Power transmission and distribution cable fault diagnosis simulation device and method

Abstract

According to the power transmission and distribution cable fault diagnosis simulation device and method, real-time adjustment of fault parameters is achieved through a programmable fault injector, various working conditions such as short circuit, open circuit, grounding, air gap discharge and arc faults can be accurately simulated, the diversity and authenticity of fault simulation are remarkably improved, and the fault diagnosis simulation efficiency is improved. Reliable test data is provided for subsequent diagnosis; the programmable load system dynamically adjusts load parameters, simulates complex load conditions in actual operation, combines an adaptive control function, and automatically adjusts the load according to a real-time feedback mechanism, thereby effectively improving the precision and reliability of fault simulation. A multi-modal sensor is arranged along a cable, a multi-dimensional monitoring system is established, precise positioning of the sensor and spatial continuity of data are realized in combination with a GIS technology, and the comprehensive sensing capability of the cable operation state is remarkably improved through real-time data acquisition and fusion.

Description

Technical Field

[0001] The present invention relates to the technical field of cable fault diagnosis, and in particular to a power transmission and distribution cable fault diagnosis simulation device and method. Background Art

[0002] As an important part of the power system, the operation status of the transmission and distribution cable directly affects the stability and safety of the power supply system. However, during the long-term operation of the cable, it is susceptible to problems such as short circuit, open circuit, grounding, air gap discharge and arc fault due to the influence of various factors such as ambient temperature, humidity, aging and external stress. Most traditional cable fault diagnosis methods rely on single-mode fault detection devices or rule-based diagnosis models. The limitations of this approach are:

[0003] Disadvantages of single detection mode: Traditional sensors can only collect a single physical quantity (such as current or temperature), which cannot realize multi-dimensional monitoring of cable operation status and it is difficult to accurately capture complex fault characteristics;

[0004] Limited data processing capabilities: Existing diagnostic systems are not able to process noise and abnormal data, which can easily lead to false positives or false negatives;

[0005] Insufficient fault simulation accuracy: Existing fault simulation equipment usually only supports fixed parameter fault injection and lacks dynamic adjustment capabilities, making it difficult to truly reflect the complex working conditions in actual operation;

[0006] Limited level of intelligent diagnosis: Most diagnostic systems rely only on simple rules or shallow learning algorithms, which makes it difficult to fully utilize the large amount of data collected by sensors, resulting in insufficient diagnostic accuracy and robustness. Summary of the invention

[0007] In order to solve the above technical problems, the present invention provides a transmission and distribution cable fault diagnosis simulation device and method.

[0008] In order to achieve the above object, the technical solution adopted by the present invention is: a transmission and distribution cable fault diagnosis simulation device, comprising:

[0009] A multi-mode fault simulation module is used to simulate various transmission and distribution cable fault modes and provide fault reference data. The fault modes include short circuit, open circuit, grounding, air gap discharge and arc fault modes. The module uses a programmable fault injector to achieve real-time fault parameter adjustment, including fault resistance, fault current, fault duration and environmental influencing factors;

[0010] The real load simulation module simulates a variety of actual load conditions by dynamically adjusting the programmable load, and has an adaptive control function that automatically adjusts load parameters through a real-time feedback mechanism to obtain more accurate fault diagnosis results;

[0011] Distributed sensors and intelligent network modules deploy multimodal sensors along the cable, including current, pressure, temperature, acoustic emission and fiber optic sensors, to form a multi-dimensional monitoring system that supports real-time data collection, data fusion and analysis to capture the spatiotemporal characteristics of the cable operation status, and assist sensor positioning through the Geographic Information System (GIS) to ensure the spatial continuity and spatiotemporal consistency of data collection;

[0012] Fault feature extraction and self-learning module, which uses sensor data to automatically extract and annotate fault features, uses reinforcement learning algorithms to continuously optimize the feature extraction process, and combines deep learning models to provide precise fault diagnosis support and improve the accuracy and robustness of the model; and

[0013] The real-time monitoring and intelligent diagnosis module includes a data preprocessing unit, an intelligent fault identification unit and a decision support unit. The data preprocessing unit uses double standardization, wavelet transform noise reduction and time series smoothing processing algorithms to remove noise and outliers to ensure data quality. The intelligent fault identification unit uses deep learning technology to determine the fault type and location in real time, and generates a fault diagnosis report based on real-time data analysis.

