Power cable fault early warning system and method based on large model

Abstract

The present invention discloses a power cable fault early warning system based on a large model, including: a data acquisition module, which is used to collect multi-dimensional operating parameters of power cables in real time, a data preprocessing module, which uses advanced data cleaning algorithms and signal processing technologies to perform denoising, outlier detection and processing, and normalization operations on the collected data to obtain high-quality preprocessed data, a data transmission module, which uses a secure transmission protocol to transmit the preprocessed data to a data analysis module, and dynamically adjusts the transmission mode according to network conditions and data importance, and a data analysis module, which uses a pre-trained deep large model to analyze and process the received data. The present invention improves data quality, early warning accuracy and reliability through designs such as multi-dimensional data acquisition, advanced preprocessing, multiple communication transmissions, intelligent large models, hierarchical early warnings and distributed databases, ensures timely processing of important faults, and protects the safety of the power system.

Description

Technical Field

[0001] The present invention relates to the technical field of power cable fault early warning, and in particular to a power cable fault early warning system and method based on a large model. Background Art

[0002] As the demand for electricity in modern society continues to grow, the stable operation of the power system is of vital importance. As a key component of power transmission, the reliability of power cables directly affects the safety of the entire power system. However, since power cables are usually buried underground or laid in hidden environments such as cable trenches, once a failure occurs, it is not only difficult to troubleshoot and repair, but may also cause large-scale power outages, bringing serious impacts to production and life.

[0003] Traditional power cable fault detection methods mainly rely on manual inspections and regular testing. Manual inspections are inefficient and difficult to detect potential faults. Although regular testing can detect cable performance to a certain extent, it cannot monitor the operating status of the cable in real time. Often, detection and repair can only be performed after the fault has occurred. In addition, traditional methods are difficult to accurately judge some early minor faults, which can easily cause the fault to deteriorate further.

[0004] In recent years, with the continuous advancement of science and technology, some new technical means have been applied to power cable fault detection. For example, partial discharge detection technology can determine the insulation condition of the cable by detecting partial discharge phenomena inside the cable, but this method often requires professional detection equipment and technicians, the cost is high, and the detection results are easily affected by external interference. There are also some sensor-based monitoring systems, which can collect some operating parameters of the cable in real time, but the data processing and analysis capabilities are limited, making it difficult to accurately evaluate the complex cable operation status and fault warning.

[0005] With the development of artificial intelligence and big data technology, new ideas and methods are provided for power cable fault warning. As a powerful data analysis tool, the big model has the ability to process large-scale data, automatically learn features and make accurate predictions. By building a power cable fault warning system based on a big model, we can make full use of the multi-dimensional operating parameters and environmental data of the power cable, monitor and analyze the operating status of the cable in real time, discover potential fault risks in time, and issue accurate warning signals so that relevant personnel can take timely and effective measures for maintenance and repair.

[0006] In this context, the development of a large-model-based power cable fault warning system and method has important practical significance. The system can overcome the shortcomings of traditional detection methods, improve the accuracy and timeliness of fault warning, and provide strong support for ensuring the stable operation of the power system. At the same time, with the continuous development of smart grids, higher requirements are put forward for the intelligent monitoring and management of power equipment. The fault warning system based on large models will also become an important part of the future smart grid construction. Summary of the invention

[0007] The present invention proposes a large model-based power cable fault early warning system and method to solve the problems mentioned in the above-mentioned prior art.

[0008] In order to achieve the above object, the present invention adopts the following technical solution: a power cable fault early warning system based on a large model, comprising:

[0009] Data acquisition module, used to collect multi-dimensional operating parameters of power cables in real time, including but not limited to current ,Voltage ,temperature , Partial discharge data and the cable's surrounding environment parameters, including humidity , magnetic field strength The collected data can be represented as a vector . The intelligent sensing technology can automatically adjust the collection frequency and accuracy according to the cable operation status to ensure the accuracy and timeliness of the data.

[0010] Assume the cable operation status parameter set ,in For different state parameters, including cable load size and ambient temperature change rate.

