A transmission line abnormal warning and fault location system

Abstract

The present invention discloses a transmission line abnormal warning and fault location system, belonging to the technical field of line detection, specifically including: dividing the warning devices in the line into each sub-region according to the distribution by the region demarcation module; when a fault occurs, the region location module quickly marks the relevant sub-regions and devices as to-be-determined status; the data acquisition module collects the data of these to-be-determined devices within a set time window before the fault, and performs preprocessing to determine whether the fault is caused by an electrical problem; if so, collect the historical power consumption data of the to-be-determined region; the fault location module performs normalization processing on the historical data and constructs a time series model, extracts the current value change sequence, divides it into multiple subsequences and then performs clustering analysis; judge the cause of the fault according to the offset situation of the category cluster where the to-be-determined current value change sequence is located in the clustering result: if the offset exceeds the preset threshold, it is considered to be caused by the abnormal operation of the electrical equipment; otherwise, it is caused by the aging of the line insulation material.

Description

Technical Field

[0001] The present invention relates to the technical field of line detection, and particularly to a system for abnormal warning and fault location of transmission lines. Background Art

[0002] With the continuous development of the power system and the continuous growth of electricity demand, the safe and stable operation of transmission lines is crucial for ensuring power supply. However, during the long-term operation of transmission lines, they may be affected by various factors, resulting in abnormalities or faults. These factors include abnormal behaviors of electrical equipment, aging of line insulation materials, environmental factors (such as lightning strikes, wind and rain, etc.), and aging and faults of the equipment itself. Therefore, timely and accurately warning of abnormal situations and locating faults in transmission lines is of great significance for improving the reliability and operation efficiency of the power system.

[0003] Traditional methods for detecting faults in transmission lines mainly rely on manual inspections and simple fault indicators. However, these methods have many limitations. Firstly, manual inspections are inefficient and it is difficult to detect potential fault hazards in a timely manner; secondly, simple fault indicators can only provide limited information, unable to accurately locate the fault position, nor can they distinguish the specific causes of faults. In addition, with the increasing complexity of the power system, traditional detection methods are difficult to meet the high-precision requirements of modern power systems for fault warning and location.

[0004] In recent years, with the rapid development of sensor technology, data acquisition technology, and data analysis methods, intelligent systems for fault warning and location of transmission lines have gradually become a research hotspot. By deploying various warning devices on transmission lines, the operating data of the lines can be collected in real time, including parameters such as current, voltage, and power. After preprocessing and analysis of these data, it can provide strong support for fault warning and location.

[0005] However, existing intelligent fault warning and location systems still have some deficiencies. On the one hand, most systems can only make a preliminary judgment on faults and it is difficult to accurately distinguish the specific causes of faults. On the other hand, existing systems often lack in-depth mining and utilization of historical data in data processing and analysis, and cannot fully utilize the value of data. Summary of the Invention

[0006] The purpose of the present invention is to provide a system for abnormal warning and fault location of transmission lines, and solve the following technical problems:

[0007] Most systems can only make a preliminary judgment on faults and it is difficult to accurately distinguish the specific causes of faults. Only relying on simple data comparison, lacking in-depth mining and analysis of historical data and time series, it is difficult to accurately distinguish the causes of faults.

[0008] The object of the present invention can be achieved by the following technical solutions:

[0009] A transmission line abnormal warning and fault location system, comprising:

[0010] A region demarcation module, configured to count all warning device information in the line, divide the line into several sub-regions according to the distribution of the warning devices, and each warning device is associated with a corresponding sub-region;

[0011] A region location module, configured to, when a fault event occurs, mark the sub-regions related to the fault event as undetermined regions, and mark the warning devices in the undetermined regions as undetermined devices;

[0012] A data acquisition module, configured to collect the monitoring data of the undetermined devices within a set time window t before the fault, preprocess the monitoring data, and determine whether the current fault event is caused by an electrical fault according to the monitoring data. If so, collect the historical power consumption data of the undetermined region before the fault event occurs;