[0014] Furthermore, the fault feature extraction and self-learning module is based on a multi-layer nonlinear conversion model, and performs multi-layer nonlinear mapping through a piecewise double sigmoid function, thereby enhancing the model's ability to express complex data and improving the stability of data in extreme ranges.

[0015] Furthermore, the multi-mode fault simulation module adopts a dynamic weight calculation mechanism to dynamically adjust the weight of each sensor through a Bayesian network model based on historical data to adapt to changes in cable status and calculate the fault probability to support real-time fault warning.

[0016] Furthermore, the data preprocessing unit in the real-time monitoring and intelligent diagnosis module further includes an anomaly detection and data fusion algorithm, which can effectively remove noise and outliers to improve data quality and ensure diagnostic accuracy.

[0017] Furthermore, the fault feature extraction and self-learning module is jointly modeled using a convolutional neural network (CNN) and a long short-term memory network (LSTM), and a transfer learning module is introduced during the training process to automatically adjust the model weights according to different cable fault types and sensor layout conditions, thereby enhancing the adaptability of the model.

[0018] Furthermore, the intelligent fault identification unit introduces dynamic thresholds in the fault diagnosis process, adjusts them according to the operating status of the system, reduces the false alarm rate, and calculates the fault probability of current data through a fault probability calculation formula combined with historical trends and spatial correlations to generate more accurate real-time warning information.

[0019] Furthermore, the decision support unit analyzes the spatial distribution of sensors through a graph neural network (GNN), and realizes fault type identification and location positioning in combination with the fault probability distribution, thereby providing more accurate fault location support.

[0020] Furthermore, the model optimization and adaptive learning module continuously trains the model through incremental learning and transfer learning, updates the model weights in combination with new fault data to ensure efficient diagnosis under different working conditions, and automatically generates fault mode analysis and historical comparison to further enhance the guidance of the diagnostic report.

[0021] Furthermore, the fault diagnosis report generated by the real-time monitoring and intelligent diagnosis module includes the fault location, type, severity, possible consequences and corresponding processing suggestions, and realizes real-time sharing and collaboration through the cloud platform, supporting multiple parties to view and respond simultaneously.

[0022] The present invention also provides a power transmission and distribution cable fault diagnosis simulation method based on the above device, comprising the following steps:

[0023] S1: In the transmission and distribution cable fault diagnosis simulation device, a multi-mode fault simulation module is configured to set the fault mode, including short circuit, open circuit, grounding, air gap discharge and arc fault; a programmable fault injector is used to adjust the fault parameters in real time, and the fault parameters include fault resistance, fault current, fault duration and environmental influencing factors;

[0024] S2: Configure the real load simulation module to dynamically adjust the load parameters through the programmable load system to simulate a variety of actual load conditions; this module includes an adaptive control function that automatically adjusts the load based on the real-time feedback mechanism to improve the accuracy of fault simulation;

[0025] S3: Distributed multimodal sensors, including current, pressure, temperature, acoustic emission and fiber optic sensors, are deployed along the transmission and distribution cables to establish a multi-dimensional monitoring system; intelligent network modules are configured to support real-time collection, fusion and analysis of sensor data to capture the operating status of the cables, and sensor positioning is achieved through the geographic information system (GIS) to ensure the spatial continuity and consistency of the data;

[0026] S4: Use the fault feature extraction and self-learning module to automatically extract and annotate fault features from sensor data; use the reinforcement learning algorithm to continuously optimize the fault feature extraction process, and combine it with the deep learning model to improve the accuracy and robustness of fault diagnosis;

[0027] S5: It is equipped with real-time monitoring and intelligent diagnosis module, including data preprocessing unit, intelligent fault identification unit and decision support unit; the data preprocessing unit adopts double standardization, wavelet transform noise reduction and time series smoothing processing algorithm to remove noise and outliers to ensure the quality and reliability of data;

[0028] S6: Introducing deep learning technology into the intelligent fault identification unit, through in-depth analysis of historical sensor data, real-time identification of fault type and location; combined with the dynamic threshold adjustment mechanism, adjusting the fault judgment standard according to the real-time operating status of the system, reducing the false alarm rate and improving the accuracy of fault identification;

[0029] S7: Analyze real-time monitoring data through the decision support unit, process the sensor spatial distribution data based on the graph neural network (GNN) technology, and combine it with the fault probability distribution to realize fault type identification and accurate positioning, and form complete fault diagnosis support information;

[0030] S8: The fault feature extraction and self-learning module combines the convolutional neural network (CNN) and the long short-term memory network (LSTM) for joint modeling. The transfer learning method is used to automatically adjust the model parameters according to different cable fault types and sensor layout conditions to improve the adaptability of the diagnosis model in complex environments.

