Fault monitoring method, system and equipment for power distribution equipment and medium

Abstract

The invention relates to a fault monitoring method, system and device for power distribution equipment and a medium, and belongs to the technical field of power equipment maintenance. The fault monitoring method comprises the following steps: receiving operation monitoring data collected by a sensor deployed on power distribution equipment; preprocessing the operation monitoring data and performing feature extraction to obtain load fluctuation features; calling a pre-established abnormity identification model based on the type information of the power distribution equipment; inputting the load fluctuation characteristics into an anomaly recognition model to obtain an anomaly recognition result of the load fluctuation characteristics; based on a preset risk level mapping table, determining a corresponding fault risk level according to the exception identification result; and associating the fault risk level with the position information of the power distribution equipment, generating a fault early warning signal, and sending the fault early warning signal to the power maintenance terminal. According to the invention, the accuracy and maintenance efficiency of fault monitoring can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of power equipment maintenance, and in particular to a method, system, equipment and medium for fault monitoring of power distribution equipment. Background Art

[0002] Distribution equipment is an indispensable and important part of the power system, and its stable operation is directly related to the reliability of the power grid and the continuity of power supply. In industrial parks, urban power grids and special environments (such as high humidity and high temperature areas), distribution equipment is prone to overload, temperature abnormalities, equipment aging and other problems due to long-term operation or external environmental interference. These problems may lead to reduced equipment operating efficiency and even cause large-scale power outages. Therefore, how to effectively monitor and maintain distribution equipment, especially to quickly identify faults and take response measures, has become a key research direction in the field of power equipment management.

[0003] At present, common distribution equipment monitoring systems mainly rely on a combination of periodic inspections and fault alarms. Although this model can detect equipment failures to a certain extent, it has many limitations. For example, the operating status of distribution equipment is often affected by many factors, including current, voltage, load fluctuations, temperature changes, etc. When the number of distribution equipment is large, periodic inspections are often difficult to cover all equipment in a timely manner, and due to differences in equipment types and operating status, it is often difficult to capture key fault characteristics; fault alarms are usually triggered based on static thresholds, and cannot dynamically adapt to complex load environments and equipment operating conditions, and cannot accurately classify anomalies. These problems result in inaccurate fault identification of distribution equipment and long response times, further increasing the risk of power grid operation and the difficulty of maintenance.

[0004] Therefore, how to accurately identify anomalies and assess failure risks to improve monitoring accuracy and maintenance efficiency is a technical problem that needs to be solved urgently. Summary of the invention

[0005] In order to improve the accuracy of fault monitoring and maintenance efficiency and ensure the safe and reliable operation of the power grid, the present application provides a fault monitoring method, system, equipment and medium for power distribution equipment.

[0006] In a first aspect, the present application provides a method for fault monitoring of power distribution equipment, which adopts the following technical solution: A fault monitoring method for power distribution equipment, the fault monitoring method comprising: Receive operation monitoring data collected by sensors deployed on power distribution equipment; Preprocessing the operation monitoring data and performing feature extraction to obtain load fluctuation features; wherein the load fluctuation features include load fluctuation amplitude, fluctuation duration and fluctuation frequency; Based on the type information of the power distribution equipment, calling a pre-established abnormality recognition model; Inputting the load fluctuation feature into the abnormality recognition model to obtain an abnormality recognition result of the load fluctuation feature; determining a corresponding fault risk level according to the abnormality recognition result based on a preset risk level mapping table; The fault risk level is associated with the location information of the power distribution equipment, and a fault warning signal is generated and sent to the power maintenance terminal.

[0007] By adopting the above technical solution, the power distribution equipment is monitored in real time in multiple dimensions, and combined with the intelligent abnormality identification model and risk assessment mechanism, it is possible to accurately and quickly discover potential equipment failures and conduct graded warnings. Through the timely transmission of fault warning signals and the efficient response of maintenance terminals, the timeliness and accuracy of power distribution equipment fault handling are effectively improved, the risk of equipment operation is reduced, and the stability and reliability of the power distribution system are ensured. Compared with traditional technical solutions, this application has significant advantages in real-time, accuracy and intelligence, and provides a strong guarantee for the efficient operation and maintenance of power distribution equipment.

[0008] Optionally, the step of preprocessing the operation monitoring data and extracting features to obtain load fluctuation features includes: Performing data cleaning on the operation monitoring data; Perform time stamp alignment and normalization on the cleaned operation monitoring data; Acquire historical load data of the power distribution equipment, and obtain a reference load value based on an average value of the historical load data; Obtaining a load fluctuation amplitude sequence according to the reference load value and the normalized operation monitoring data; According to the load fluctuation amplitude sequence and the preset fluctuation amplitude threshold, determine the time period during which the load fluctuation amplitude exceeds the preset fluctuation amplitude threshold, and obtain the fluctuation duration and fluctuation frequency; The load fluctuation amplitude sequence, fluctuation duration and fluctuation frequency are structured to obtain load fluctuation characteristics.

[0009] By adopting the above technical solution, the load fluctuation characteristics are described based on a combination of multiple features (amplitude, duration, frequency), which improves the feature expression capability and achieves seamless connection with subsequent model input, ensuring the integrity of data flow and the accuracy of analysis.

