate real-time force visualization results. An embedded integrated module for fault prediction model is used to trigger a fault early warning mechanism based on force anomalies according to the real-time force visualization results. The low-code UI and one-click deployment module is used to generate a user interface and complete platform deployment based on the fault warning mechanism. The twin tree structure construction module is further used to generate a basic twin framework based on the physical structure of the wind turbine, add force transmission attributes to each component in the basic twin framework using a recursive algorithm based on the force data, and realize the dynamic propagation of force data in the twin tree structure based on the force transmission attributes.

9. An electronic device comprising a memory, a processor, and a computing program stored in the memory and executable on the processor, characterized in that, When the processor executes the computing program, it implements the method of any one of claims 1-7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1-7.

Images