[0031] S9: In the model optimization and adaptive learning module, incremental learning and transfer learning methods are used to continuously train and update the model based on newly collected fault data, automatically generate fault mode analysis and historical comparison results, and ensure the stability of the diagnosis model under different working conditions;

[0032] S10: The real-time monitoring and intelligent diagnosis module generates a fault diagnosis report, including the fault location, type, severity, possible consequences and corresponding handling suggestions, and shares and collaborates in real time through the cloud platform, supporting real-time viewing and response by multiple parties to achieve efficient management and processing of fault information.

[0033] Compared with the prior art, the present invention has the following beneficial effects:

[0034] 1. The present invention realizes real-time adjustment of fault parameters (including fault resistance, fault current, fault duration and environmental influencing factors) through a programmable fault injector, and can accurately simulate various working conditions such as short circuit, open circuit, grounding, air gap discharge and arc fault, significantly improve the diversity and authenticity of fault simulation, and provide reliable test data for subsequent diagnosis.

[0035] 2. The programmable load system of the present invention simulates the complex load conditions in actual operation by dynamically adjusting the load parameters, and automatically adjusts the load according to the real-time feedback mechanism in combination with the adaptive control function, thereby effectively improving the accuracy and reliability of fault simulation.

[0036] 3. The present invention arranges multimodal sensors (such as current, pressure, temperature, acoustic emission and optical fiber sensors) along the cable, establishes a multi-dimensional monitoring system, combines GIS technology to achieve precise positioning of sensors and spatial continuity of data, and significantly improves the comprehensive perception of the cable operation status through real-time data collection and fusion.

[0037] 4. The present invention adopts a reinforcement learning algorithm combined with a deep learning model to automatically extract fault features from multimodal sensor data and optimize the labeling process, significantly improving the accuracy and robustness of fault diagnosis and avoiding the excessive reliance of traditional methods on manual experience.

[0038] 5. The present invention ensures high data quality through double standardization, wavelet transform denoising and time series smoothing. It combines CNN, LSTM and graph neural network (GNN) technologies to dynamically adjust fault judgment criteria, identify fault types and locations in real time, and reduce false alarm rates through a dynamic threshold mechanism to improve diagnostic accuracy.

[0039] 6. The present invention uses incremental learning and transfer learning methods to continuously train and optimize the diagnostic model based on newly collected fault data, ensuring the efficient adaptability and stability of the model under different working conditions and improving the long-term effectiveness of fault diagnosis.

[0040] 7. The fault diagnosis report generated in the present invention not only includes the fault location, type and severity, but also provides corresponding processing suggestions, and realizes real-time sharing among multiple parties through the cloud platform, thereby improving the coordination and management efficiency of fault response. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 This is a schematic diagram of a power transmission and distribution cable fault diagnosis simulation device according to an embodiment of the present invention;

[0042] Figure 2 The present invention is a flowchart of a method for simulating power transmission and distribution cable fault diagnosis according to an embodiment of the present invention. DETAILED DESCRIPTION

[0043] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0044] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present application belongs.

[0045] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0046] like Figure 1 As shown, this embodiment provides a power transmission and distribution cable fault diagnosis simulation device, including: a multi-mode fault simulation module, a real load simulation module, a distributed sensor and intelligent network module, a fault feature extraction and self-learning module and a real-time monitoring and intelligent diagnosis module.

[0047] The multi-mode fault simulation module is used to simulate various transmission and distribution cable fault modes and provide fault reference data. The fault modes include short circuit, open circuit, grounding, air gap discharge and arc fault modes. The module realizes real-time fault parameter adjustment through a programmable fault injector, including fault resistance, fault current, fault duration and environmental influencing factors.

[0048] The real load simulation module simulates a variety of actual load conditions by dynamically adjusting the programmable load and has an adaptive control function to automatically adjust load parameters through a real-time feedback mechanism to obtain more accurate fault diagnosis results.