[0013] A fault location module, configured to perform normalization processing on the historical power consumption data, construct a time series model, extract the current value change sequence before the fault event occurs, divide the current value change sequence into several sub-current value change sequences with a duration of t, mark the sub-current value change sequences within the set time window as undetermined current value change sequences, perform clustering on all the sub-current value change sequences, check whether the category cluster where the undetermined current value change sequence is located deviates during the clustering process. If the deviation exceeds a preset threshold, it is determined that the electrical fault is caused by an abnormal electrical device; if the deviation does not exceed the preset threshold, it is determined that the electrical fault is due to the aging of the line insulation material;

[0014] The warning devices include electric energy meters, circuit protection devices, electrical fault warning sensors, intelligent sockets, and electrical fault prevention and control devices; the monitoring data of the warning devices includes the fundamental wave of the current signal, the amplitude of the current signal, harmonic components, and voltage;

[0015] The warning devices calculate power-related parameters according to the real-time voltage value and the real-time current value, and the power-related parameters include apparent power, active power, reactive power, and power factor;

[0016] The process by which the data acquisition module determines whether the fault event is caused by an electrical fault is as follows:

[0017] Obtain the values of all monitoring data and power - related parameters within a set time window before the occurrence of a fault event. If the amplitude of the real - time current value is 0, it is considered that this fault event is not an electrical fault. If the amplitude of the real - time current value exceeds the preset standard interval, it is considered that this fault event is an electrical fault. If the amplitude of the real - time current value is within the preset standard interval, obtain any real - time monitoring data or power parameter. When the value of any real - time monitoring data or power parameter exceeds the corresponding preset standard interval, it is considered that this fault event is an electrical fault.

[0018] As a further solution of the present invention: The process of the data acquisition module pre - processing the monitoring data is as follows:

[0019] Verify the monitoring data collected from each warning device, supplement missing values or outliers, convert the date - time field from a string to a unified format, detect and remove duplicate monitoring data records, remove outliers and invalid characters from the monitoring data, and normalize the values of different types of monitoring data.

[0020] As a further solution of the present invention: The process of normalization is as follows:

[0021]

[0022] where x i represents the normalized monitoring data, X i represents the original monitoring data, X max represents the maximum value in the original monitoring data, X min represents the minimum value in the original monitoring data.

[0023] As a further solution of the present invention: The process of the fault location module extracting the current value change sequence before the occurrence of a fault event is as follows:

[0024] Obtain the initial time series of historical power consumption data, collect the length of the initial time series and mark it as L, obtain the sliding step size of the time series and mark it as S. If the length of the current value change sequence is t, then the number of splits of the historical power consumption data is n = L / (S×t).

[0025] As a further solution of the present invention: The process of the fault location module clustering all sub - current value change sequences and checking whether the category cluster where the pending current value change sequence is located shifts during the clustering process is as follows:

[0026] Set a control radius and a minimum similarity number. For a sequence of undetermined current value changes, detect the number m of similar sequences within its control radius. If m is less than the minimum similarity number, directly determine that the cause of the electrical fault is an abnormality in the electrical equipment. If m is greater than or equal to the minimum similarity number, put the sequence of undetermined current value changes into a category cluster, and sequentially put all the sub-sequences of current value change values into the corresponding category clusters or mark them as abnormal;

[0027] Repeat the above process, extract the cluster centers of the category clusters after several clustering operations, compare the number of cluster centers where the sequence of undetermined current change values is located and the differences in the cluster center values. If the offset of the cluster center exceeds a preset threshold, determine that the electrical fault is caused by an abnormality in the electrical equipment; if the offset does not exceed the preset threshold, determine that the electrical fault is due to the aging of the line insulation material.

[0028] As a further solution of the present invention: The formula for judging whether the cluster center shifts is:

[0029]

[0030] Δc i =|c i,T+1 -c i,T |;

[0031] Wherein, S i is a set of data points, ∣S i ∣ is the number of data points in the category cluster, Δc i represents the change value of the i-th cluster center between two consecutive iterations; c i,T and c i,T+1 respectively represent the values of the i-th cluster center at time T and time T + 1. When Δc i is greater than the preset threshold, it indicates that the cluster center has shifted.