[0011] The acquisition frequency adjustment function is: ,in is the weight coefficient corresponding to each state parameter, satisfying . The meaning of this formula is to perform a weighted sum based on the importance of different state parameters, and then take its reciprocal as the acquisition frequency. For example, if a state parameter has a greater impact on cable faults, then the corresponding weight coefficient will be larger, thereby reducing the acquisition frequency to obtain data more frequently. Here, represents the acquisition frequency adjustment function, is a set of cable operation status parameters, Indicates The weight coefficient of each state parameter, It is state parameters.

[0012] The precision adjustment function is ,in and is the adjustment factor. Here Represents the natural exponential function. The formula adjusts the accuracy by weighted summation of state parameters. When the state parameter value is large, the accuracy will decrease accordingly. is the precision adjustment function, is a set of cable operation status parameters, It is state parameters.

[0013] The data preprocessing module uses advanced data cleaning algorithms and signal processing techniques to perform denoising, outlier detection and processing, and normalization operations on the collected data to obtain high-quality preprocessed data.

[0014] Assume the original data is , the denoising function is ,in , using the sliding average method to denoise, is the window size. This formula means that for each data point, the average value of the data points before and after it is taken as the denoised value. Here, is the denoising function, is the original data set, The first data points, is the denoised data point, Is an index variable used to represent the summation range, which is used to determine the position offset of the original data points involved in the summation when calculating the sliding average. Specifically: When changing from -r to r, When These data points are summed and then the average is taken as the denoised value.

[0015] The outlier processing function is ,in , and are the lower quartile and the upper quartile, respectively. , is the outlier judgment coefficient. This formula determines whether it is an outlier by comparing the denoised data with the quartile range. If it is within the range, the original data is retained, otherwise it is replaced by the midpoint of the quartile. Here, is the outlier processing function, is the original data set, is the data point after outlier processing. is the denoised data point, is the interquartile range.

[0016] The normalization function is ,in This formula scales the processed data set to the interval [0,1] to facilitate subsequent data analysis. Here, is the normalization function, is the original data set, is the data point after outlier processing. is the minimum value in the processed data set, is the maximum value in the processed data set.

[0017] This module also innovatively introduces data fusion strategy to integrate multi-source data. The data fusion formula can be expressed as ,in For the The data vector of the data source, For the The fusion weight coefficient of the data sources satisfies , is the fused data vector, and n is the total number of data sources. This formula represents the weighted summation based on the importance of different data sources to obtain the fused data vector.

[0018] The data transmission module uses an efficient and secure transmission protocol to quickly and stably transmit the pre-processed data to the data analysis module. The module has adaptive transmission capabilities and can dynamically adjust the transmission method according to network conditions and data importance to ensure the reliability of data transmission.

[0019] Set network status parameter set , such as network bandwidth, delay parameters, and data importance parameters are .

[0020] The transmission rate function is ,in and is the adjustment coefficient. This formula indicates that the transmission rate is related to the network status parameters and the importance of the data. When the network status is poor, the transmission rate will decrease, and the higher the importance of the data, the higher the transmission rate will increase accordingly. Here, is the transmission rate function, is a set of network status parameters, It is network status parameters.

[0021] The data analysis module uses a pre-trained deep model to analyze and process the received data. The model integrates multiple deep learning architectures, such as convolutional neural networks, recurrent neural networks, and long short-term memory networks, and innovatively introduces attention mechanisms and reinforcement learning algorithms. Processing, failure risk probability of output power cable .

[0022] Assume that the large model function is ,in are model parameters, is the weight matrix, is the bias vector, is the activation function, and through training The value of accurately reflects the failure risk probability. Here, is a large model function, is the set of model parameters.

[0023] Early warning module, when the data analysis module determines the probability of failure risk Exceeding the set threshold When a fault occurs, an early warning signal is issued. Early warning methods include sound and light alarms, remote communication units, and mobile terminal push notifications to ensure that relevant personnel can take timely measures. The module also has an intelligent early warning grading system that provides different levels of early warning according to the severity of the fault risk.

[0024] Assume the warning level function is ,in is the threshold value of different warning levels. This formula divides the warning into three levels according to the size of the failure risk probability. Here, is the warning level function, is the failure risk probability, and are the thresholds for different warning levels.

[0025] The database module is used to store the collected data, analysis results and warning records. The database uses distributed storage and backup technology to ensure the security and reliability of the data.

