Power cable fault diagnosis and analysis method and system based on multi-source data

Abstract

The invention discloses a power cable fault diagnosis and analysis method and system based on multi-source data, and relates to the technical field of diagnosis and analys.During operation of the system, various sensors are installed in a cable and the surrounding environment of the cable, operation data of the power cable are collected in real time, and after the collected multi-source data are preprocessed, the power cable fault diagnosis and analysis method and system based on the multi-source data are obtained. The method comprises the following steps: calculating key diagnosis parameters by using a machine learning algorithm, obtaining a comprehensive fault diagnosis index CDI, comparing the comprehensive fault diagnosis index CDI with a preset first threshold M and a preset second threshold N, obtaining a fault level and outputting a corresponding processing strategy, and carrying out fault point positioning by using a reflected wave method in combination with multi-source data according to fault characteristics and a cable model. And outputting a specific position of a fault point, accurately positioning the fault point in the cable, integrating analysis results of the modules, and automatically generating a cable fault diagnosis report which comprises a fault type, a fault point position, potential risk analysis and maintenance suggestions.

Description

Technical Field

[0001] The present invention relates to the field of diagnosis and analysis technology, and in particular to a method and system for diagnosing and analyzing power cable faults based on multi-source data. Background Art

[0002] In the process of power transmission, cables, as the core component of the transmission network, bear the heavy responsibility of transmitting electricity from power stations to users. However, as the use time increases, cables may fail due to various factors such as insulation aging, mechanical damage, and changes in the external environment. Traditional cable fault diagnosis methods usually rely on a single detection method, which cannot fully obtain the operating status data of the cable and is difficult to accurately locate the fault point. At present, most power cable fault diagnosis systems have some shortcomings. The traditional single data source diagnosis method relies on single electrical parameters such as current and voltage, ignoring other important factors that affect the cable status, such as temperature changes, partial discharge signal characteristics, etc., resulting in insufficient accuracy and reliability of the diagnosis results. Even the diagnosis method that combines multiple data sources is often difficult to achieve comprehensive data fusion and intelligent analysis due to the limitations of data processing and analysis capabilities. In addition, most existing fault location technologies rely on the reflection wave method, but fail to combine the comprehensive analysis results of multi-source data, thus failing to achieve more accurate fault location and comprehensive diagnosis.

[0003] The above status quo and shortcomings are mainly caused by the fact that traditional diagnostic methods rely on single or simple multi-source data analysis methods and lack systematic data fusion and intelligent algorithm support. When a cable fails, if it fails to be diagnosed and located in a timely and accurate manner, it may lead to further damage to the power system or the expansion of the accident, causing serious economic losses and safety hazards. For example, in the long-term operation of high-voltage cable lines, the frequent occurrence of partial discharge and insulation aging problems cannot be discovered and handled in time, which may lead to serious faults such as cable short circuits or disconnections. By developing a new intelligent diagnostic analysis system based on multi-source data fusion, it is possible to better monitor and evaluate the cable status, issue warnings in time and take corresponding maintenance measures, thereby reducing the risk of power system failures and improving the safety and reliability of the overall power grid. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention provides a method and system for diagnosing and analyzing power cable faults based on multi-source data, which solves the problems mentioned in the background technology.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: a power cable fault diagnosis and analysis system based on multi-source data, including a data acquisition module, a data preprocessing module, a feature extraction module, a fault type analysis module, a fault location module and a diagnosis report generation module;

[0006] The data acquisition module is used to use a variety of sensors installed on the cable and its surrounding environment to collect the operation data of the power cable in real time, and send the collected data to the data preprocessing module through the wireless transmission module;

[0007] The data preprocessing module is used to clean, denoise and normalize the collected multi-source data;

[0008] The feature extraction module is used to extract the features of electrical signals and temperature signals from the preprocessed data through time domain and frequency domain analysis, and calculate key diagnostic parameters using a machine learning algorithm to obtain: a comprehensive fault diagnosis index CDI;

[0009] The fault type analysis module is used to analyze the fault type of the cable, including short circuit, disconnection or insulation aging, based on the calculation result of the comprehensive fault diagnosis index CDI, match the multi-source data features with the historical fault database, and compare the comprehensive fault diagnosis index CDI with the preset first threshold M and second threshold N to obtain the fault level and output the corresponding processing strategy;

[0010] The fault location module is used to locate the fault point according to the fault characteristics and cable model using the reflection wave method combined with multi-source data, output the specific location of the fault point, the distance from the cable starting point to the fault point, and accurately locate the fault point in the cable;

[0011] The diagnostic report generation module is used to integrate the analysis results of each module and automatically generate a cable fault diagnosis report, including fault type, fault point location, potential risk analysis and maintenance suggestions.