[0032] Advantages of the present invention:

[0033] The present invention divides the transmission line into several sub-regions according to the distribution of warning devices. By collecting the monitoring data within a set time window before the fault, and through analyzing the current value amplitude and other power-related parameters, it can quickly determine whether the fault event is caused by an electrical fault, avoiding misjudgment and improving the accuracy and reliability of early warning. Using historical electricity consumption data to construct a time series model, extracting the sequence of current value changes before the fault event occurs, and dividing it into several sub-sequences for clustering analysis. Through clustering analysis and cluster center offset detection, the system can effectively distinguish whether the electrical fault is caused by an abnormality in the electrical equipment or the aging of the line insulation material. This data-driven method for analyzing the cause of the fault avoids misjudgment caused by insufficient experience in traditional methods and provides a clear direction for subsequent repair and maintenance work. Description of the Drawings

[0034] The present invention will be further described below in conjunction with the accompanying drawings.

[0035] Figure 1 It is a schematic diagram of the modules of the present invention. Specific embodiments

[0036] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0037] Please refer to Figure 1 As shown, the present invention is a transmission line abnormal warning and fault location system, including:

[0038] A region demarcation module, which is a basic component of the system. Its main function is to count the information of all warning devices in the transmission line. These warning devices include electricity meters, circuit protection devices, electrical fault warning sensors, intelligent sockets, and electrical fault prevention and control devices, etc. According to the distribution of these warning devices, the module divides the entire transmission line into several sub-regions and ensures that each warning device is associated with a specific sub-region. This region division method provides a clear geographical and logical framework for subsequent fault location and data collection.

[0039] A region location module, which will be quickly activated when a fault event occurs. By analyzing the relevant information of the fault event, it marks the sub-regions related to the fault event as "pending regions". At the same time, the module further marks all warning devices within the pending regions as "pending devices". The purpose of this process is to quickly narrow down the fault troubleshooting scope and focus on the regions and devices that may be involved in the fault, thereby improving the efficiency of fault location.

[0040] A data collection module, which is the data processing core of the system. It is responsible for collecting the monitoring data of the pending devices within a set time window (denoted as t) before the fault occurs. These monitoring data include key parameters such as the fundamental wave, amplitude, harmonic components, and voltage of the current signal. To ensure the availability and accuracy of the data, the module preprocesses the collected monitoring data. The preprocessing process includes validating the data, supplementing missing values or outliers, converting the date and time fields from string format to a unified standard format, detecting and removing duplicate monitoring data records, removing outliers and invalid characters, and normalizing the numerical values of different types of monitoring data.

[0041] After the preprocessing is completed, the module determines whether the current fault event is caused by an electrical fault based on the monitoring data. If it is determined to be an electrical fault, historical power consumption data of the area to be determined before the fault event is further collected to provide more comprehensive data support for subsequent fault location and cause analysis.

[0042] Fault location module. The fault location module is the core functional module of the system, which is used to deeply analyze the collected historical power consumption data to achieve accurate fault location and cause judgment. The module first normalizes the historical power consumption data and then constructs a time series model. Through this model, the module extracts the sequence of current value changes before the fault event and divides this sequence into several sub-sequences of current value changes with a duration of t. Among them, the sub-sequence of current value changes within the set time window is marked as the "pending sub-sequence of current value changes".

[0043] Next, the module performs clustering analysis on all sub-sequences of current value changes. The purpose of clustering analysis is to classify the sub-sequences of current value changes with similar characteristics into the same category cluster. During the clustering process, the module pays special attention to whether the category cluster where the pending sub-sequence of current value changes is located has shifted. Specifically, the module sets a control radius and a minimum similarity number. For the pending sub-sequence of current value changes, the module detects the number of similar sequences within its control radius (denoted as m). If m is less than the minimum similarity number, it is directly determined that the electrical fault is caused by the abnormal behavior of the electrical equipment; if m is greater than or equal to the minimum similarity number, the pending sub-sequence of current value changes is placed into the corresponding category cluster, and all sub-sequences of current value changes are processed in turn.

[0044] To further confirm the cause of the fault, the module repeats the above clustering process several times and extracts the clustering centers of the category clusters after each clustering. By comparing the number and the numerical differences of the clustering centers where the pending sub-sequence of current value changes is located, the module determines whether the clustering center has shifted by more than a preset threshold. If the shift exceeds the preset threshold, it is determined that the electrical fault is caused by the abnormal behavior of the electrical equipment; if the shift does not exceed the preset threshold, it is determined that the electrical fault is caused by the aging of the line insulation material.