[0010] Optionally, the method further includes a training step of the abnormality recognition model, wherein the training step includes: Acquire a historical sample data set; the historical sample data includes historical monitoring data corresponding to the device type of the power distribution device and a pre-marked abnormal result label; Preprocessing and feature extraction of the historical sample data set to obtain a feature matrix and corresponding abnormal result labels; Divide the historical sample data set after feature extraction into a training set and a test set; Inputting the feature matrix in the training set into a pre-built random forest network model to obtain a training result label; The training result label is compared with the abnormal result label pre-marked in the training set to obtain the training comparison result; the random forest network model is iterated according to the training comparison result and the loss function is calculated until the loss function meets the preset conditions to obtain the trained abnormality recognition model.

[0011] By adopting the above technical solutions, a random forest-based anomaly recognition model was established, which can automatically learn the failure mode and operation characteristics of the equipment, so as to accurately judge the abnormal operation of the distribution equipment, such as overload, voltage fluctuation, etc., using multi-dimensional characteristics, and provide risk assessment. This process is highly automated and intelligent, which can greatly improve the response speed and accuracy of equipment maintenance, effectively reduce manual intervention and failure losses, and enhance the reliability and stability of power equipment.

[0012] Optionally, after the step of obtaining the trained abnormality recognition model, the step further includes: Inputting the feature matrix in the test set into the trained anomaly recognition model to obtain a test result label; Comparing the abnormal result labels in the test set with the test result labels to obtain a test comparison result; Based on the test comparison results, the performance indicators of the trained anomaly recognition model are evaluated and iterated until a preset performance index is met or a preset number of iterations is reached, thereby obtaining the trained anomaly recognition model.

[0013] By adopting the above technical solutions, it is ensured that the model can not only perform well on the training data, but also maintain high performance on the test data in actual applications, ultimately providing accurate support for equipment failure prediction.

[0014] Optionally, after the step of generating a fault warning signal and sending it to the power maintenance terminal, the step further includes: Receiving operation monitoring data of the power distribution equipment after a preset period of time; Based on the abnormality recognition model, abnormality recognition is performed on the operation monitoring data after the preset time period to obtain an abnormality recognition result; It is determined whether the power distribution equipment has an abnormality after a preset period of time based on the abnormality identification result. If so, a maintenance verification prompt message is sent to the management terminal.

[0015] By adopting the above technical solution, continuous monitoring of the equipment operating status is added after the fault warning signal is sent to verify whether the abnormality persists or worsens, thus achieving dynamic tracking and risk review of equipment abnormalities, and realizing a closed-loop management system from early warning, monitoring to verification, helping the maintenance team to quickly respond to and handle abnormalities.

[0016] Optionally, after the step of determining that the power distribution equipment is abnormal after a preset time period, the method further includes: According to the type information of the power distribution equipment, real-time load data and location distribution data of backup equipment of the same type are obtained; based on the real-time load data and location distribution data, a comprehensive score of each backup equipment is obtained; The backup equipment with the largest comprehensive score is selected as the optimal backup equipment; Calculating the maximum load distribution amount of the optimal standby device according to the real-time load data of the optimal standby device; The load distribution of the power distribution equipment is adjusted according to the maximum load distribution amount, and part of the load of the power distribution equipment is switched to the optimal standby equipment.

[0017] By adopting the above technical solution, this application can not only quickly respond to single equipment failures, but also improve the overall reliability and operating efficiency of the power grid system through scientific backup equipment selection and load distribution, providing strong support for the intelligent operation and maintenance of complex power systems.

[0018] In a second aspect, the present application provides a fault monitoring system for power distribution equipment, which adopts the following technical solution: A fault monitoring system for power distribution equipment, the fault monitoring system comprising: A data receiving module is used to receive operation monitoring data collected by sensors deployed on the power distribution equipment; A data processing module, used to pre-process the operation monitoring data and perform feature extraction to obtain load fluctuation characteristics; wherein the load fluctuation characteristics include load fluctuation amplitude, fluctuation duration and fluctuation frequency; A model calling module, used to call a pre-established abnormality recognition model based on the type information of the power distribution equipment; an abnormality identification module, used for inputting the load fluctuation feature into the abnormality identification model to obtain an abnormality identification result of the load fluctuation feature; A risk level generating module, configured to determine a corresponding fault risk level according to the abnormality identification result based on a preset risk level mapping table; The fault warning module is used to associate the fault risk level with the location information of the power distribution equipment, generate a fault warning signal and send it to the power maintenance terminal.

[0019] Optionally, the fault monitoring system further includes: The data receiving module is also used to receive the operation monitoring data of the power distribution equipment after a preset time period; The abnormality identification module is further used to perform abnormality identification on the operation monitoring data after the preset time period based on the abnormality identification model to obtain an abnormality identification result; A judgment module, used to judge whether the power distribution equipment is abnormal after a preset time period according to the abnormality identification result, and if so, output the abnormality judgment result; The maintenance check prompt module is used to send maintenance check prompt information to the management terminal in response to the abnormality judgment result.

[0020] In a third aspect, the present application provides a computer device, which adopts the following technical solution: A computer device comprises a memory, a processor and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method as described in the first aspect.