[0026] Furthermore, the data acquisition module includes a high-precision current sensor, a voltage sensor, a temperature sensor, a partial discharge monitor and an environmental parameter sensor, which can accurately obtain multi-dimensional operation information of the power cable.

[0027] Furthermore, the data preprocessing module uses advanced signal processing algorithms and data cleaning techniques to remove outliers and noise interference and improve data quality.

[0028] Furthermore, the data transmission module supports multiple communication modes, including wired network, wireless network and satellite communication, to ensure the stability of data transmission in different environments.

[0029] Furthermore, the large model has self-learning and self-adaptation capabilities, and can continuously optimize and adjust model parameters according to new data to improve the accuracy of fault warning.

[0030] Furthermore, the early warning module can set different early warning modes and priorities according to the level of fault risk to ensure that important faults are handled in a timely manner.

[0031] Furthermore, the database module adopts distributed storage and backup technology to ensure the security and reliability of data.

[0032] Furthermore, a power cable fault early warning method based on a large model comprises the following steps:

[0033] The data acquisition module collects the operating parameters of the power cable and the surrounding environment parameters in real time to obtain the data vector .

[0034] The data preprocessing module is used to clean, denoise and normalize the collected data to obtain .

[0035] The data transmission module is used to transmit the preprocessed data to the data analysis module.

[0036] The data analysis module uses a pre-trained large model to analyze and process the data, that is, the input , get the failure risk probability .

[0037] like Exceeding the set threshold , the early warning module sends out corresponding early warning signals.

[0038] The collected data, analysis results and warning records are stored in the database module for subsequent query and analysis.

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] The present invention uses a data acquisition module to collect multi-dimensional operating parameters of power cables and surrounding environmental parameters in real time, including current, voltage, temperature, partial discharge data, humidity, magnetic field strength, etc., to fully grasp the operating status of the cable and provide a rich data basis for accurate early warning. At the same time, the intelligent sensor can automatically adjust the acquisition frequency and accuracy according to the operating status of the cable to ensure the accuracy and timeliness of the data.

[0041] Through the data preprocessing module, advanced data cleaning algorithms, signal processing technology and data fusion strategies are used to denoise, detect and process outliers, and normalize the collected data. Multi-source data is effectively integrated to improve data quality and comprehensiveness, providing reliable input for subsequent analysis.

[0042] The data transmission module uses an efficient and secure transmission protocol and has adaptive transmission capabilities. It can dynamically adjust the transmission method according to network conditions and data importance to ensure the reliability of data transmission, so that the pre-processed data can be quickly and stably transmitted to the data analysis module.

[0043] The data analysis module uses a pre-trained deep large model, integrates multiple deep learning architectures, and introduces attention mechanisms and reinforcement learning algorithms to analyze and process the received data and accurately output the fault risk probability of the power cable. The large model has powerful data analysis and learning capabilities, and can continuously optimize and adjust model parameters to improve the accuracy of fault warnings.

[0044] The early warning module conducts intelligent graded early warning according to the probability of fault risk. When it is judged that the probability of fault risk exceeds the set threshold, it sends out early warning signals through sound and light alarms, remote communication units, and mobile terminal push, ensuring that relevant personnel can take timely measures to effectively prevent the occurrence and expansion of faults.

[0045] The database module uses distributed storage and backup technology to store collected data, analysis results and warning records to ensure data security and reliability. It also supports big data analysis and mining, providing data support for system optimization and improvement. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 A schematic block diagram of a power cable fault early warning system based on a large model proposed by the present invention;

[0047] Figure 2 This is a schematic block diagram of a large model-based power cable fault early warning method proposed by the present invention. DETAILED DESCRIPTION

[0048] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0049] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise" and "counterclockwise" indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as limiting the present invention.

[0050] In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined. In addition, the terms "installed", "connected" and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, and it can be the internal connection of two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances. The present invention will be further described in detail below in conjunction with the accompanying drawings.

[0051] Reference Figure 1-2 :A power cable fault early warning system based on a large model, comprising:

[0052] Data acquisition module, used to collect multi-dimensional operating parameters of power cables in real time, including but not limited to current ,Voltage ,temperature , Partial discharge data and the cable's surrounding environment parameters, including humidity , magnetic field strength The collected data can be represented as a vector . The intelligent sensing technology can automatically adjust the collection frequency and accuracy according to the cable operation status to ensure the accuracy and timeliness of the data.