[0045] Workflow:

[0046] 1. Area division and location:

[0047] The system first counts all the warning device information in the transmission line through the area delineation module and divides the line into several sub-areas according to the device distribution. When a fault event occurs, the area location module quickly marks the sub-area related to the fault as the area to be determined and marks the warning devices in this area as the devices to be determined.

[0048] 2. Data collection and preprocessing:

[0049] The data acquisition module collects the monitoring data of the device to be determined within the set time window before the fault, and preprocesses the data, including operations such as verification, filling in missing values, format conversion, deduplication, removing outliers, and normalization. The preprocessed data is used to determine whether the fault is caused by an electrical fault. If it is an electrical fault, the module further collects the historical power consumption data of the area to be determined.

[0050] 3. Fault Location and Cause Analysis:

[0051] The fault location module normalizes the historical power consumption data, constructs a time series model, and extracts the sequence of current value changes before the occurrence of the fault event. The module divides the sequence of current value changes into several subsequences and performs clustering analysis on these subsequences. By detecting the deviation of the category clusters during the clustering process, the module finally determines whether the fault is caused by the abnormality of the electrical equipment or the aging of the line insulation material.

[0052] Through the collaborative work of the above modules, the transmission line anomaly warning and fault location system of the present invention can realize real-time warning, accurate location, and accurate cause analysis of the transmission line anomalies, providing strong support for the safe operation and efficient maintenance of the power system.

[0053] In another preferred embodiment of the present invention, the warning devices in the system are further clarified and optimized to improve the monitoring ability and data quality of the system. Specifically, the warning devices include the following key devices:

[0054] Electric energy meter: used to accurately measure the electric energy consumption in the line and provide basic parameters such as current, voltage, and power.

[0055] Circuit protection device: used to monitor the current and voltage of the line to ensure that the power supply can be cut off in time under abnormal conditions to protect the safety of equipment and personnel.

[0056] Electrical fault warning sensor: specifically used to detect the early signs of electrical faults, such as partial discharge, overheating, etc., and can send out warning signals in advance.

[0057] Intelligent socket: as the access point of the terminal device, it can real-time monitor the current, voltage, and power of the access device and upload the data to the system.

[0058] Electrical fault prevention and control device: used to actively monitor and intervene in electrical faults to prevent the expansion of faults.

[0059] These warning devices can real-time collect a variety of monitoring data, including the fundamental wave of the current signal, the amplitude of the current signal, harmonic components, and key parameters such as voltage. These data provide a rich information basis for the fault warning and location of the system.

[0060] Calculation of power-related parameters

[0061] In a preferred case of this embodiment, the warning device not only collects basic current and voltage data, but also further calculates and generates the following power-related parameters based on the real-time voltage value and real-time current value:

[0062] Apparent power: It represents the total power actually consumed by the device during operation and is the product of current and voltage.

[0063] Active power: It represents the power part of the device that actually does work and reflects the effective utilization of electrical energy by the device.

[0064] Reactive power: It represents the power part exchanged between the device and the power grid and reflects the reactive load of the device on the power grid.

[0065] Power factor: It represents the ratio of active power to apparent power and reflects the degree of effective utilization of electrical energy by the device.

[0066] By calculating these power-related parameters, the system can more comprehensively evaluate the operating state of the device and the electrical characteristics of the line, thereby providing a more accurate basis for fault judgment.

[0067] Preprocessing function of the data acquisition module

[0068] To ensure the quality and usability of the monitoring data, the data acquisition module will perform a series of preprocessing operations after collecting the data. The specific steps are as follows:

[0069] Data verification: Verify the monitoring data collected from each warning device to ensure the integrity and accuracy of the data.

[0070] Supplement missing values or outliers: For missing data or outliers, the system will reasonably supplement or correct them according to the context information.

[0071] Format conversion: Convert the date and time fields from string format to a unified time format for subsequent time series analysis.

[0072] Duplicate removal operation: Detect and remove duplicate monitoring data records to avoid the influence of redundant data on the analysis results.

[0073] Remove outliers and invalid characters: Remove outliers and invalid characters in the monitoring data through statistical analysis methods to ensure the rationality of the data.

[0074] Normalization processing: Normalize the numerical values of different types of monitoring data to unify the data dimension and improve the accuracy of analysis.