[0021] In a fourth aspect, the present application provides a computer-readable storage medium, which adopts the following technical solution: A computer-readable storage medium stores a computer program that can be loaded by a processor and execute any one of the methods in the first aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a first flow chart of a method for fault monitoring of power distribution equipment according to one of the embodiments of the present application.

[0023] Figure 2 This is a second flow chart of a method for fault monitoring of power distribution equipment according to one of the embodiments of the present application.

[0024] Figure 3 It is a third flow chart of a method for fault monitoring of power distribution equipment according to one of the embodiments of the present application.

[0025] Figure 4 It is a fourth flow chart of a method for fault monitoring of power distribution equipment according to one of the embodiments of the present application.

[0026] Figure 5 It is a fifth flow chart of a method for fault monitoring of power distribution equipment according to one of the embodiments of the present application.

[0027] Figure 6 It is a sixth flow chart of a method for fault monitoring of power distribution equipment according to one of the embodiments of the present application. DETAILED DESCRIPTION

[0028] In order to make the purpose, technical solutions and advantages of this application more clear, the following Figure 1-6 It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0029] An embodiment of the present application discloses a fault monitoring method for power distribution equipment.

[0030] Reference Figure 1 , a fault monitoring method for power distribution equipment, the fault monitoring method comprising: Step S101, receiving operation monitoring data collected by sensors deployed on power distribution equipment; Among them, sensors (such as voltage sensors, current sensors, temperature sensors and load monitoring modules) arranged on the power distribution equipment collect operation monitoring data, such as voltage, current, temperature, load fluctuation value and operation time, etc. These sensors sample the equipment operation status at a fixed frequency, and transmit the collected data to the central processing system through the communication module, realizing real-time monitoring of the equipment operation status, ensuring the timeliness and comprehensiveness of the data, and providing high-quality basic data for subsequent analysis.

[0031] Step S102, preprocessing the operation monitoring data and extracting features to obtain load fluctuation features; wherein the load fluctuation features include load fluctuation amplitude, fluctuation duration and fluctuation frequency; Among them, the data collected by the sensor is preprocessed, including denoising, outlier removal and formatting; at the same time, the load fluctuation characteristic values ​​are extracted, including the load fluctuation amplitude (the percentage change between the real-time load and the benchmark load), the fluctuation duration (the accumulated time of load anomalies) and the fluctuation frequency (the number of times anomalies occur within a certain period of time).

[0032] It should be noted that the load fluctuation characteristics not only include the fluctuation characteristics of the load itself (such as power or load), but also cover the fluctuation characteristics of other key parameters (such as voltage, current, and temperature). These parameters are related to each other and together reflect the operating status of the equipment.

[0033] Step S103, based on the type information of the power distribution equipment, calling a pre-established abnormality recognition model; Each type of power distribution equipment has different operating characteristics and failure modes. Therefore, the system calls the corresponding abnormality recognition model according to the equipment type (such as transformer, distribution cabinet or load circuit breaker, etc.). The model is usually established based on the equipment's historical data, operating environment and failure mode, and can identify abnormal conditions that may occur during equipment operation (such as overload, overtemperature, voltage fluctuation, etc.).

[0034] For example, for transformer equipment, when the temperature exceeds 80°C, it is marked as "overheating risk"; when the fluctuation amplitude is continuously greater than 10% for more than 20 seconds, it is marked as "overload risk"; when the voltage fluctuation exceeds ±10%, it is marked as "voltage instability".

[0035] Step S104, inputting the load fluctuation feature into an abnormality recognition model to obtain an abnormality recognition result of the load fluctuation feature; Specifically, the load fluctuation characteristic values ​​(fluctuation amplitude, duration, frequency, etc.) are input into the anomaly recognition model, and the logic rules are gradually matched to determine whether the abnormal conditions are met and generate a marking result. The marking result includes the anomaly type (such as "overload anomaly") and the time period when the anomaly occurs.

[0036] For example, the input feature values ​​are fluctuation amplitude = 12%, duration = 30 seconds, and frequency = 3 times per minute. The abnormality recognition model performs abnormality analysis: the fluctuation amplitude exceeds 10%, the duration exceeds 20 seconds, and the "overload abnormality" rule is met. The final marking result is: the abnormality type is overload abnormality, and the abnormal duration is 20 seconds.

[0037] Step S105, based on the preset risk level mapping table, the corresponding fault risk level is determined according to the abnormality identification result; wherein, the abnormality identification result is mapped to the fault risk level through the preset risk level mapping table; the mapping table comprehensively calculates the risk level (such as low risk, medium risk and high risk) according to the abnormality type, duration, frequency and other parameters. Through risk quantification, the system can intuitively reflect the severity of the fault and provide a basis for subsequent maintenance priority.

[0038] For example, assuming that the abnormality identification model is marked as "overload abnormality", according to the risk level mapping table: if the overload abnormality lasts for more than 30 seconds and the load fluctuation amplitude is greater than 15%, the risk level is "high"; if the overload abnormality lasts between 10-30 seconds and the fluctuation amplitude is between 10%-15%, the risk level is "medium"; if the overload abnormality lasts for less than 10 seconds and the fluctuation amplitude is less than 10%, the risk level is "low".

[0039] Step S106, associating the fault risk level with the location information of the power distribution equipment, generating a fault warning signal and sending it to the power maintenance terminal.