[0049] The distributed sensors and intelligent network modules deploy multimodal sensors along the cable, including current, pressure, temperature, acoustic emission and optical fiber sensors, to form a multi-dimensional monitoring system that supports real-time acquisition, data fusion and analysis functions to capture the spatiotemporal characteristics of the cable operation status, and assist sensor positioning through the geographic information system (GIS) to ensure the spatial continuity and spatiotemporal consistency of data acquisition.

[0050] The fault feature extraction and self-learning module uses sensor data to automatically extract and annotate fault features, uses a reinforcement learning algorithm to continuously optimize the feature extraction process, and combines a deep learning model to provide accurate fault diagnosis support and improve the accuracy and robustness of the model. The fault feature extraction and self-learning module is based on a multi-layer nonlinear conversion model, and uses a piecewise double sigmoid function to perform multi-layer nonlinear mapping to enhance the model's ability to express complex data and improve the stability of data in extreme intervals; the fault feature extraction and self-learning module uses a convolutional neural network (CNN) and a long short-term memory network (LSTM) for joint modeling, and introduces a transfer learning module during the training process to automatically adjust the model weights according to different cable fault types and sensor layout conditions to enhance the adaptability of the model.

[0051] The real-time monitoring and intelligent diagnosis module includes a data preprocessing unit, an intelligent fault identification unit and a decision support unit. The data preprocessing unit uses double standardization, wavelet transform noise reduction and time series smoothing processing algorithms to remove noise and outliers to ensure data quality. The intelligent fault identification unit uses deep learning technology to determine the fault type and location in real time, and generates a fault diagnosis report based on real-time data analysis. The data preprocessing unit in the real-time monitoring and intelligent diagnosis module further includes anomaly detection and data fusion algorithms, which can effectively remove noise and outliers to improve data quality and ensure diagnostic accuracy; the intelligent fault identification unit introduces dynamic thresholds in the fault diagnosis process, adjusts according to the operating status of the system, reduces the false alarm rate, and calculates the fault probability of the current data through the fault probability calculation formula combined with historical trends and spatial correlation to generate more accurate real-time warning information; the decision support unit analyzes the spatial distribution of sensors through a graph neural network (Graph Neural Network, GNN), combines the fault probability distribution to realize fault type identification and location positioning, and provides more accurate fault location support.

[0052] The multi-mode fault simulation module adopts a dynamic weight calculation mechanism to dynamically adjust the weight of each sensor through a Bayesian network model based on historical data to adapt to cable state changes and calculate the fault probability to support real-time fault warning.

[0053] The model optimization and adaptive learning module continuously trains the model through incremental learning and transfer learning, updates the model weights in combination with new fault data, ensures efficient diagnosis under different working conditions, and automatically generates fault mode analysis and historical comparison, further enhancing the guidance of the diagnostic report.

[0054] The fault diagnosis report generated by the real-time monitoring and intelligent diagnosis module includes the fault location, type, severity, possible consequences and corresponding processing suggestions, and realizes real-time sharing and collaboration through the cloud platform, supporting multiple parties to view and respond at the same time.

[0055] like Figure 2 As shown, this embodiment also provides a power transmission and distribution cable fault diagnosis simulation method based on the above device, comprising the following steps:

[0056] S1: In the power transmission and distribution cable fault diagnosis simulation device, a multi-mode fault simulation module is configured to set the fault mode, including short circuit, open circuit, grounding, air gap discharge and arc fault, etc.; a programmable fault injector is used to realize real-time adjustment of fault parameters, and the fault parameters include fault resistance, fault current, fault duration and environmental influencing factors.

[0057] S2: Configure the real load simulation module to dynamically adjust the load parameters through the programmable load system to simulate a variety of actual load conditions; this module includes an adaptive control function that automatically adjusts the load based on the real-time feedback mechanism to improve the accuracy of fault simulation.

[0058] S3: Distributed multimodal sensors, including current, pressure, temperature, acoustic emission and fiber optic sensors, are deployed along the transmission and distribution cables to establish a multi-dimensional monitoring system. Intelligent network modules are configured to support real-time collection, fusion and analysis of sensor data to capture the operating status of the cables, and sensor positioning is achieved through the geographic information system (GIS) to ensure the spatial continuity and consistency of the data.

[0059] S4: Use the fault feature extraction and self-learning module to automatically extract and annotate fault features from sensor data; use the reinforcement learning algorithm to continuously optimize the fault feature extraction process, combined with the deep learning model to improve the accuracy and robustness of fault diagnosis.