[0053] Assume the cable operation status parameter set ,in For different state parameters, including cable load size and ambient temperature change rate.

[0054] The acquisition frequency adjustment function is: ,in is the weight coefficient corresponding to each state parameter, satisfying . The meaning of this formula is to perform a weighted sum based on the importance of different state parameters, and then take its reciprocal as the acquisition frequency. For example, if a state parameter has a greater impact on cable faults, then the corresponding weight coefficient will be larger, thereby reducing the acquisition frequency to obtain data more frequently. Here, represents the acquisition frequency adjustment function, is a set of cable operation status parameters, Indicates The weight coefficient of each state parameter, It is state parameters.

[0055] The precision adjustment function is ,in and is the adjustment factor. Here Represents the natural exponential function. The formula adjusts the accuracy by weighted summation of state parameters. When the state parameter value is large, the accuracy will decrease accordingly. is the precision adjustment function, is a set of cable operation status parameters, It is state parameters.

[0056] The data preprocessing module uses advanced data cleaning algorithms and signal processing techniques to perform denoising, outlier detection and processing, and normalization operations on the collected data to obtain high-quality preprocessed data.

[0057] Assume the original data is , the denoising function is ,in , using the sliding average method to denoise, is the window size. This formula means that for each data point, the average value of the data points before and after it is taken as the denoised value. Here, is the denoising function, is the original data set, The original data is data points, is the denoised data point, Is an index variable used to represent the summation range, which is used to determine the position offset of the original data points involved in the summation when calculating the sliding average. Specifically: When changing from -r to r, When These data points are summed and then the average is taken as the denoised value.

[0058] The outlier processing function is ,in , and are the lower quartile and the upper quartile, respectively. , is the outlier judgment coefficient. This formula determines whether it is an outlier by comparing the denoised data with the quartile range. If it is within the range, the original data is retained, otherwise it is replaced by the midpoint of the quartile. Here, is the outlier processing function, is the original data set, is the data point after outlier processing. is the denoised data point, is the interquartile range.

[0059] The normalization function is ,in This formula scales the processed data set to the interval [0,1] to facilitate subsequent data analysis. Here, is the normalization function, is the original data set, is the data point after outlier processing. is the minimum value in the processed data set, is the maximum value in the processed data set.

[0060] This module also innovatively introduces data fusion strategy to effectively integrate multi-source data and improve the comprehensiveness and value of data. After integrating multi-source data, the data fusion formula can be expressed as ,in For the The data vector of the data source, For the The fusion weight coefficient of the data sources satisfies , is the fused data vector, and n is the total number of data sources. This formula represents the weighted summation based on the importance of different data sources to obtain the fused data vector.

[0061] The data transmission module uses an efficient and secure transmission protocol to quickly and stably transmit the pre-processed data to the data analysis module. The module has adaptive transmission capabilities and can dynamically adjust the transmission method according to network conditions and data importance to ensure the reliability of data transmission.

[0062] Set network status parameter set , such as network bandwidth, delay parameters, and data importance parameters are .

[0063] The transmission rate function is ,in and is the adjustment coefficient. This formula indicates that the transmission rate is related to the network status parameters and the importance of the data. When the network status is poor, the transmission rate will decrease, and the higher the importance of the data, the higher the transmission rate will increase accordingly. Here, is the transmission rate function, is a set of network status parameters, It is network status parameters.

[0064] The data analysis module uses a pre-trained deep model to analyze and process the received data. The model integrates multiple deep learning architectures, such as convolutional neural networks, recurrent neural networks, and long short-term memory networks, and innovatively introduces attention mechanisms and reinforcement learning algorithms. Processing, failure risk probability of output power cable .

[0065] Assume that the large model function is ,in are model parameters, is the weight matrix, is the bias vector, is the activation function, and through training The value of accurately reflects the failure risk probability. Here, is a large model function, is the set of model parameters.

[0066] Early warning module, when the data analysis module determines the probability of failure risk Exceeding the set threshold When a fault occurs, an early warning signal is issued. Early warning methods include sound and light alarms, remote communication units, and mobile terminal push notifications to ensure that relevant personnel can take timely measures. The module also has an intelligent early warning grading system that provides different levels of early warning according to the severity of the fault risk.