[0040] The fault risk level is integrated with the location information and marking results of the equipment to generate a detailed warning signal, which is sent to the maintenance terminal through the communication module. In some embodiments, the fault warning signal may include information such as equipment ID, location, fault type, and fault risk level to provide maintenance personnel with clear action guidance.

[0041] In the above implementation, the power distribution equipment is monitored in real time in multiple dimensions, and combined with the intelligent abnormality identification model and risk assessment mechanism, it is possible to accurately and quickly discover potential equipment failures and conduct graded warnings. Through the timely transmission of fault warning signals and the efficient response of maintenance terminals, the timeliness and accuracy of power distribution equipment fault handling are effectively improved, the risk of equipment operation is reduced, and the stability and reliability of the power distribution system are ensured. Compared with traditional technical solutions, this application has significant advantages in real-time, accuracy and intelligence, and provides a strong guarantee for the efficient operation and maintenance of power distribution equipment.

[0042] Reference Figure 2 As an implementation method of step S102, the step of preprocessing the operation monitoring data and extracting features to obtain load fluctuation features includes: Step S201, cleaning the operation monitoring data; Specifically, data cleaning includes: denoising, which can use the sliding window averaging method to smooth sudden changes in data, for example, by taking the average of five consecutive data points to smooth load fluctuation data; outlier detection and elimination, which can use the statistically based triple standard deviation method (3σ method) to detect outliers, for example, when the load fluctuation value exceeds the normal range (such as ±50%), it is marked as abnormal and eliminated; data completion, for missing values ​​caused by sensor failure or communication interruption, interpolation methods (such as linear interpolation or previous value completion) can be used to fill in the missing data.

[0043] Step S202, aligning the timestamps of the cleaned operation monitoring data and performing normalization processing; Among them, the cleaned high-quality monitoring data stream is sampled at a uniform time interval (such as once per second) to ensure the synchronization of all sensor data, and the data of different dimensions (such as voltage and current) are normalized to the [0,1] interval to ensure the comparability of the data.

[0044] Step S203, obtaining historical load data of the power distribution equipment, and obtaining a reference load value based on an average value of the historical load data; Among them, the average value of the historical load data of the power distribution equipment in the past period of time (such as the average value in the last 30 minutes) can be extracted as the benchmark load value.

[0045] Step S204, obtaining a load fluctuation amplitude sequence according to the reference load value and the normalized operation monitoring data; The load fluctuation amplitude A at each time point is calculated according to the following formula: It can be understood that by integrating the load fluctuation amplitude at each time point, a load fluctuation amplitude sequence can be obtained; Step S205, according to the load fluctuation amplitude sequence and the preset fluctuation amplitude threshold, determine the time period in which the load fluctuation amplitude exceeds the preset fluctuation amplitude threshold, and obtain the fluctuation duration and fluctuation frequency; The time period during which the cumulative load fluctuation amplitude exceeds the preset fluctuation amplitude threshold (such as 10%) is recorded as the fluctuation duration; for example, if the load fluctuation amplitude is greater than 10% within three minutes, the fluctuation duration is 3 minutes. The number of times the load fluctuation amplitude exceeds the preset fluctuation amplitude threshold within a unit time (such as 1 minute) is counted and recorded as the fluctuation frequency; for example, if there are 5 load fluctuations within 1 minute, the fluctuation frequency is 5 times / minute.

[0046] Step S206, structurally processing the load fluctuation amplitude sequence, fluctuation duration and fluctuation frequency to obtain load fluctuation characteristics.

[0047] The extracted load fluctuation features are structured and output as input data for the subsequent abnormality recognition model. For example, the load fluctuation features are {amplitude sequence: [10%, 15%, 8%], fluctuation duration: 5 seconds, fluctuation frequency: 3 times / minute}.

[0048] In the above implementation, the load fluctuation characteristics are described based on a combination of multiple features (amplitude, duration, frequency), which improves the feature expression capability and achieves seamless connection with subsequent model input, ensuring the integrity of data flow and the accuracy of analysis.

[0049] Reference Figure 3 As an implementation of the anomaly recognition model, the training steps of the anomaly recognition model include: Step S301, obtaining a historical sample data set; the historical sample data includes historical monitoring data corresponding to the equipment type of the power distribution equipment and pre-marked abnormal result labels; Among them, historical monitoring data corresponding to the equipment types of distribution equipment (such as transformers, distribution cabinets, etc.) are collected, including multi-dimensional features such as voltage, current, temperature, load fluctuation value and timestamp, as well as corresponding pre-labeled abnormal result labels (such as "overload abnormality" and "voltage instability"); these data come from historical operation records or manual labeling by experts and are the basis for model training.

[0050] It can be understood that providing personalized, high-quality training data for the model based on the device type ensures that the training data can cover various historical working states associated with the device type.

[0051] Step S302, preprocessing and feature extraction of the historical sample data set to obtain a feature matrix and corresponding abnormal result labels; Among them, the historical sample data is preprocessed, including denoising, normalization and feature extraction, to form a feature matrix suitable for model input and the corresponding abnormal result label; the focus of feature extraction is to screen key parameters related to anomalies, such as load fluctuation amplitude, fluctuation duration and fluctuation frequency.