[0012] Preferably, the data acquisition module includes an electrical parameter acquisition unit, a temperature monitoring unit and a partial discharge monitoring unit;

[0013] The electrical parameter acquisition unit is used to collect the electrical operating status parameters of the cable, including current, voltage and phase unbalance coefficient Ub. The mutual inductor is installed on each phase line of the cable to monitor the three-phase current and voltage in real time. The three-phase current I is collected by the sensor. a , I b , I c Real-time data of voltage and current;

[0014] The temperature monitoring unit is used to collect temperature data of the cable surface and the environment in real time using infrared temperature sensors and thermocouples. The infrared temperature sensors are installed at key points of the cable, including cable joints or bends. Thermocouples are distributed at different positions of the cable to obtain more accurate temperature data to monitor the temperature of the cable in real time and record the highest temperature T at each sampling point. max and the minimum temperature T min and the length L of the cable.

[0015] Preferably, the partial discharge monitoring unit is used to use an acceleration sensor installed on the outer casing of the cable to detect mechanical vibration signals, and an ultrasonic sensor and a high-frequency current sensor are installed at the cable joint or where insulation defects are likely to occur to detect the amplitude, frequency and phase mode of partial discharge, and obtain: amplitude A i , frequency f i and phase angle θ i ;

[0016] All collected data are aggregated and transferred to the data preprocessing module.

[0017] Preferably, the data preprocessing module includes a data preprocessing unit;

[0018] The data preprocessing unit is used to receive cable operation data, remove missing values, incomplete data and abnormal values, select a suitable filter by analyzing the spectral characteristics of the signal, remove high-frequency noise, and normalize the collected data from different sources to ensure that data of different units and magnitudes can be analyzed within the same range.

[0019] Preferably, the feature extraction module includes a feature calculation unit;

[0020] The feature calculation unit is used to extract features from the preprocessed electrical signal and temperature signal data, analyze the time domain and frequency domain features of the signal, and calculate key diagnostic parameters through a machine learning algorithm to generate: a comprehensive fault diagnosis index CDI, an interphase imbalance coefficient Ub, a temperature gradient coefficient Tgi, and a partial discharge signal characteristic coefficient PDf.

[0021] Preferably, the fault type analysis module includes a fault type identification unit and a fault level assessment unit;

[0022] The fault type identification unit is used to identify the specific fault type of the cable based on the calculation result of the comprehensive fault diagnosis index CDI, perform similarity matching between the extracted multi-source data features and the known fault modes in the historical fault database, analyze the value and trend of the comprehensive fault diagnosis index CDI, and preliminarily judge the fault type. If a high CDI value indicates serious insulation aging or short circuit fault, and a low CDI value indicates a minor line break problem, it is transmitted to the fault level evaluation unit for evaluation.

[0023] Preferably, the fault level evaluation unit is used to compare the comprehensive fault diagnosis index CDI with the preset first threshold value M and second threshold value N, evaluate the fault level, and output a corresponding processing strategy:

[0024] If the comprehensive fault diagnosis index CDI < the first threshold M: the fault level is level 1, indicating that the cable has a potential fault but does not exceed the safe operation range, and no risk factors that need to be dealt with urgently are detected;

[0025] If the first threshold M ≤ comprehensive fault diagnosis index CDI < second threshold N: the fault level is level 2, and the cable has hidden dangers of aging insulation, local overheating or loose joints, indicating that maintenance is required;

[0026] If the comprehensive fault diagnosis index CDI ≥ the second threshold N: the fault level is level three, the cable is completely faulty, and there are problems including short circuit, wire breakage or serious insulation damage, indicating that immediate measures need to be taken.

[0027] Preferably, the fault location module includes a fault point location unit;

[0028] The fault point location unit is used to send a pulse signal in the cable, record the reflected wave signal at the fault point, analyze the time delay and attenuation characteristics of the reflected wave, and calculate the distance from the starting point of the cable to the fault point by using the return time and signal strength of the reflected wave;

[0029] The impedance change of electrical characteristics, local overheating of temperature characteristics and abnormal vibration data of vibration signals are fused and analyzed with the reflection wave positioning results. Based on the comprehensive analysis results of the reflection wave method and multi-source data fusion, the specific location of the fault point is output, including the precise distance from the cable starting point to the fault point.

[0030] Preferably, the diagnostic report generating module comprises a report generating unit;

[0031] The report generation unit is used to automatically collect various analysis result data, including fault types such as short circuit, wire break, insulation aging, the precise distance from the cable starting point to the fault point, and the diagnosis conclusion based on the comprehensive fault diagnosis index CDI, and generate automatic chart drawing, key data point annotation and text optimization.