[0060] S5: Equipped with real-time monitoring and intelligent diagnosis module, including data preprocessing unit, intelligent fault identification unit and decision support unit; the data preprocessing unit adopts double standardization, wavelet transform noise reduction and time series smoothing processing algorithm to remove noise and outliers to ensure the quality and reliability of data.

[0061] S6: Introduce deep learning technology into the intelligent fault identification unit to identify the fault type and location in real time through in-depth analysis of historical sensor data; combine with the dynamic threshold adjustment mechanism to adjust the fault judgment standard according to the real-time operating status of the system, reduce the false alarm rate, and improve the accuracy of fault identification.

[0062] S7: Analyze real-time monitoring data through the decision support unit, process the sensor spatial distribution data based on the graph neural network (GNN) technology, and combine it with the fault probability distribution to realize fault type identification and precise positioning, and form complete fault diagnosis support information.

[0063] S8: The fault feature extraction and self-learning module combines the convolutional neural network (CNN) and the long short-term memory network (LSTM) for joint modeling. Using the transfer learning method, the model parameters are automatically adjusted according to different cable fault types and sensor layout conditions to improve the adaptability of the diagnosis model in complex environments.

[0064] S9: In the model optimization and adaptive learning module, incremental learning and transfer learning methods are used to continuously train and update the model based on newly collected fault data, automatically generate fault mode analysis and historical comparison results, and ensure the stability of the diagnostic model under different working conditions.

[0065] S10: The real-time monitoring and intelligent diagnosis module generates a fault diagnosis report, including the fault location, type, severity, possible consequences and corresponding handling suggestions, and shares and collaborates in real time through the cloud platform, supporting real-time viewing and response by multiple parties to achieve efficient management and processing of fault information.

[0066] In order to verify the performance of the power transmission and distribution cable fault diagnosis simulation method of the present invention, the following experimental scheme is designed in this embodiment:

[0067] Experimental equipment preparation:

[0068] Using a fault diagnosis simulation device, including a multi-mode fault simulation module, a real load simulation module, a distributed multi-modal sensor system and an intelligent network module;

[0069] Equipped with high-precision current sensor (accuracy ±0.1A), optical fiber sensor (accuracy ±0.05mm), acoustic emission sensor (sampling rate 1MHz) and other sensors;

[0070] Configure the fault injector to support real-time adjustment of the fault resistance range (0.01Ω-100Ω) and fault duration (0.1ms-10s);

[0071] Use a high-performance computer that supports deep learning and install the TensorFlow and PyTorch frameworks.

[0072] Experimental scene setting:

[0073] Fault mode: simulate three typical faults such as short circuit, ground fault and arc fault;

[0074] Load conditions: The load simulation module dynamically sets three load types (light load 50%, medium load 70%, heavy load 90%);

[0075] Environmental conditions: Set different temperature and humidity environments (25℃ / 60% humidity and 40℃ / 80% humidity).

[0076] Experimental process:

[0077] Steps S1-S3: set the fault mode (e.g., short circuit, fault resistance of 5Ω, duration of 2s), dynamically adjust the load parameters through the real load simulation module, and arrange multi-modal sensors along the transmission and distribution cables to collect data in real time;

[0078] Steps S4-S5: The sensor data is collected and preprocessed in real time through the intelligent network module, using wavelet noise reduction and time series smoothing to remove abnormal values ​​and extract fault features;

[0079] Steps S6-S7: using a joint model based on a convolutional neural network (CNN) and a long short-term memory network (LSTM) to identify the fault type and location, and combining a dynamic threshold adjustment mechanism to optimize the diagnosis accuracy;

[0080] Steps S8-S10: Adopt incremental learning methods to continuously optimize the diagnostic model based on newly collected data, generate a fault report including fault type, location, severity and handling suggestions, and upload it to the cloud platform for real-time sharing.

[0081] The experimental data are shown in the following table:

[0082]

[0083] According to the data in the above table, the power transmission and distribution cable fault diagnosis simulation method of the present invention has significant advantages over the prior art:

[0084] 1. Significantly shortened diagnosis time:

[0085] The diagnosis time of Experiment 1 is 12ms, which is 52% less than the 25ms of the prior art; Experiments 2 and 3 also show that the diagnosis time is reduced by 50% and 54.5% respectively. This advantage mainly comes from the real-time acquisition of multimodal sensor data and the application of deep learning technology in the present invention.