[0052] Step S303, dividing the historical sample data set after feature extraction into a training set and a test set; The data set is divided proportionally (such as 70% for training and 30% for testing), the training set is used for model fitting, and the test set is used to verify model performance to ensure that the trained model has generalization capabilities.

[0053] Step S304, inputting the feature matrix in the training set into the pre-built random forest network model to obtain the training result label; Specifically, the feature matrix of the training set is input into the pre-built random forest model, and the relationship between features and abnormal labels is learned by fitting the data one by one through decision trees, and the training result labels are generated. Among them, the random forest model is an integrated learning method, which obtains the final prediction result through the voting mechanism of multiple decision trees. In this process, the model automatically adjusts the parameters of the tree according to the training data, performs feature selection and splitting, until the final classification model is generated.

[0054] For example, assuming that the feature matrix in the training set includes current, temperature, load fluctuation amplitude, etc., the random forest model will learn the patterns of abnormal situations such as "overload abnormality" and "voltage instability" based on these features.

[0055] Step S305, comparing the training result label with the abnormal result label pre-labeled in the training set to obtain a training comparison result; After the model training is completed, the predicted labels output by the model are compared with the real labels in the training set to determine the difference between the model's predicted results and the actual labels. By comparing the training result labels with the real labels, the performance of the model can be analyzed and the deficiencies in the model training can be found, thus providing a basis for subsequent optimization and iteration.

[0056] Step S306, iterate the random forest network model according to the training comparison result and calculate the loss function until the loss function meets the preset conditions to obtain the trained anomaly recognition model.

[0057] Specifically, based on the training comparison results, the model evaluates the error between the predicted result and the true label by calculating the loss function (such as cross entropy loss or mean square error). The model adjusts the parameters according to the value of the loss function and performs iterative optimization. The training process continues until the loss function reaches the preset convergence condition (such as the loss is less than 0.01 or the number of iterations reaches a certain threshold).

[0058] For example, if the difference between the model's predicted "overload anomaly" label and the true label is large, the loss function will be high, and the model will be optimized by adjusting parameters to reduce the difference. Through repeated iterations and adjustments, the model can continuously optimize and gradually reduce the prediction error, and finally obtain accurate anomaly recognition capabilities.

[0059] In the above implementation, a random forest-based anomaly recognition model is established, which can automatically learn the failure mode and operation characteristics of the equipment, so as to accurately judge the abnormal operation of the distribution equipment, such as overload, voltage fluctuation, etc., using multi-dimensional characteristics, and provide risk assessment. This process is highly automated and intelligent, which can greatly improve the response speed and accuracy of equipment maintenance, effectively reduce manual intervention and failure losses, and enhance the reliability and stability of power equipment.

[0060] Reference Figure 4 As a further implementation of the abnormality recognition model, after the step of obtaining the trained abnormality recognition model, the method further includes: Step S401, input the feature matrix in the test set into the trained anomaly recognition model to obtain a test result label; wherein, the feature matrix in the test set is input into the trained model, and the model outputs the test result label according to the previously learned rules. This process evaluates the prediction ability of the trained model for new data (unseen data), can verify the generalization ability of the trained model, and evaluate the effect of the model in practical applications.

[0061] Step S402, comparing the abnormal result labels in the test set with the test result labels to obtain a test comparison result; Step S403, based on the test comparison result, the performance index of the trained anomaly recognition model is evaluated and iterated until a preset performance index is met or a preset number of iterations is reached, thereby obtaining a trained anomaly recognition model.

[0062] Specifically, by comparing the test result labels with the real labels in the test set, the model's performance indicators, such as accuracy, precision, recall, F1 score, etc., are calculated; based on these evaluation indicators, the model may need to be further adjusted and optimized until it meets the preset performance indicators (such as accuracy reaching more than 90%). For example, if the test results show that the model is 85% accurate in predicting "overload anomalies", the model can be further optimized based on other indicators such as precision and recall.

[0063] In the above implementation, it is ensured that the model can not only perform well on the training data, but also maintain high performance on the test data in actual applications, and ultimately provide accurate support for equipment failure prediction.

[0064] Reference Figure 5 As a further implementation of the fault monitoring method, after the step of generating a fault warning signal and sending it to the power maintenance terminal, the method further includes: Step S501, receiving operation monitoring data of the power distribution equipment after a preset time period; Among them, after the fault warning signal is sent to the maintenance terminal to prompt the maintenance personnel to perform maintenance, the system needs to continuously monitor the operating status of the power distribution equipment. The sensor collects the real-time operating data of the equipment again according to the preset time (such as 30 minutes) to ensure that the changing trend of the equipment operation can be dynamically grasped. These data may include parameters such as voltage, current, temperature, and load fluctuation.

[0065] Step S502, based on the abnormality recognition model, performing abnormality recognition on the operation monitoring data after a preset time period to obtain an abnormality recognition result; The collected operation monitoring data is input into the anomaly recognition model, and the operation status of the equipment is analyzed again through the model. This process is used to identify abnormal features in the newly collected data and output abnormal categories and related information.

[0066] Step S503, judging whether the power distribution equipment has an abnormality after a preset time period according to the abnormality identification result, if so, jumping to step S504; if not, no operation is performed; Step S504: Send maintenance verification prompt information to the management terminal.