[0032] A method for diagnosing and analyzing power cable faults based on multi-source data comprises the following steps:

[0033] Step 1: Use a variety of sensors installed on the cable and its surrounding environment to collect the operation data of the power cable in real time, and send the collected data to the data preprocessing module through the wireless transmission module;

[0034] Step 2: Clean, denoise and normalize the collected multi-source data;

[0035] Step 3: Extract the characteristics of electrical signals and temperature signals from the preprocessed data through time domain and frequency domain analysis, and use machine learning algorithms to calculate key diagnostic parameters to obtain: Comprehensive Fault Diagnosis Index CDI;

[0036] Step 4: Based on the calculation result of the comprehensive fault diagnosis index CDI, analyze the fault type of the cable, including short circuit, disconnection or insulation aging, match the multi-source data features with the historical fault database, and compare the comprehensive fault diagnosis index CDI with the preset first threshold M and second threshold N to obtain the fault level and output the corresponding processing strategy;

[0037] Step 5: According to the fault characteristics and cable model, the reflected wave method is used in combination with multi-source data to locate the fault point, output the specific location of the fault point, the distance from the cable starting point to the fault point, and accurately locate the fault point in the cable;

[0038] Step 6: Based on the analysis results of each module, a cable fault diagnosis report is automatically generated, including fault type, fault point location, potential risk analysis and maintenance recommendations.

[0039] (III) Beneficial effects

[0040] The present invention provides a method and system for diagnosing and analyzing power cable faults based on multi-source data, which has the following beneficial effects:

[0041] (1) When the system is in operation, a variety of sensors are installed on the cable and its surrounding environment to collect the operation data of the power cable in real time. After preprocessing the collected multi-source data, the key diagnostic parameters are calculated using a machine learning algorithm to obtain: a comprehensive fault diagnosis index CDI. The comprehensive fault diagnosis index CDI is compared with the preset first threshold M and second threshold N to obtain the fault level and output the corresponding processing strategy. According to the fault characteristics and cable model, the reflection wave method is used in combination with multi-source data to locate the fault point, output the specific location of the fault point, accurately locate the fault point in the cable, and automatically generate a cable fault diagnosis report based on the analysis results of each module, including the fault type, fault point location, potential risk analysis and maintenance recommendations.

[0042] (2) Through the comprehensive use of data acquisition module, data preprocessing module, feature extraction module, fault type analysis module, fault location module and diagnostic report generation module, the power cable fault diagnosis and analysis system based on multi-source data has achieved significant improvements in the accuracy and efficiency of cable fault diagnosis and location. The data acquisition module obtains multi-dimensional data such as electrical, temperature and partial discharge in real time through a variety of sensors to fully reflect the operating status of the cable; the data preprocessing module cleans, denoises and normalizes these multi-source data to ensure the reliability and consistency of the data; the feature extraction module calculates key diagnostic parameters such as the comprehensive fault diagnosis index CDI through time domain and frequency domain analysis and machine learning algorithms, providing a solid data foundation for subsequent fault analysis. These modules work together to achieve comprehensive monitoring and analysis of the cable status.

[0043] (3) In terms of fault diagnosis and location, the introduction of the fault type analysis module and the fault location module greatly improves the accuracy of fault identification and location. The fault type analysis module uses the comprehensive fault diagnosis index CDI to compare with the preset threshold, which can effectively identify fault types such as short circuit, wire break or insulation aging, and output corresponding processing strategies according to different fault levels; the fault location module uses the reflection wave method combined with multi-source data to accurately locate the fault point, and provides specific distance information from the cable starting point to the fault point. This fault diagnosis and location technology based on multi-source data fusion is more efficient and accurate than traditional methods, reducing diagnostic errors and the uncertainty of manual judgment, thereby improving the response speed of fault maintenance.

[0044] (4) The diagnostic report generation module automatically generates a cable fault diagnostic report based on the analysis results of each module, intuitively displays the fault type, location, potential risks and maintenance recommendations, and provides a scientific decision-making basis for operation and maintenance personnel. Compared with existing technical means, this system integrates multiple sensor data and intelligent analysis algorithms to realize an integrated intelligent diagnostic process from data acquisition, preprocessing to fault diagnosis, positioning and report generation. Its advantage is that it greatly improves the automation level and information visualization capabilities of cable fault diagnosis, significantly reduces operation and maintenance costs and system downtime, and comprehensively improves the safety and reliability of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 This is a block diagram of a power cable fault diagnosis and analysis system based on multi-source data of the present invention;

[0046] Figure 2 The present invention is a schematic diagram of the steps of a power cable fault diagnosis and analysis method based on multi-source data. DETAILED DESCRIPTION

[0047] 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.