[0075] Detailed process of normalization processing

[0076] In a preferred case of this embodiment, the specific formula for normalization is as follows:

[0077]

[0078] Where x i represents the normalized monitoring data, X i represents the original monitoring data, X max represents the maximum value in the original monitoring data, X min represents the minimum value in the original monitoring data.

[0079] In another preferred embodiment of the present invention, when the data acquisition module determines whether a fault event is caused by an electrical fault, a systematic analysis process is adopted to ensure the accuracy and reliability of the determination. The specific process is as follows:

[0080] Data acquisition:

[0081] The data acquisition module first obtains the values of all monitoring data and power - related parameters within a set time window before the occurrence of the fault event. These data include the fundamental wave, amplitude, harmonic components of the current signal, voltage, and power - related parameters (such as apparent power, active power, reactive power, power factor, etc.) calculated by the warning device. These data provide a comprehensive information basis for subsequent fault determination.

[0082] Current amplitude analysis:

[0083] If the amplitude of the real - time current value is 0: In this case, it indicates that there is no current passing through the line, which may be due to the equipment not being started or the line being completely disconnected. Therefore, it can be determined that this fault event is not caused by an electrical fault, but may be caused by equipment shutdown, switch disconnection, or other non - electrical factors.

[0084] If the amplitude of the real - time current value exceeds the preset standard range: This indicates that the current value is abnormal and exceeds the normal operating range. In this case, it can be clearly determined that this fault event is caused by an electrical fault. For example, too high a current amplitude may indicate a short - circuit, overload, or other electrical abnormalities.

[0085] Further monitoring data and power parameter analysis:

[0086] If the amplitude of the real - time current value is within the preset standard range, it indicates that the current value itself does not show obvious abnormalities. At this time, the module will further obtain any real - time monitoring data or power parameter for more detailed analysis:

[0087] Monitoring data or power parameters exceed the preset standard range: If the value of any real-time monitoring data (such as voltage, harmonic components) or power parameters (such as apparent power, active power, power factor) exceeds the corresponding preset standard range, then this fault event is considered an electrical fault. For example, too high or too low voltage, abnormal increase in harmonic components, abnormal change in power factor, etc., may all be signs of electrical faults.

[0088] Both monitoring data and power parameters are within the standard range: If all monitoring data and power parameters are within the preset standard range, then it can be preliminarily judged that this fault event may not be caused by an electrical fault, but by other factors (such as normal startup, stop of the equipment, or external environmental factors).

[0089] In another preferred embodiment of the present invention, the process by which the fault location module extracts the change sequence of current values before the occurrence of a fault event is as follows:

[0090] Obtain the initial time series of historical electricity consumption data, collect the length of the initial time series and label it as L, obtain the sliding step size of the time series and label it as S, and the length of the current value change sequence is t, then the number of splits of the historical electricity consumption data is n = L / (S×t).

[0091] In another preferred embodiment of the present invention, the fault location module adopts a method based on cluster analysis to process all sub-current value change sequences, and judges the specific cause of the electrical fault by checking the changes of the category cluster where the pending current value change sequence is located during the clustering process. The specific implementation steps of this process are as follows:

[0092] 1. Initialization of cluster analysis:

[0093] The fault location module first normalizes the historical electricity consumption data and constructs a time series model. Through this model, the module extracts the change sequence of current values before the occurrence of the fault event and divides it into several sub-current value change sequences with a duration of t. These sub-sequences cover the key current change information before the occurrence of the fault and provide the basic data for subsequent cluster analysis.

[0094] 2. Similarity detection and category cluster division:

[0095] The module sets two key parameters for cluster analysis: the control radius and the minimum similarity number. The control radius is used to define the search range of the pending current value change sequence in the feature space, while the minimum similarity number is the minimum quantity requirement for judging sequence similarity.

[0096] For each pending current value change sequence, the module detects the number of similar sequences within the control radius, denoted as m;

[0097] If m is less than the minimum similarity number: This indicates that within the control radius, the number of sequences similar to the sequence of the to-be-determined current value changes is insufficient. In this case, the module directly determines that the cause of the electrical fault is the abnormal behavior of the electrical equipment. Because the abnormality of the electrical equipment usually leads to the randomness and uniqueness of the current change, making it difficult to match with other normal sequences.