[0086] 2. Significantly improved diagnostic accuracy:

[0087] The diagnostic accuracy of the present invention exceeds 97%, and Experiment 3 reaches 99.2%, which is 5%-8% higher than the prior art. This improvement is due to the joint modeling of convolutional neural network (CNN) and long short-term memory network (LSTM), as well as the optimization of the reinforcement learning algorithm for fault feature extraction.

[0088] 3. False alarm rate is significantly reduced:

[0089] The false alarm rate of the present invention is lower than 1.5%, which is much lower than the prior art (generally above 5%). This is due to the application of the dynamic threshold adjustment mechanism and the time series smoothing algorithm.

[0090] In summary, the experimental data fully proves that the present invention has significant advantages in diagnosis time, accuracy and false alarm rate, embodies its innovation and practicality, and can effectively make up for the shortcomings of the existing technology.

[0091] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0092] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0093] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0094] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0095] The above is only a preferred embodiment of the present invention, and does not limit the present invention in other forms. Any technician familiar with the profession may use the above disclosed technical content to change or modify it into an equivalent embodiment with equivalent changes. However, any simple modification, equivalent change and modification made to the above embodiment according to the technical essence of the present invention without departing from the technical solution of the present invention still belongs to the protection scope of the technical solution of the present invention.

Claims

1. A power transmission and distribution cable fault diagnosis simulation device, characterized in that: include: A multi-mode fault simulation module is used to simulate various transmission and distribution cable fault modes and provide fault reference data. The fault modes include short circuit, open circuit, grounding, air gap discharge and arc fault modes. The module uses a programmable fault injector to achieve real-time fault parameter adjustment, including fault resistance, fault current, fault duration and environmental influencing factors; The real load simulation module simulates a variety of actual load conditions by dynamically adjusting the programmable load, and has an adaptive control function that automatically adjusts load parameters through a real-time feedback mechanism to obtain more accurate fault diagnosis results; Distributed sensors and intelligent network modules deploy multimodal sensors along the cable, including current, pressure, temperature, acoustic emission and fiber optic sensors, to form a multi-dimensional monitoring system that supports real-time data collection, data fusion and analysis functions to capture the spatiotemporal characteristics of the cable operation status, and assist sensor positioning through geographic information systems to ensure the spatial continuity and spatiotemporal consistency of data collection; The fault feature extraction and self-learning module uses sensor data to automatically extract and annotate fault features, uses a reinforcement learning algorithm to continuously optimize the feature extraction process, and combines deep learning models to provide precise fault diagnosis support and improve the accuracy and robustness of the model. as well as The real-time monitoring and intelligent diagnosis module includes a data preprocessing unit, an intelligent fault identification unit and a decision support unit. The data preprocessing unit uses double standardization, wavelet transform noise reduction and time series smoothing processing algorithms to remove noise and outliers to ensure data quality. The intelligent fault identification unit uses deep learning technology to determine the fault type and location in real time, and generates a fault diagnosis report based on real-time data analysis.

2. A power transmission and distribution cable fault diagnosis simulation device according to claim 1, characterized in that: The fault feature extraction and self-learning module is based on a multi-layer nonlinear conversion model, and performs multi-layer nonlinear mapping through a piecewise double sigmoid function, thereby enhancing the model's ability to express complex data and improving the stability of data in extreme ranges.

3. A power transmission and distribution cable fault diagnosis simulation device according to claim 1, characterized in that: The multi-mode fault simulation module adopts a dynamic weight calculation mechanism to dynamically adjust the weight of each sensor through a Bayesian network model based on historical data to adapt to cable state changes and calculate the fault probability to support real-time fault warning.

4. A power transmission and distribution cable fault diagnosis simulation device according to claim 1, characterized in that: The data preprocessing unit in the real-time monitoring and intelligent diagnosis module further includes anomaly detection and data fusion algorithms, which can effectively remove noise and outliers to improve data quality and ensure diagnostic accuracy.

5. A power transmission and distribution cable fault diagnosis simulation device according to claim 1, characterized in that: The fault feature extraction and self-learning module uses convolutional neural network and long short-term memory network to jointly model, and introduces a transfer learning module during the training process to automatically adjust the model weight according to different cable fault types and sensor layout conditions to enhance the adaptability of the model.