[0048] Example 1

[0049] The present invention provides a power cable fault diagnosis and analysis system based on multi-source data. Figure 1 , including data acquisition module, data preprocessing module, feature extraction module, fault type analysis module, fault location module and diagnosis report generation module;

[0050] The data acquisition module is used to use a variety of sensors installed on the cable and its surrounding environment to collect the operation data of the power cable in real time, and send the collected data to the data preprocessing module through the wireless transmission module;

[0051] The data preprocessing module is used to clean, denoise and normalize the collected multi-source data;

[0052] The feature extraction module is used to extract the features of electrical signals and temperature signals from the preprocessed data through time domain and frequency domain analysis, and calculate key diagnostic parameters using a machine learning algorithm to obtain: a comprehensive fault diagnosis index CDI;

[0053] The fault type analysis module is used to analyze the fault type of the cable, including short circuit, disconnection or insulation aging, based on the calculation result of the comprehensive fault diagnosis index CDI, match the multi-source data features with the historical fault database, and compare the comprehensive fault diagnosis index CDI with the preset first threshold M and second threshold N to obtain the fault level and output the corresponding processing strategy;

[0054] The fault location module is used to locate the fault point according to the fault characteristics and cable model using the reflection wave method combined with multi-source data, output the specific location of the fault point, the distance from the cable starting point to the fault point, and accurately locate the fault point in the cable;

[0055] The diagnostic report generation module is used to integrate the analysis results of each module and automatically generate a cable fault diagnosis report, including fault type, fault point location, potential risk analysis and maintenance suggestions.

[0056] In this embodiment, by using a variety of sensors installed on the cable and its surrounding environment, the operation data of the power cable is collected in real time, the collected data is sent to the data preprocessing module through the wireless transmission module, the collected multi-source data is cleaned, denoised and normalized, the characteristics of the electrical signal and the temperature signal are extracted from the preprocessed data through time domain and frequency domain analysis, and the key diagnostic parameters are calculated using a machine learning algorithm to obtain: a comprehensive fault diagnosis index CDI, based on the calculation result of the comprehensive fault diagnosis index CDI, the fault type of the cable is analyzed, including short circuit, disconnection or insulation aging, the multi-source data characteristics are matched with the historical fault database, the comprehensive fault diagnosis index CDI is compared with the preset first threshold M and second threshold N, the fault level is obtained and the corresponding processing strategy is output, according to the fault characteristics and the cable model, the reflection wave method is used in combination with the multi-source data to locate the fault point, the specific location of the fault point, the distance from the cable starting point to the fault point are output, the fault point in the cable is accurately located, and the analysis results of each module are integrated to automatically generate a cable fault diagnosis report, including the fault type, the location of the fault point, the potential risk analysis and maintenance suggestions.

[0057] Example 2

[0058] This embodiment is explained in Example 1, please refer to Figure 1 ,Specifically: the data acquisition module includes an electrical parameter acquisition unit, a temperature monitoring unit and a partial discharge monitoring unit;

[0059] The electrical parameter acquisition unit is used to collect the electrical operating status parameters of the cable, including current, voltage and phase unbalance coefficient Ub. The mutual inductor is installed on each phase line of the cable to monitor the three-phase current and voltage in real time. The three-phase current I is collected by the sensor. a , I b , I c Real-time data of voltage and current;

[0060] The temperature monitoring unit is used to collect temperature data of the cable surface and the environment in real time using infrared temperature sensors and thermocouples. The infrared temperature sensors are installed at key points of the cable, including cable joints or bends. Thermocouples are distributed at different positions of the cable to obtain more accurate temperature data to monitor the temperature of the cable in real time and record the highest temperature T at each sampling point. max and the minimum temperature T min and the length L of the cable;

[0061] The partial discharge monitoring unit is used to use an acceleration sensor installed on the outer casing of the cable to detect mechanical vibration signals. The ultrasonic sensor and the high-frequency current sensor are installed at the cable joint or where insulation defects are likely to occur to detect the amplitude, frequency and phase mode of partial discharge, and obtain: Amplitude A i , frequency f i and phase angle θ i ;

[0062] All collected data are aggregated and transferred to the data preprocessing module.

[0063] The data preprocessing module includes a data preprocessing unit;

[0064] The data preprocessing unit is used to receive cable operation data, remove missing values, incomplete data and abnormal values, select a suitable filter by analyzing the spectral characteristics of the signal, remove high-frequency noise, and normalize the collected data from different sources to ensure that data of different units and magnitudes can be analyzed within the same range.

[0065] In this embodiment, the system significantly improves the operating status monitoring and fault diagnosis capabilities of power cables by integrating electrical parameter acquisition, temperature monitoring and partial discharge monitoring units. The electrical parameter acquisition unit monitors electrical parameters such as current, voltage and phase imbalance coefficient in real time, providing reliable data for cable fault diagnosis. The temperature monitoring unit combines infrared sensors and thermocouples to accurately collect temperature information at key locations, which helps to provide early warning of thermal faults. The partial discharge monitoring unit detects the amplitude, frequency and phase pattern of partial discharge, and can effectively identify cable insulation defects and aging problems. The data preprocessing module ensures data consistency and reliability by removing outliers, selecting appropriate filters for denoising, and normalizing multi-source data. This module improves the accuracy of data analysis and the accuracy of fault diagnosis, and reduces the risk of misjudgment and missed judgment.