[0098] If m is greater than or equal to the minimum similarity number: This indicates that there are sufficient similar sequences for the sequence of the to-be-determined current value changes within the control radius. The module places the sequence of the to-be-determined current value changes into the corresponding category cluster and processes all sub-sequences of the current value changes in turn, classifying them into the corresponding category clusters or marking them as abnormal. This process aims to classify the current change sequences with similar characteristics for subsequent analysis.

[0099] 3. Extraction and comparison of the clustering centers:

[0100] To further confirm the cause of the fault, the module repeats the above clustering process several times, and extracts the clustering centers of the category clusters after each clustering. The clustering center is the feature representative of all sequences in the category cluster, reflecting the typical current change pattern of the category cluster.

[0101] Calculation of the clustering center offset: By comparing the change value of the clustering center between two consecutive clustering iterations, it is judged whether the clustering center has shifted.

[0102] Judgment of the offset threshold: If the change value of the clustering center exceeds the preset threshold, it is determined that the electrical fault is caused by the abnormality of the electrical equipment. Because the abnormal behavior of the electrical equipment usually leads to a significant shift in the current change pattern, and this shift is particularly obvious in the change of the clustering center.

[0103] Case of no significant shift: If the change value of the clustering center does not exceed the preset threshold, it is determined that the electrical fault is due to the aging of the line insulation material. The aging of the line insulation material usually leads to a slow degradation of the current change pattern, and this change is manifested as a small offset value in the change of the clustering center.

[0104] 4. Through the above clustering analysis and judgment of the clustering center offset, the fault location module can accurately distinguish whether the electrical fault is caused by the abnormality of the electrical equipment or by the aging of the line insulation material. The advantages of this method are as follows:

[0105] Data-driven judgment: Based on the clustering analysis of the current value change sequences, it can mine the fault feature patterns from a large amount of historical data, avoiding misjudgments caused by insufficient experience in traditional methods.

[0106] Dynamic adaptability: Through multiple clusterings and dynamic comparisons of the clustering centers, the module can adapt to the changes in fault characteristics in different scenarios, improving the robustness and reliability of the system.

[0107] Precise positioning and cause analysis: It can not only accurately locate the position where the fault occurs, but also further analyze the specific causes of the fault, providing a clear direction for subsequent repair and maintenance work.

[0108] 5. In practical applications, this fault judgment method based on clustering analysis can effectively handle complex and changeable transmission line fault scenarios. For example, in the current change sequence of a certain area, if the clustering center shows a significant deviation, the module can quickly judge that there is an abnormality in the electrical equipment and prompt the operation and maintenance personnel to check the relevant equipment. For the case where the deviation of the clustering center is small, the module can accurately identify that the line insulation material is aging and remind the operation and maintenance personnel to carry out line maintenance and replacement.

[0109] Through this optimized fault positioning and cause analysis method, the transmission line abnormal warning and fault positioning system of the present invention can significantly improve the operation efficiency and reliability of the power system, providing a strong guarantee for the safe and stable operation of the power system.

[0110] In a preferred case of this embodiment, the formula for judging whether the clustering center deviates is:

[0111]

[0112] Δc i =|c i,T+1 -c i,T |;

[0113] where S i is the set of data points, ∣S i ∣ is the number of data points in the category cluster, Δc i represents the change value of the i-th clustering center between two consecutive iterations; c i,T and c i,T+1 represent the values of the i-th clustering center at time T and time T + 1 respectively. When Δc i is greater than the preset threshold, it indicates that the clustering center has deviated.

[0114] The above has described a detailed description of an embodiment of the present invention, but the content described is only a preferred embodiment of the present invention and cannot be considered as limiting the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the application of the present invention should still fall within the scope covered by the patent of the present invention.