[0067] Assume the warning level function is ,in is the threshold value of different warning levels. This formula divides the warning into three levels according to the size of the failure risk probability. Here, is the warning level function, is the failure risk probability, and are the thresholds for different warning levels.

[0068] The database module is used to store the collected data, analysis results and warning records. The database uses distributed storage and backup technology to ensure the security and reliability of the data.

[0069] In the present invention, the data acquisition module plays a vital role, and includes a variety of advanced sensing devices such as high-precision current sensors, voltage sensors, temperature sensors, partial discharge monitors, and environmental parameter sensors.

[0070] Among them, high-precision current sensors can accurately detect the size and changes of current in power cables. Through real-time monitoring of current, abnormal current fluctuations, such as overcurrent, short circuit and other potential fault signs, can be discovered in time. With its high sensitivity and accuracy, current sensors provide the system with key current data and provide an important basis for fault warning.

[0071] The voltage sensor is responsible for accurately measuring the voltage value of the power cable. Voltage stability is crucial to the normal operation of the power system. The voltage sensor can quickly capture voltage changes, and abnormal voltage conditions, whether too high or too low, can be monitored in time. This helps to detect cable faults that may be caused by voltage problems in advance, and provides strong support for maintaining the safe and stable operation of the power system.

[0072] Temperature sensors also play an indispensable role in the system. Power cables generate heat during operation. Excessive temperature may damage the cable insulation, leading to failures. Temperature sensors can monitor the temperature changes of the cable in real time. Once the temperature exceeds the normal range, the system can immediately issue an early warning. Its high-precision measurement capability ensures accurate control of the cable temperature and effectively prevents failures caused by excessive temperature.

[0073] The partial discharge monitor is specially used to detect partial discharge inside power cables. Partial discharge is one of the early signs of cable insulation degradation. The monitor can sensitively capture partial discharge signals and analyze and judge them. By monitoring partial discharge, timely measures can be taken before cable faults occur to avoid further deterioration of the fault.

[0074] In addition, environmental parameter sensors can accurately obtain environmental information around power cables, such as humidity, magnetic field strength, etc. Changes in humidity may affect the insulation performance of the cable, while abnormal magnetic field strength may indicate external interference or internal faults in the cable. These environmental parameter sensors provide the system with comprehensive cable operating environment information, which helps to more accurately evaluate the operating status of the cable and improve the accuracy of fault warning.

[0075] In summary, the data acquisition module of the present invention can accurately obtain multi-dimensional operation information of the power cable through these high-precision sensors, providing a solid data foundation for subsequent fault warning and analysis.

[0076] In the present invention, the data preprocessing module plays a key role. It innovatively adopts advanced signal processing algorithms and data cleaning technologies, laying a solid foundation for subsequent data analysis and fault warning.

[0077] The advanced signal processing algorithms in the data preprocessing module are like a precise craftsman, finely carving the collected raw data. These algorithms can identify and analyze various signal features in the data, extract valuable information, and remove irrelevant signals that may interfere with the analysis results. For example, through the filtering algorithm, high-frequency noise and low-frequency interference can be effectively removed, making the data clearer and more stable. The spectrum analysis algorithm can reveal the frequency components in the data and help identify potential abnormal signal patterns. Through these advanced signal processing algorithms, the quality of the data has been significantly improved, providing a more reliable input for subsequent analysis.

[0078] At the same time, data cleaning technology also plays a vital role in the data preprocessing module. It is like a careful cleaner, dedicated to removing outliers and noise interference in the data. Outliers may be caused by sensor failure, measurement errors or other abnormal conditions, which will seriously affect the accuracy of data analysis. Data cleaning technology can accurately identify these outliers and use reasonable methods to deal with them, such as replacement, interpolation or deletion. For noise interference, data cleaning technology can effectively reduce the noise level and make the data purer through smoothing, denoising algorithms and other means. By removing outliers and noise interference, the accuracy and reliability of the data are greatly improved.

[0079] By adopting advanced signal processing algorithms and data cleaning technology, the data preprocessing module can significantly improve the data quality. High-quality data not only helps to more accurately analyze the operating status of the power cable, but also improves the accuracy and reliability of fault warning. This enables the power cable fault warning system of the present invention to detect potential faults more timely and effectively, providing strong support for ensuring the safe and stable operation of the power system.