[0066] Example 3

[0067] This embodiment is explained in Example 1, please refer to Figure 1 ,Specifically: the feature extraction module includes a feature calculation unit;

[0068] The feature calculation unit is used to extract features from the preprocessed electrical signal and temperature signal data, analyze the time domain and frequency domain features of the signal, and calculate key diagnostic parameters through a machine learning algorithm to generate: a comprehensive fault diagnosis index CDI, an interphase unbalance coefficient Ub, a temperature gradient coefficient Tgi, and a partial discharge signal characteristic coefficient PDf;

[0069] The comprehensive fault diagnosis index CDI is calculated by the following formula:

[0070] CDI=α1*Ub+α2*PDf+α3*Tgi;

[0071] Wherein, CDI represents the comprehensive fault diagnosis index, Ub represents the phase imbalance coefficient, Tgi represents the temperature gradient coefficient, PDf represents the partial discharge signal characteristic coefficient, α1, α2 and α3 represent the proportional coefficients of the phase imbalance coefficient Ub, the temperature gradient coefficient Tgi and the partial discharge signal characteristic coefficient PDf respectively;

[0072] The phase unbalance coefficient Ub is calculated by the following formula:

[0073]

[0074] In the formula, I a , I b , I c Represents three-phase current, I avg Represents the average value of the three-phase current. The collected data is synchronously sampled and converted into digital-to-analog format to calculate the average value of the three-phase current I avgAnd the phase unbalance coefficient Ub is calculated by the formula.

[0075] The temperature gradient coefficient Tgi is calculated by the following formula:

[0076]

[0077] Where, T max Indicates the maximum temperature, T min represents the minimum temperature, L represents the length of the cable, and the temperature gradient index Tgi is calculated based on the temperature difference in each time period and the length L of the cable.

[0078] The local discharge signal characteristic coefficient PDf is calculated by the following formula:

[0079]

[0080] In the formula, A i represents the amplitude of partial discharge, f i represents the discharge frequency, θ i Indicates the phase angle of the discharge.

[0081] In this embodiment, the feature extraction module extracts key diagnostic parameters from the preprocessed multi-source data through the feature calculation unit, thereby realizing a comprehensive analysis of the cable fault. The phase-to-phase imbalance coefficient Ub is calculated based on the three-phase current difference, and can monitor the electrical state of the cable in real time and quickly identify the phase-to-phase current imbalance fault. The temperature gradient coefficient Tgi is calculated by the temperature difference at different positions of the cable, and can accurately evaluate the thermal fault situation, which helps to detect overheating problems caused by overload or external environmental influences. The local discharge signal characteristic coefficient PDf combines signal characteristics such as amplitude, frequency and phase angle to effectively identify insulation aging and local discharge anomalies. The comprehensive fault diagnosis index CDI is used as the core diagnostic indicator. By combining the proportional relationship between the phase-to-phase imbalance coefficient Ub, the temperature gradient coefficient Tgi and the local discharge signal characteristic coefficient PDf, a unified fault assessment standard is formed. This comprehensive index can quantify and integrate multiple fault characteristics, greatly improving the accuracy and efficiency of cable fault diagnosis. Compared with the traditional single feature analysis method, it uses machine learning algorithms to analyze time domain and frequency domain characteristics from multiple angles, effectively reducing the misjudgment rate and missed judgment rate, and improving the accuracy of fault location. It provides a scientific basis and optimized solutions for intelligent cable fault management, reducing the operating risks and maintenance costs of the power system.

[0082] Example 4

[0083] This embodiment is explained in Example 1, please refer to Figure 1 ,Specifically: the fault type analysis module includes a fault type identification unit and a fault level evaluation unit;

[0084] The fault type identification unit is used to identify the specific fault type of the cable based on the calculation result of the comprehensive fault diagnosis index CDI, perform similarity matching between the extracted multi-source data features and the known fault modes in the historical fault database, analyze the value and trend of the comprehensive fault diagnosis index CDI, and preliminarily judge the fault type. If a high CDI value indicates a serious insulation aging or short circuit fault, and a low CDI value indicates a minor disconnection problem, it is transmitted to the fault level evaluation unit for evaluation;

[0085] The fault level evaluation unit is used to compare the comprehensive fault diagnosis index CDI with the preset first threshold value M and second threshold value N, evaluate the fault level, and output a corresponding processing strategy:

[0086] If the comprehensive fault diagnosis index CDI < the first threshold M: the fault level is level 1, indicating that the cable has a potential fault but does not exceed the safe operating range, and no risk factors that need to be dealt with urgently are detected. It is recommended to conduct regular monitoring of the cable's operating status, including regular collection of parameters such as electrical signals, temperature and vibration signals, and record their data in the database for trend analysis. If multiple sampling results show that the CDI value tends to increase, the monitoring frequency should be increased, and more detailed data analysis should be performed to detect potential risks early;