Claims

1. A power transmission line abnormality warning and fault location system, characterized in that: include: The area demarcation module is used to collect statistics on all warning equipment information in the line and divide the line into several sub-areas according to the distribution of warning equipment. Each warning equipment is associated with a corresponding sub-area. The regional positioning module is used to mark the sub-region related to the fault event as a pending area and mark the early warning equipment in the pending area as a pending equipment when a fault event occurs; A data acquisition module is used to collect monitoring data of the pending equipment within a set time window t before the fault, and pre-process the monitoring data, and determine whether the current fault event is caused by an electrical fault based on the monitoring data. If so, collect historical power consumption data of the pending area before the fault event occurs; A fault location module is used to normalize the historical power consumption data, build a time series model, extract the current value change sequence before the fault event occurs, and divide the current value change sequence into a number of sub-current value change sequences with a duration of t, mark the sub-current value change sequence within a set time window as a pending current value change sequence, cluster all the sub-current value change sequences, and check whether the category cluster to which the pending current value change sequence belongs is offset during the clustering process. If the offset exceeds a preset threshold, it is determined that the electrical fault is caused by an abnormality in the electrical equipment; If the deviation does not exceed the preset threshold, the electrical fault is determined to be aging of the line insulation material; The early warning equipment includes an electric energy meter, a circuit protection device, an electrical fault early warning sensor, an intelligent socket, and an electrical fault prevention and control device; the monitoring data of the early warning equipment includes a current signal fundamental wave, a current signal amplitude, a harmonic component, and a voltage; The early warning device calculates and generates power parameters according to the real-time voltage value and the real-time current value, and the power parameters include apparent power, active power, reactive power, and power factor; The process of the data acquisition module determining whether the fault event is caused by an electrical fault is as follows: Get the values ​​of all monitoring data and power parameters within the set time window before the fault event occurs. If the real-time current value amplitude is 0, it is considered that the fault event is not an electrical fault; if the real-time current value amplitude exceeds the preset standard range, it is considered that the fault event is an electrical fault; if the real-time current value amplitude is within the preset standard range, get any real-time monitoring data or power parameter. When the value of any real-time monitoring data or power parameter exceeds the corresponding preset standard range, it is considered that the fault event is an electrical fault.

2. A power transmission line abnormality warning and fault location system according to claim 1, characterized in that: The process of preprocessing the monitoring data by the data acquisition module is as follows: Verify the monitoring data collected from each warning device, supplement missing values ​​or outliers, convert date and time fields from strings to a unified format, detect and remove duplicate monitoring data records, remove outliers and invalid characters in monitoring data, and normalize the values ​​of different types of monitoring data.

3. A power transmission line abnormality warning and fault location system according to claim 1, characterized in that: The normalization process is: where x i represents the normalized monitoring data, X i represents the original monitoring data, X max Indicates the maximum value in the original monitoring data, X min Indicates the minimum value in the original monitoring data.

4. A power transmission line abnormality warning and fault location system according to claim 1, characterized in that: The process of the fault location module extracting the current value change sequence before the fault event occurs is as follows: Obtain the initial time series of historical electricity consumption data, collect the length of the initial time series and mark it as L, obtain the sliding step of the time series and mark it as S, the length of the current value change sequence is t, then the number of splits of the historical electricity consumption data is n=L / (S×t).

5. A power transmission line abnormality warning and fault location system according to claim 4, characterized in that: The process of clustering all sub-current value change sequences and checking whether the category cluster to which the pending current value change sequence belongs is offset during the clustering process is as follows: Set the control radius and the minimum similarity number. For the pending current value change sequence, detect the number m of similar sequences within the control radius. If m is less than the minimum similarity number, directly judge that the cause of the electrical fault is the abnormality of the electrical equipment. If m is greater than or equal to the minimum similarity number, put the pending current value change sequence into the category cluster, and put all the sub-current value change value sequences into the corresponding category cluster or mark them as abnormal in turn. Repeat the above process, extract the cluster centers of the category clusters after several clusterings, compare the number of cluster centers where the current change value sequence to be determined is located and the difference in the cluster center values, if the cluster center is offset by more than a preset threshold, it is determined that the electrical fault is caused by abnormal electrical equipment; If the deviation does not exceed the preset threshold, the electrical fault is determined to be aging of the line insulation material.

6. A power transmission line abnormality warning and fault location system according to claim 5, characterized in that: The formula for determining whether the cluster center is offset is: Δc i =|c i,T+1 -c i,T |; Among them, S i is a set of data points, |S i ∣ is the number of data points in the class cluster, Δc i represents the change value of the i-th cluster center between the previous and next two iterations; c i,T and c i,T+1 Represent the values ​​of the i-th cluster center at time T and time T+1 respectively. When Δc i When it is greater than the preset threshold, it indicates that the cluster center has shifted.

Images