[0080] In the present invention, the data transmission module shows strong adaptability and reliability. It actively supports multiple communication modes, including wired network, wireless network and satellite communication. This design decision provides a solid guarantee for the data transmission stability of the system in various complex environments.

[0081] As a traditional and stable communication method, wired network has the advantages of fast transmission speed and high reliability. In an environment with relatively complete infrastructure, the data transmission module can make full use of the characteristics of wired network to ensure that data can be transmitted in an efficient and stable manner. Whether inside the power control center or at some fixed monitoring sites, the wired network can provide a reliable channel for data transmission and ensure the integrity and accuracy of the data.

[0082] Wireless networks bring greater flexibility and convenience to the system. In some areas where it is difficult to lay wired networks or in scenarios where mobile monitoring is required, wireless networks play an irreplaceable role. It allows seamless connection between data acquisition equipment and analysis systems without being restricted by physical lines. Whether it is different types of wireless network technologies such as Wi-Fi, Bluetooth or cellular networks, the data transmission module can intelligently select and switch according to actual conditions to ensure stable data transmission in different wireless environments.

[0083] The addition of satellite communications provides a strong guarantee for data transmission in extreme environments. In remote areas, mountainous areas, oceans and other places where wired and wireless networks are difficult to cover, satellite communications can ensure timely data transmission. Even in natural disasters or other emergencies, when ground communication facilities are destroyed, satellite communications can serve as a last resort to ensure the normal operation of the power cable fault warning system.

[0084] By supporting multiple communication modes, the data transmission module can adapt to different environmental requirements, whether between high-rise buildings in the city or in the wilderness, it can ensure the stability of data transmission. This enables the power cable fault early warning system of the present invention to operate reliably in various complex environments, providing strong support for the safety and stability of the power system.

[0085] In the present invention, the large model exhibits excellent intelligent characteristics. It has strong self-learning and self-adaptive capabilities, and this unique advantage provides a key guarantee for continuously improving the accuracy of fault warning.

[0086] The self-learning ability of the big model is like a diligent scholar, constantly drawing knowledge and experience from new data. As the power cable system operates, a steady stream of new data is constantly generated, and the big model can actively analyze and learn from this data. It can identify potential patterns and rules in the data, and constantly update and enrich its own knowledge system. Through in-depth learning of new data, the big model can better understand the operating status and fault characteristics of the power cable, so as to provide more accurate fault warning.

[0087] At the same time, the adaptive ability of the large model enables it to flexibly respond to various changes. The operating environment of the power cable system may change over time, such as load changes, ambient temperature fluctuations, external interference, etc. The large model can adjust its model parameters in time according to these changes to adapt to new situations. It can automatically adjust the weight distribution of different parameters, optimize the performance of the algorithm, and ensure that high-accuracy fault warnings can be maintained under different operating conditions.

[0088] When new data indicates that the previous warning is biased, the big model will quickly start the self-adjustment mechanism. It will re-evaluate the model parameters, correct the prediction results, and continuously improve them in subsequent analysis. This continuous optimization and adjustment process enables the big model to always maintain the best state and continuously improve the accuracy of fault warnings.

[0089] In summary, the large model in the present invention can continuously optimize and adjust model parameters according to new data by virtue of its self-learning and adaptive capabilities, providing highly accurate and reliable analysis results for the power cable fault early warning system, and effectively ensuring the safe and stable operation of the power system.

[0090] In the present invention, the early warning module demonstrates a high degree of intelligence and flexibility. It can finely set different early warning modes and priorities according to the level of fault risk. This important feature provides a strong guarantee for ensuring that important faults are handled in a timely manner.

[0091] The early warning module first accurately assesses and classifies the fault risk. Through in-depth analysis of power cable operation data, combined with advanced algorithms and models, the early warning module can classify fault risks into different levels. Low-level risks may indicate that potential problems are just beginning to emerge, while high-level risks may mean that a fault is about to occur or is already in a more serious state. This accurate risk classification provides the basis for subsequent targeted early warnings.