[0067] Among them, by judging whether the equipment is still in an abnormal state after a preset period of time, if the abnormal state persists, it means that the equipment problem has not been alleviated or has worsened, and the system needs to issue further maintenance verification prompts. The maintenance prompt information includes the latest operating data of the equipment, abnormal identification results, risk level and recommended operations, providing managers with intuitive maintenance verification basis.

[0068] It is understandable that after sending the fault warning signal to the maintenance terminal for a preset period of time, if the power distribution equipment still has abnormal conditions, it may be caused by the maintenance personnel failing to perform maintenance in a timely manner or the maintenance measures being ineffective, and the management personnel need to conduct further maintenance inspections.

[0069] In the above implementation, after sending the fault warning signal, continuous monitoring of the equipment operating status is added to verify whether the abnormality persists or worsens, thereby realizing dynamic tracking and risk review of equipment abnormalities, and realizing a closed-loop management system from early warning, monitoring to verification, helping the maintenance team to quickly respond to and handle abnormalities.

[0070] Reference Figure 6 As a further implementation of the fault monitoring method, after the step of determining that the power distribution equipment is abnormal after a preset time period, the method further includes: Step S601, acquiring real-time load data and location distribution data of backup equipment of the same type according to type information of the power distribution equipment; Among them, when the abnormality of the faulty equipment persists, the system first obtains the real-time operating status information of the same type of backup equipment, including real-time load data and location distribution data, to provide input basis for subsequent analysis. The real-time load data of the backup equipment reflects its current operating pressure, while the location distribution data determines the physical distance of the backup equipment response.

[0071] Specifically, the power grid monitoring system is used to extract information about standby equipment of the same type as the faulty equipment, including real-time load rate (such as the percentage of current load relative to rated load) and geographic location (such as GPS coordinates or topological location).

[0072] Step S602, based on the real-time load data and location distribution data, obtain a comprehensive score for each backup device; wherein, through the comprehensive scoring method, the system performs quantitative calculations based on multi-dimensional factors such as the load rate, geographical location, and reliability of the backup device, providing a standard for screening the optimal backup device. The scoring method ensures the optimal selection of the backup device in terms of load capacity and response speed, making the selection of the optimal backup device more scientific and reasonable In one embodiment of the present application, the scoring formula is: S=w1·F load +w2·F distance +w3·F reliability ; In the above formula, S is the comprehensive score, F load Load score, based on the equipment load rate (the lower the load rate, the higher the score); F distance The distance score is based on the distance between the device and the faulty device (the closer the distance, the higher the score); F reliability It is a reliability score based on the historical failure rate or maintenance record of the equipment (e.g. the lower the failure rate, the higher the score). w1, w2, and w3 are weight factors, reflecting the importance of each factor.

[0073] Step S603, selecting the standby device with the largest comprehensive score as the optimal standby device; Among them, according to the comprehensive scoring results, the backup equipment with the highest score is selected as the optimal backup equipment to ensure the execution effect of the load distribution plan.

[0074] Step S604, calculating the maximum load distribution amount of the optimal standby device according to the real-time load data of the optimal standby device; Among them, according to the rated load and current load of the optimal backup equipment, the maximum load that can be taken over is calculated to ensure that the load distribution does not exceed the safe operating range of the backup equipment to avoid causing secondary failures.

[0075] Specifically, the calculation formula is: L max =L rated -L current ; In the above formula, L max is the maximum load that can be taken over, L rated is the rated load of the standby equipment, L current The current load of the standby device.

[0076] Step S605: adjust the load distribution of the power distribution equipment according to the maximum load distribution amount, and switch part of the load of the power distribution equipment to the optimal standby equipment.

[0077] Among them, according to the maximum load distribution amount of the optimal backup equipment, part of the load of the faulty equipment is transferred to the backup equipment to reduce the operating pressure of the faulty equipment.

[0078] In some embodiments, the step of adjusting the load distribution of the power distribution equipment includes: calculating the amount of load that the power distribution equipment needs to transfer; if the amount of load that needs to be transferred is less than or equal to the maximum load distribution amount of the optimal backup equipment, then directly transfer it; if the amount of load that needs to be transferred exceeds the maximum distribution amount of the optimal backup equipment, then transfer it in batches or seek other backup equipment for collaborative processing.

[0079] In the above implementation mode, the present application can not only quickly respond to single equipment failures, but also improve the overall reliability and operating efficiency of the power grid system through scientific backup equipment selection and load distribution, and provide strong support for the intelligent operation and maintenance of complex power systems.

[0080] The embodiment of the present application also discloses a fault monitoring system for power distribution equipment.

[0081] A fault monitoring system for power distribution equipment, the fault monitoring system comprising: A data receiving module is used to receive operation monitoring data collected by sensors deployed on the power distribution equipment; A data processing module is used to pre-process the operation monitoring data and extract features to obtain load fluctuation features; wherein the load fluctuation features include load fluctuation amplitude, fluctuation duration and fluctuation frequency; A model calling module, used to call a pre-established abnormality recognition model based on the type information of the power distribution equipment; An abnormality recognition module, used for inputting the load fluctuation characteristics into the abnormality recognition model to obtain an abnormality recognition result of the load fluctuation characteristics; A risk level generation module is used to determine the corresponding fault risk level according to the abnormality identification result based on a preset risk level mapping table; The fault warning module is used to associate the fault risk level with the location information of the distribution equipment, generate a fault warning signal and send it to the power maintenance terminal.