6. A power transmission and distribution cable fault diagnosis simulation device according to claim 1, characterized in that: The intelligent fault identification unit introduces a dynamic threshold in the fault diagnosis process, adjusts it according to the operating status of the system, reduces the false alarm rate, and calculates the fault probability of the current data through the fault probability calculation formula combined with historical trends and spatial correlation to generate more accurate real-time warning information.

7. A power transmission and distribution cable fault diagnosis simulation device according to claim 1, characterized in that: The decision support unit analyzes the spatial distribution of sensors through a graph neural network, and realizes fault type identification and location positioning in combination with the fault probability distribution, providing more accurate fault location support.

8. A power transmission and distribution cable fault diagnosis simulation device according to claim 1, characterized in that: The model optimization and adaptive learning module continuously trains the model through incremental learning and transfer learning, updates the model weights in combination with new fault data, ensures efficient diagnosis under different working conditions, and automatically generates fault mode analysis and historical comparison, further enhancing the guidance of the diagnostic report.

9. A power transmission and distribution cable fault diagnosis simulation device according to claim 1, characterized in that: The fault diagnosis report generated by the real-time monitoring and intelligent diagnosis module includes the fault location, type, severity, possible consequences and corresponding processing suggestions, and realizes real-time sharing and collaboration through the cloud platform, supporting multiple parties to view and respond at the same time.

10. A method for simulating fault diagnosis of power transmission and distribution cables based on the device according to any one of claims 1 to 9, characterized in that: The following steps are involved: S1: In the transmission and distribution cable fault diagnosis simulation device, a multi-mode fault simulation module is configured to set the fault mode, including short circuit, open circuit, grounding, air gap discharge and arc fault; a programmable fault injector is used to adjust the fault parameters in real time, and the fault parameters include fault resistance, fault current, fault duration and environmental influencing factors; S2: Configure the real load simulation module to dynamically adjust the load parameters through the programmable load system to simulate a variety of actual load conditions; this module includes an adaptive control function that automatically adjusts the load based on the real-time feedback mechanism to improve the accuracy of fault simulation; S3: Distributed multimodal sensors, including current, pressure, temperature, acoustic emission and fiber optic sensors, are deployed along the transmission and distribution cables to establish a multi-dimensional monitoring system; intelligent network modules are configured to support real-time collection, fusion and analysis of sensor data to capture the operating status of the cables, and sensor positioning is achieved through geographic information systems to ensure spatial continuity and consistency of data; S4: Automatically extract and annotate fault features from sensor data using the fault feature extraction and self-learning module; Adopt reinforcement learning algorithm to continuously optimize the fault feature extraction process, and combine it with deep learning model to improve the accuracy and robustness of fault diagnosis; S5: It is equipped with real-time monitoring and intelligent diagnosis module, including data preprocessing unit, intelligent fault identification unit and decision support unit; the data preprocessing unit adopts double standardization, wavelet transform noise reduction and time series smoothing processing algorithm to remove noise and outliers to ensure the quality and reliability of data; S6: Introducing deep learning technology into the intelligent fault identification unit, through in-depth analysis of historical sensor data, real-time identification of fault type and location; combined with the dynamic threshold adjustment mechanism, adjusting the fault judgment standard according to the real-time operating status of the system, reducing the false alarm rate and improving the accuracy of fault identification; S7: Analyze real-time monitoring data through the decision support unit, process the sensor spatial distribution data based on the graph neural network technology, and combine it with the fault probability distribution to realize fault type identification and accurate positioning, and form complete fault diagnosis support information; S8: The fault feature extraction and self-learning module combines convolutional neural networks and long short-term memory networks for joint modeling. It uses transfer learning methods to automatically adjust model parameters according to different cable fault types and sensor layout conditions to improve the adaptability of the diagnostic model in complex environments. S9: In the model optimization and adaptive learning module, incremental learning and transfer learning methods are used to continuously train and update the model based on newly collected fault data, automatically generate fault mode analysis and historical comparison results, and ensure the stability of the diagnosis model under different working conditions; S10: The real-time monitoring and intelligent diagnosis module generates a fault diagnosis report, including the fault location, type, severity, possible consequences and corresponding handling suggestions, and shares and collaborates in real time through the cloud platform, supporting real-time viewing and response by multiple parties to achieve efficient management and processing of fault information.

Images