[0092] For different levels of failure risks, the early warning module has carefully set up a variety of early warning methods. In the case of low-level risks, a milder early warning method may be used, such as system pop-up prompts, sending short email notifications, etc. This can attract the attention of relevant personnel without causing excessive tension and interference. When the risk level increases, the early warning method will become more eye-catching and intense. For example, the sound and light alarm will emit a harsh alarm and flashing lights to attract the attention of people around; at the same time, the remote communication unit will send an emergency text message or push notification to the mobile phone of the relevant person in charge to ensure that they can learn about the failure at the first time. For particularly serious high-level risks, the early warning module can also start automatic emergency procedures, such as cutting off part of the power supply, starting the backup system, etc., to minimize the losses caused by the failure.

[0093] In addition, the early warning module sets clear priorities based on the level of fault risk. Important high-level faults are given the highest priority to ensure that relevant personnel can focus on handling these emergencies at the first time. For low-level risks, although the urgency of handling is relatively low, they will also be recorded and tracked for further inspection and handling when appropriate. By setting priorities reasonably, the early warning module can effectively allocate resources to ensure that important faults are handled promptly and effectively, while not ignoring potential problems.

[0094] In summary, the early warning module in the present invention provides an efficient and accurate early warning mechanism for the power cable fault early warning system by setting different early warning modes and priorities according to the level of fault risk, which effectively ensures the safe and stable operation of the power system.

[0095] In the present invention, the design of the database module fully considers the security and reliability of data, and innovatively adopts distributed storage and backup technology.

[0096] Distributed storage technology provides a strong guarantee for the secure storage of data. It stores data in multiple physical nodes in a dispersed manner, avoiding the single point failure risk brought by centralized data storage. Even if a node fails, the data on other nodes can still be accessed and used normally, without affecting the operation of the entire system. This decentralized storage method greatly improves the availability and risk resistance of data. At the same time, distributed storage can be dynamically adjusted according to the importance and access frequency of data, optimize the storage layout of data, and improve storage efficiency.

[0097] Backup technology further enhances data reliability. The database module regularly backs up data and copies it to multiple different storage locations, including local storage devices and remote servers. In this way, even in the event of unexpected situations such as hardware failure, natural disasters or human errors, the integrity and availability of data can be guaranteed by restoring the backup data. The backup process is automated to ensure that data is backed up in a timely and accurate manner. At the same time, the backup data is encrypted to prevent data leakage and malicious tampering.

[0098] By adopting distributed storage and backup technology, the database module can effectively ensure the security and reliability of data. Whether in daily operation or in the face of various emergencies, it can provide stable and reliable data support for the power cable fault early warning system, providing a solid foundation for the system's accurate analysis and timely early warning.

[0099] In the present invention, a power cable fault early warning method based on a large model comprises the following steps:

[0100] The data acquisition module collects the operating parameters of the power cable and the surrounding environment parameters in real time to obtain the data vector .

[0101] The data preprocessing module is used to clean, denoise and normalize the collected data to obtain .

[0102] The data transmission module is used to transmit the preprocessed data to the data analysis module.

[0103] The data analysis module uses a pre-trained large model to analyze and process the data, that is, the input , get the failure risk probability .

[0104] like Exceeding the set threshold , the early warning module sends out corresponding early warning signals.

[0105] The collected data, analysis results and warning records are stored in the database module for subsequent query and analysis.