[0087] If the first threshold M≤comprehensive fault diagnosis index CDI<second threshold N: the fault level is level 2, and the cable has hidden dangers of aging insulation layer, local overheating or loose joints, indicating that maintenance is needed. It is recommended to immediately activate the early warning mechanism and notify the operation and maintenance personnel to conduct on-site inspection and diagnosis to eliminate potential hidden dangers. Use more sophisticated diagnostic tools (such as infrared imagers, partial discharge testers, etc.) to conduct detailed inspections on key nodes of the cable to identify and locate possible fault points;

[0088] If the comprehensive fault diagnosis index CDI ≥ the second threshold N: the fault level is level three, the cable is completely faulty, and there are problems including short circuit, broken wire or serious insulation damage, indicating that immediate measures need to be taken to activate the emergency response mechanism to automatically cut off the power supply of the cable to avoid possible safety accidents, including fire or electric shock. After confirming the cause of the fault, perform necessary repair or replacement operations, such as replacing damaged cable segments, repairing or replacing cable connectors, adding or replacing insulating materials, etc.

[0089] The fault location module includes a fault point location unit;

[0090] The fault point location unit is used to send a pulse signal in the cable, record the reflected wave signal at the fault point, analyze the time delay and attenuation characteristics of the reflected wave, and calculate the distance from the starting point of the cable to the fault point by using the return time and signal strength of the reflected wave;

[0091] The distance calculation formula is as follows:

[0092] In the formula, D is the distance from the starting point of the cable to the fault point, v is the propagation speed of the radio wave in the cable, and t is the round-trip time of the pulse signal;

[0093] The impedance change of electrical characteristics, local overheating of temperature characteristics and abnormal vibration data of vibration signals are fused and analyzed with the reflection wave positioning results. Based on the comprehensive analysis results of the reflection wave method and multi-source data fusion, the specific location of the fault point is output, including the precise distance from the cable starting point to the fault point.

[0094] The diagnostic report generating module comprises a report generating unit;

[0095] The report generation unit is used to automatically collect various analysis result data, including fault types such as short circuit, wire break, insulation aging, the precise distance from the cable starting point to the fault point, and the diagnosis conclusion based on the comprehensive fault diagnosis index CDI, and generate automatic chart drawing, key data point annotation and text optimization.

[0096] In this embodiment, the fault type analysis module can accurately identify the specific fault type of the cable and perform graded processing through the cooperation of the fault type identification unit and the fault level evaluation unit. The module uses the similarity matching analysis of the comprehensive fault diagnosis index CDI and the historical fault database to quickly identify different types of faults such as short circuit, disconnection or insulation aging of the cable, and compares the CDI with the preset thresholds M and N to accurately evaluate the severity of the fault and give corresponding maintenance or emergency treatment strategies, thereby improving the efficiency and accuracy of fault handling. The fault location module accurately determines the fault point location of the cable through the pulse signal method. The module uses the time delay and attenuation characteristics of the reflected wave, combined with the results of multi-source data analysis, such as impedance change, temperature anomaly and vibration signal anomaly, to fuse and analyze various features, and can accurately calculate the distance from the cable starting point to the fault point, ensuring the accuracy of fault location. Compared with traditional methods, it significantly reduces the time and cost of fault finding, and improves the reliability and maintenance efficiency of the power system. The diagnostic report generation module can automatically generate a detailed cable fault diagnosis report, covering fault type, fault level, fault point location and maintenance recommendations. This module automatically integrates and analyzes the results of each module, provides visual charts and textual diagnostic suggestions, and provides intuitive reference for maintenance personnel, which significantly improves the scientificity and timeliness of cable maintenance management and optimizes the operation and maintenance process of the power system.

[0097] Example 5

[0098] A power cable fault diagnosis and analysis method based on multi-source data, please refer to Figure 2 , specifically: including the following steps:

[0099] Step 1: Use a variety of sensors installed on the cable and its surrounding environment to collect the operation data of the power cable in real time, and send the collected data to the data preprocessing module through the wireless transmission module;

[0100] Step 2: Clean, denoise and normalize the collected multi-source data;

[0101] Step 3: Extract the characteristics of electrical signals and temperature signals from the preprocessed data through time domain and frequency domain analysis, and use machine learning algorithms to calculate key diagnostic parameters to obtain: Comprehensive Fault Diagnosis Index CDI;

[0102] Step 4: Based on the calculation result of the comprehensive fault diagnosis index CDI, analyze the fault type of the cable, including short circuit, disconnection or insulation aging, match the multi-source data features with the historical fault database, and compare the comprehensive fault diagnosis index CDI with the preset first threshold M and second threshold N to obtain the fault level and output the corresponding processing strategy;

[0103] Step 5: According to the fault characteristics and cable model, the reflected wave method is used in combination with multi-source data to locate the fault point, output the specific location of the fault point, the distance from the cable starting point to the fault point, and accurately locate the fault point in the cable;

[0104] Step 6: Based on the analysis results of each module, a cable fault diagnosis report is automatically generated, including fault type, fault point location, potential risk analysis and maintenance recommendations.