[0082] In the above implementation, the operation monitoring data of the power distribution equipment is received in real time, and combined with data preprocessing and feature extraction, the load fluctuation characteristics of the equipment, such as fluctuation amplitude, duration and frequency, can be accurately captured. By matching the extracted features with the anomaly recognition model, the system can quickly identify abnormal conditions in the operation of the equipment, such as overload, voltage instability and other problems, and generate risk levels based on the abnormal conditions to quantify the severity of the fault. This technical solution not only achieves high efficiency and accuracy in anomaly detection, but also provides a scientific decision-making basis for the maintenance team, which helps to shorten the fault response time and reduce the risk of interruption in power grid operation.

[0083] This application generates intuitive and clear fault warning signals by associating fault risk levels with equipment location information, and transmits them to the maintenance terminal in a timely manner. This design further enhances the intelligence and operability of the system, allowing maintenance personnel to quickly locate problematic equipment, understand its risk level, and develop corresponding treatment plans. The practical significance of this system is that it significantly improves the operation and monitoring efficiency of distribution equipment, reduces the labor cost of fault maintenance, and at the same time ensures the safety and reliability of the power system, providing advanced technical support for the intelligent management of complex power grids.

[0084] As a further implementation of the fault monitoring system, the fault monitoring system further includes: The data receiving module is also used to receive the operation monitoring data of the power distribution equipment after a preset time period; The anomaly recognition module is also used to perform anomaly recognition on the operation monitoring data after a preset time based on the anomaly recognition model to obtain an anomaly recognition result; A judgment module is used to judge whether the power distribution equipment is abnormal after a preset time period according to the abnormality identification result, and if so, output the abnormality judgment result; The maintenance verification prompt module is used to send maintenance verification prompt information to the management terminal in response to the abnormality judgment result.

[0085] In the above implementation, by continuously receiving the operation monitoring data of the power distribution equipment after a preset period of time, and analyzing the new data in combination with the abnormality recognition model, the dynamic tracking of the equipment status is effectively achieved. By judging whether the equipment abnormality persists, the system can promptly confirm the severity and evolution trend of the equipment operation problem, and automatically generate maintenance verification prompt information when necessary, and quickly notify the management terminal. This technical solution not only enhances the accuracy and timeliness of fault identification, but also constructs a closed-loop process from fault identification to maintenance prompts through dynamic monitoring and automated judgment, avoiding secondary faults that may be caused by continued unresolved problems.

[0086] This application provides an intelligent, real-time decision-making support solution for power grid operation and maintenance, effectively shortening the detection and response time of abnormal problems, improving the operating reliability and maintenance efficiency of distribution equipment, while reducing the workload of maintenance personnel, and providing a solid guarantee for the efficient and safe operation of the power system.

[0087] A fault monitoring system for power distribution equipment in an embodiment of the present application can implement any of the above-mentioned fault monitoring methods for power distribution equipment, and the specific working process of each module in the fault monitoring system can refer to the corresponding process in the above-mentioned method embodiment.

[0088] In the several embodiments provided in this application, it should be understood that the provided methods and systems can be implemented in other ways. For example, the system embodiments described above are only illustrative; for example, the division of a certain module is only a logical function division, and there may be other division methods in actual implementation, such as multiple modules can be combined or integrated into another system, or some features can be ignored or not executed.

[0089] The embodiment of the present application also discloses a computer device.

[0090] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, a fault monitoring method for a power distribution device as described above is implemented.

[0091] The embodiment of the present application also discloses a computer-readable storage medium.

[0092] A computer-readable storage medium stores a computer program that can be loaded by a processor and execute any one of the above-mentioned methods for fault monitoring of power distribution equipment.

[0093] Among them, computer-readable storage media can be any tangible medium that contains or stores a program, which can be used by or in conjunction with an instruction execution system, apparatus or device; the program code contained on the computer-readable medium can be transmitted using any appropriate medium, including but not limited to wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0094] It should be noted that in the above embodiments, the description of each embodiment has different emphases, and for the parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0095] The above are all preferred embodiments of the present application, and are not intended to limit the protection scope of the present application. Any feature disclosed in this specification (including the abstract and drawings), unless otherwise stated, can be replaced by other equivalent or alternative features with similar purposes. That is, unless otherwise stated, each feature is only an example of a series of equivalent or similar features.

Claims

1. A method for monitoring faults of power distribution equipment, characterized in that: The fault monitoring method comprises: Receive operation monitoring data collected by sensors deployed on power distribution equipment; Preprocessing the operation monitoring data and performing feature extraction to obtain load fluctuation features; wherein the load fluctuation features include load fluctuation amplitude, fluctuation duration and fluctuation frequency; Based on the type information of the power distribution equipment, calling a pre-established abnormality recognition model; Inputting the load fluctuation feature into the abnormality recognition model to obtain an abnormality recognition result of the load fluctuation feature; Based on a preset risk level mapping table, determining a corresponding fault risk level according to the abnormality identification result; The fault risk level is associated with the location information of the power distribution equipment, and a fault warning signal is generated and sent to the power maintenance terminal.