[0106] The above are only preferred specific implementation modes of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can replace or change the technical solution and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A power cable fault early warning system based on a large model, characterized in that: include: Data acquisition module, used to collect multi-dimensional operating parameters of power cables in real time, including current ,Voltage ,temperature , Partial discharge data and the cable's surrounding environment parameters, including humidity , magnetic field strength , the collected data can be represented as a vector , using intelligent sensing technology to automatically adjust the acquisition frequency and accuracy according to the cable operation status; Assume the cable operation status parameter set ,in For different state parameters, including cable load size and ambient temperature change rate; The acquisition frequency adjustment function is: ,in is the weight coefficient corresponding to each state parameter, satisfying , represents the acquisition frequency adjustment function, is a set of cable operation status parameters, Indicates The weight coefficient of each state parameter, It is state parameters; The precision adjustment function is ,in and is the adjustment factor, here represents the natural exponential function, and the precision is adjusted by the weighted sum of the state parameters, is the precision adjustment function, is a set of cable operation status parameters, It is state parameters; The data preprocessing module uses data cleaning algorithms and signal processing technology to perform denoising, outlier detection and processing, and normalization operations on the collected data to obtain high-quality preprocessed data; Assume the original data is , the denoising function is ,in , using the sliding average method to denoise, is the window size, is the denoising function, is the original data set, The original data is data points, is the denoised data point; The outlier processing function is ,in , and are the lower quartile and the upper quartile, respectively. , is the outlier judgment coefficient, is the outlier processing function, is the original data set, is the data point after outlier processing. is the denoised data point, is the interquartile range; The normalization function is ,in , the processed data set is scaled to the interval [0,1], is the normalization function, is the original data set, is the data point after outlier processing. is the minimum value in the processed data set, is the maximum value in the processed data set; Introducing data fusion strategy to integrate multi-source data, the data fusion formula can be expressed as ,in For the The data vector of the data source, For the The fusion weight coefficient of the data sources satisfies , is the fused data vector, n is the total number of data sources; The data transmission module uses a secure transmission protocol to transmit the pre-processed data to the data analysis module, and dynamically adjusts the transmission mode according to the network status and data importance; Set network status parameter set , including network bandwidth, delay parameters, and data importance parameters are ; The transmission rate function is ,in and is the adjustment factor, is the transmission rate function, is a set of network status parameters, It is Network status parameters; The data analysis module uses a pre-trained deep model to analyze and process the received data. The large model integrates multiple deep learning architectures, including convolutional neural networks, recurrent neural networks, and long short-term memory networks, and introduces attention mechanisms and reinforcement learning algorithms. Processing, failure risk probability of output power cable ; Assume that the large model function is ,in are model parameters, is the weight matrix, is the bias vector, is the activation function, and through training The value of accurately reflects the probability of failure risk. is a large model function, is the set of model parameters; Early warning module, when the data analysis module determines the probability of failure risk Exceeding the set threshold When a fault occurs, an early warning signal is issued. The early warning methods include sound and light alarm, remote communication unit and mobile terminal push. There is also an intelligent early warning classification system to issue different levels of early warning according to the severity of the fault risk. Assume the warning level function is ,in are the thresholds for different warning levels. The warnings are divided into three levels according to the probability of failure risk. is the warning level function, is the failure risk probability, and are the thresholds for different warning levels; The database module is used to store the collected data, analysis results and warning records. The database adopts distributed storage and backup technology.

2. A large model-based power cable fault early warning system according to claim 1, characterized in that: The data acquisition module includes a high-precision current sensor, a voltage sensor, a temperature sensor, a partial discharge monitor and an environmental parameter sensor, which are used to obtain multi-dimensional operation information of the power cable.

3. The large model-based power cable fault early warning system according to claim 1 is characterized in that: The data preprocessing module uses signal processing algorithms and data cleaning techniques to remove outliers and noise interference.

4. The large model-based power cable fault early warning system according to claim 1 is characterized in that: The data transmission module supports multiple communication modes, including wired network, wireless network and satellite communication.

5. The large model-based power cable fault early warning system according to claim 1 is characterized in that: The large model has self-learning and self-adaptation capabilities, and continuously optimizes and adjusts model parameters according to new data.

6. The large model-based power cable fault early warning system according to claim 1 is characterized in that: The early warning module sets different early warning modes and priorities according to the level of failure risk.

7. The large model-based power cable fault early warning system according to claim 1 is characterized in that: The database module adopts distributed storage and backup technology.

8. A method for applying the large model-based power cable fault early warning system according to any one of claims 1 to 7, characterized in that: The following steps are involved: S1. The data acquisition module collects the operating parameters of the power cable and the surrounding environment parameters in real time to obtain the data vector ; S2. Use the data preprocessing module to clean, denoise and normalize the collected data to obtain ; S3, using the data transmission module to transmit the pre-processed data to the data analysis module; S4, the data analysis module uses the pre-trained large model to analyze and process the data, that is, input , get the failure risk probability ; S5. If Exceeding the set threshold , the early warning module sends out a corresponding early warning signal; S6. Store the collected data, analysis results and warning records in the database module for subsequent query and analysis.

Images