[0105] In this embodiment, the power cable fault diagnosis and analysis system based on multi-source data provides an efficient, accurate and automated fault detection and diagnosis solution. Through the close integration of six steps, the efficiency and safety of cable operation and maintenance are effectively improved. First, steps one and two realize the real-time collection and high-quality preprocessing of power cable operation data, laying a solid data foundation for subsequent analysis. The collaborative work of multiple sensors can collect comprehensive electrical, temperature, partial discharge and other data. Combined with the wireless transmission module, remote and continuous data monitoring and transmission can be achieved. The data cleaning, denoising and normalization process ensures the accuracy and consistency of the analysis data, eliminates the errors and potential risks caused by data quality problems, and steps three and four further improve the intelligent level of fault diagnosis through feature extraction and machine learning algorithms. By extracting time domain and frequency domain feature data and calculating the comprehensive fault diagnosis index CDI, the system can accurately identify the type of cable fault. By matching the comprehensive fault diagnosis index with the historical database and comparing the threshold, the system can quickly determine the type and level of the fault and provide corresponding processing strategies. This intelligent fault analysis and classification method greatly shortens the time of manual analysis and the probability of misjudgment in the traditional diagnosis process. Steps 5 and 6 achieve accurate positioning of the fault point through the fusion analysis of the reflection wave method and multi-source data, and generate a detailed diagnostic report. Compared with the traditional single detection method, the reflection wave method combined with the analysis method of multi-source data can more accurately locate the location of the fault point and provide the specific distance from the cable starting point to the fault point, which significantly improves the accuracy and efficiency of maintenance work. The introduction of the diagnostic report generation module provides scientific decision-making support for maintenance personnel, and intuitively displays the fault type, location, potential risks and maintenance recommendations. Compared with existing technical means, this system not only improves the accuracy and automation of cable fault diagnosis, but also greatly reduces the operation and maintenance costs and system downtime, and comprehensively improves the safety and reliability of the power system.

[0106] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A power cable fault diagnosis and analysis system based on multi-source data, characterized by: It includes data acquisition module, data preprocessing module, feature extraction module, fault type analysis module, fault location module and diagnosis report generation module; The data acquisition module is used to use a variety of sensors installed on the cable and its surrounding environment to collect the operation data of the power cable in real time, and send the collected data to the data preprocessing module through the wireless transmission module; The data preprocessing module is used to clean, denoise and normalize the collected multi-source data; The feature extraction module is used to extract the features of electrical signals and temperature signals from the preprocessed data through time domain and frequency domain analysis, and calculate key diagnostic parameters using a machine learning algorithm to obtain: a comprehensive fault diagnosis index CDI; The fault type analysis module is used to analyze the fault type of the cable, including short circuit, disconnection or insulation aging, based on the calculation result of the comprehensive fault diagnosis index CDI, match the multi-source data features with the historical fault database, and compare the comprehensive fault diagnosis index CDI with the preset first threshold M and second threshold N to obtain the fault level and output the corresponding processing strategy; The fault location module is used to locate the fault point according to the fault characteristics and cable model using the reflection wave method combined with multi-source data, output the specific location of the fault point, the distance from the cable starting point to the fault point, and accurately locate the fault point in the cable; The diagnostic report generation module is used to integrate the analysis results of each module and automatically generate a cable fault diagnosis report, including fault type, fault point location, potential risk analysis and maintenance suggestions.

2. The power cable fault diagnosis and analysis system based on multi-source data according to claim 1 is characterized in that: The data acquisition module includes an electrical parameter acquisition unit, a temperature monitoring unit and a partial discharge monitoring unit; The electrical parameter acquisition unit is used to collect the electrical operating status parameters of the cable, including current, voltage and phase unbalance coefficient Ub. The mutual inductor is installed on each phase line of the cable to monitor the three-phase current and voltage in real time. The three-phase current I is collected by the sensor. a , I b , I c Real-time data of voltage and current; The temperature monitoring unit is used to collect temperature data of the cable surface and the environment in real time using infrared temperature sensors and thermocouples. The infrared temperature sensors are installed at key points of the cable, including cable joints or bends. Thermocouples are distributed at different positions of the cable to obtain more accurate temperature data to monitor the temperature of the cable in real time and record the highest temperature T at each sampling point. max and the minimum temperature T min and the length L of the cable.

3. The power cable fault diagnosis and analysis system based on multi-source data according to claim 2 is characterized in that: The partial discharge monitoring unit is used to use an acceleration sensor installed on the outer casing of the cable to detect mechanical vibration signals. The ultrasonic sensor and the high-frequency current sensor are installed at the cable joint or where insulation defects are likely to occur to detect the amplitude, frequency and phase mode of partial discharge, and obtain: Amplitude A i , frequency f i and phase angle θ i ; All collected data are aggregated and transferred to the data preprocessing module.