2. A method for fault monitoring of power distribution equipment according to claim 1, characterized in that: The steps of preprocessing the operation monitoring data and extracting features to obtain load fluctuation features include: Performing data cleaning on the operation monitoring data; Perform time stamp alignment and normalization on the cleaned operation monitoring data; Acquire historical load data of the power distribution equipment, and obtain a reference load value based on an average value of the historical load data; Obtaining a load fluctuation amplitude sequence according to the reference load value and the normalized operation monitoring data; According to the load fluctuation amplitude sequence and the preset fluctuation amplitude threshold, determine the time period during which the load fluctuation amplitude exceeds the preset fluctuation amplitude threshold, and obtain the fluctuation duration and fluctuation frequency; The load fluctuation amplitude sequence, fluctuation duration and fluctuation frequency are structured to obtain load fluctuation characteristics.

3. A method for fault monitoring of power distribution equipment according to claim 1, characterized in that: The method further includes a training step of the abnormality recognition model, wherein the training step includes: Acquire a historical sample data set; the historical sample data includes historical monitoring data corresponding to the device type of the power distribution device and a pre-marked abnormal result label; Preprocessing and feature extraction of the historical sample data set to obtain a feature matrix and corresponding abnormal result labels; Divide the historical sample data set after feature extraction into a training set and a test set; Inputting the feature matrix in the training set into a pre-built random forest network model to obtain a training result label; Compare the training result label with the abnormal result label pre-marked in the training set to obtain a training comparison result; The random forest network model is iterated according to the training comparison result and the loss function is calculated until the loss function meets the preset conditions, thereby obtaining the trained anomaly recognition model.

4. A method for fault monitoring of power distribution equipment according to claim 3, characterized in that: After the step of obtaining the trained abnormality recognition model, the method further includes: Inputting the feature matrix in the test set into the trained anomaly recognition model to obtain a test result label; Comparing the abnormal result labels in the test set with the test result labels to obtain a test comparison result; Based on the test comparison results, the performance indicators of the trained anomaly recognition model are evaluated and iterated until a preset performance index is met or a preset number of iterations is reached, thereby obtaining the trained anomaly recognition model.

5. A method for monitoring faults of power distribution equipment according to any one of claims 1 to 4, characterized in that: After the step of generating a fault warning signal and sending it to the power maintenance terminal, the method further includes: Receiving operation monitoring data of the power distribution equipment after a preset period of time; Based on the abnormality recognition model, abnormality recognition is performed on the operation monitoring data after the preset time period to obtain an abnormality recognition result; It is determined whether the power distribution equipment has an abnormality after a preset period of time based on the abnormality identification result. If so, a maintenance verification prompt message is sent to the management terminal.

6. A method for fault monitoring of power distribution equipment according to claim 5, characterized in that: After the step of determining that the power distribution equipment is abnormal after a preset time, the step further includes: According to the type information of the power distribution equipment, real-time load data and location distribution data of backup equipment of the same type are obtained; Based on the real-time load data and location distribution data, a comprehensive score of each backup device is obtained; The backup equipment with the largest comprehensive score is selected as the optimal backup equipment; Calculating the maximum load distribution amount of the optimal standby device according to the real-time load data of the optimal standby device; The load distribution of the power distribution equipment is adjusted according to the maximum load distribution amount, and part of the load of the power distribution equipment is switched to the optimal standby equipment.

7. A fault monitoring system for power distribution equipment, characterized in that: The fault monitoring system comprises: A data receiving module is used to receive operation monitoring data collected by sensors deployed on the power distribution equipment; A data processing module, used to pre-process the operation monitoring data and perform feature extraction to obtain load fluctuation characteristics; wherein the load fluctuation characteristics include load fluctuation amplitude, fluctuation duration and fluctuation frequency; A model calling module, used to call a pre-established abnormality recognition model based on the type information of the power distribution equipment; an abnormality identification module, used for inputting the load fluctuation feature into the abnormality identification model to obtain an abnormality identification result of the load fluctuation feature; A risk level generating module, configured to determine a corresponding fault risk level according to the abnormality identification result based on a preset risk level mapping table; The fault warning module is used to associate the fault risk level with the location information of the power distribution equipment, generate a fault warning signal and send it to the power maintenance terminal.

8. A fault monitoring system for power distribution equipment according to claim 7, characterized in that: The fault monitoring system further comprises: The data receiving module is also used to receive the operation monitoring data of the power distribution equipment after a preset time period; The abnormality identification module is further used to perform abnormality identification on the operation monitoring data after the preset time period based on the abnormality identification model to obtain an abnormality identification result; A judgment module, used to judge whether the power distribution equipment is abnormal after a preset time period according to the abnormality identification result, and if so, output the abnormality judgment result; The maintenance check prompt module is used to send maintenance check prompt information to the management terminal in response to the abnormality judgment result.

9. A computer device, characterized in that: The method comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method according to any one of claims 1 to 6 is implemented.

10. A computer-readable storage medium, characterized in that: A computer program is stored which can be loaded by a processor and execute the method according to any one of claims 1 to 6.

Images