4. The power cable fault diagnosis and analysis system based on multi-source data according to claim 2 is characterized in that: The data preprocessing module includes a data preprocessing unit; The data preprocessing unit is used to receive cable operation data, remove missing values, incomplete data and abnormal values, select a suitable filter by analyzing the spectral characteristics of the signal, remove high-frequency noise, and normalize the collected data from different sources to ensure that data of different units and magnitudes can be analyzed within the same range.

5. The power cable fault diagnosis and analysis system based on multi-source data according to claim 2 is characterized in that: The feature extraction module includes a feature calculation unit; The feature calculation unit is used to extract features from the preprocessed electrical signal and temperature signal data, analyze the time domain and frequency domain features of the signal, and calculate key diagnostic parameters through a machine learning algorithm to generate: a comprehensive fault diagnosis index CDI, an interphase imbalance coefficient Ub, a temperature gradient coefficient Tgi, and a partial discharge signal characteristic coefficient PDf.

6. The power cable fault diagnosis and analysis system based on multi-source data according to claim 1 is characterized in that: The fault type analysis module includes a fault type identification unit and a fault level assessment unit; The fault type identification unit is used to identify the specific fault type of the cable based on the calculation result of the comprehensive fault diagnosis index CDI, perform similarity matching between the extracted multi-source data features and the known fault modes in the historical fault database, analyze the value and trend of the comprehensive fault diagnosis index CDI, and preliminarily judge the fault type. If a high CDI value indicates serious insulation aging or short circuit fault, and a low CDI value indicates a minor line break problem, it is transmitted to the fault level evaluation unit for evaluation.

7. The power cable fault diagnosis and analysis system based on multi-source data according to claim 6 is characterized in that: The fault level evaluation unit is used to compare the comprehensive fault diagnosis index CDI with the preset first threshold value M and second threshold value N, evaluate the fault level, and output a corresponding processing strategy: If the comprehensive fault diagnosis index CDI < the first threshold M: the fault level is level 1, indicating that the cable has a potential fault but does not exceed the safe operation range, and no risk factors that need to be dealt with urgently are detected; If the first threshold M ≤ comprehensive fault diagnosis index CDI < second threshold N: the fault level is level 2, and the cable has hidden dangers of aging insulation, local overheating or loose joints, indicating that maintenance is required; If the comprehensive fault diagnosis index CDI ≥ the second threshold N: the fault level is level three, the cable is completely faulty, and there are problems including short circuit, wire breakage or serious insulation damage, indicating that immediate measures need to be taken.

8. The power cable fault diagnosis and analysis system based on multi-source data according to claim 1, characterized in that: The fault location module includes a fault point location unit; The fault point locating unit is used to send a pulse signal in the cable and record the reflected wave signal at the fault point, analyze the time delay and attenuation characteristics of the reflected wave, and use the return time and signal strength of the reflected wave to calculate the distance from the cable starting point to the fault point; fuse and analyze the impedance change of the electrical characteristics, the local overheating of the temperature characteristics, and the abnormal vibration data of the vibration signal with the reflected wave locating result, and output the specific location of the fault point, including the precise distance from the cable starting point to the fault point, based on the comprehensive analysis results of the reflected wave method and multi-source data fusion.

9. The power cable fault diagnosis and analysis system based on multi-source data according to claim 1, characterized in that: The diagnostic report generating module comprises a report generating unit; The report generation unit is used to automatically collect various analysis result data, including fault types such as short circuit, wire break, insulation aging, the precise distance from the cable starting point to the fault point, and the diagnosis conclusion based on the comprehensive fault diagnosis index CDI, and generate automatic chart drawing, key data point annotation and text optimization.

10. A method for diagnosing and analyzing power cable faults based on multi-source data, applied to a system for diagnosing and analyzing power cable faults based on multi-source data as claimed in any one of claims 1 to 9, characterized in that: The following steps are involved: Step 1: Use a variety of sensors installed on the cable and its surrounding environment to collect the operation data of the power cable in real time, and send the collected data to the data preprocessing module through the wireless transmission module; Step 2: Clean, denoise and normalize the collected multi-source data; Step 3: Extract the characteristics of electrical signals and temperature signals from the preprocessed data through time domain and frequency domain analysis, and use machine learning algorithms to calculate key diagnostic parameters to obtain: Comprehensive Fault Diagnosis Index CDI; Step 4: Based on the calculation result of the comprehensive fault diagnosis index CDI, analyze the fault type of the cable, including short circuit, disconnection or insulation aging, match the multi-source data features with the historical fault database, and compare the comprehensive fault diagnosis index CDI with the preset first threshold M and second threshold N to obtain the fault level and output the corresponding processing strategy; Step 5: According to the fault characteristics and cable model, the reflected wave method is used in combination with multi-source data to locate the fault point, output the specific location of the fault point, the distance from the cable starting point to the fault point, and accurately locate the fault point in the cable; Step 6: Based on the analysis results of each module, a cable fault diagnosis report is automatically generated, including fault type, fault point location, potential risk analysis and maintenance recommendations.

Images