An abnormality alarm method of a power distribution system and related device

Abstract

The application discloses an abnormality alarm method of a power distribution system and related devices, and relates to the technical field of data analysis, which comprises the following steps: preprocessing monitoring data of power distribution equipment of the power distribution system, performing current phase vector analysis based on the preprocessed monitoring data; performing operation loss analysis based on the monitoring data combined with environmental parameters; analyzing the correlation between the operation states of the power distribution equipment based on an operation state matrix; performing load state analysis of the power distribution system based on the correlation combined with the monitoring data; performing abnormality occurrence uncertainty analysis of the power distribution system based on operation loss data and load state data; and performing abnormality risk assessment of the power distribution system based on current phase vector data, operation loss data, load state data and abnormality occurrence uncertainty data to determine whether alarm information needs to be sent. The application effectively improves the accuracy of abnormality risk assessment and avoids false positives and false negatives of abnormality alarm events.

Description

Technical Field

[0001] This invention relates to the field of data analysis technology, and in particular to an abnormal alarm method and related device for a power distribution system. Background Technology

[0002] Power distribution systems are an indispensable part of production and daily life, distributing power to the necessary equipment. To ensure the safety of these systems, monitoring is essential for timely anomaly alerts, preventing serious impacts on power system operation. Load status analysis (CSSA) is a crucial step in CSSA monitoring and analysis. Currently, load status is typically determined by comparing monitoring data with indicator data. However, this method relies heavily on the expertise of personnel, compromising accuracy. Furthermore, current CSSA neglects the interrelationships between the operating states of distribution equipment, resulting in insufficient data support and a high number of false alarms and missed alarms. Moreover, most current CSSA anomaly risk assessments rely solely on load status or operating losses, failing to consider all factors comprehensively and thus lacking accurate results, leading to frequent errors and omissions in CSSA anomaly alerts. Summary of the Invention

[0003] The purpose of this invention is to overcome the shortcomings of the prior art. This invention provides an abnormal alarm method and related device for power distribution systems, which effectively improves the accuracy of abnormal risk assessment and avoids false alarms and missed alarms.

[0004] To address the aforementioned technical problems, this invention provides an abnormal alarm method for a power distribution system, the method comprising:

[0005] The monitoring data of the power distribution equipment in the power distribution system is preprocessed to obtain preprocessed monitoring data, and current phase vector analysis is performed based on the preprocessed monitoring data to obtain current phase vector data.

[0006] Operational loss data is obtained by combining preprocessed monitoring data with environmental parameters to analyze operational loss.

[0007] Analysis of the correlation between the operating states of power distribution equipment based on the operating state matrix;

[0008] Based on the correlation relationship and the preprocessed monitoring data, the load status analysis of the power distribution system is carried out to obtain load status data.

[0009] Uncertainty analysis of anomalies in power distribution systems is performed based on operational loss data and load status data to obtain anomaly occurrence uncertainty data.

[0010] Anomaly risk assessment of the power distribution system is performed based on current phase vector data, operating loss data, load status data, and anomaly occurrence uncertainty data. Anomaly risk assessment results are obtained, and it is determined whether an alarm message needs to be issued based on the anomaly risk assessment results.

[0011] Optionally, the monitoring data of the power distribution equipment in the power distribution system is preprocessed to obtain preprocessed monitoring data, and current phase vector analysis is performed based on the preprocessed monitoring data to obtain current phase vector data, including:

[0012] The monitoring data is cleaned to obtain cleaned monitoring data.

[0013] The monitoring data after data cleaning is integrated to obtain preprocessed monitoring data.

[0014] Extract current vector data and current values ​​from the preprocessed monitoring data, and perform amplitude correction on the current vector data based on the amplitude correction coefficient table to obtain corrected current vector data;

[0015] A current flow direction matrix is ​​constructed based on the corrected current vector data, and current phase vector analysis is performed based on the current flow direction matrix, the corrected current vector data, and the current values ​​to obtain current phase vector data.

[0016] Optionally, the step of performing operational loss analysis based on preprocessed monitoring data and environmental parameters to obtain operational loss data includes:

[0017] Determine the environmental and climatic characteristics of the power distribution equipment, and conduct an environmental and climatic state impact analysis of the power distribution equipment based on the environmental and climatic characteristics to obtain environmental and climatic state impact data.

[0018] Determine the environmental parameters of the environment where the power distribution equipment is located and the range of safe parameters for normal operation of the power distribution equipment, and calculate the target difference between the pre-processed monitoring data and the endpoints of the safe parameter range;

[0019] The loss coefficient is determined based on environmental parameters and the actual usage time of the power distribution equipment.

[0020] Operational loss analysis is conducted based on target difference, loss coefficient, and environmental and climatic conditions to obtain operational loss data.

[0021] Optionally, the analysis of the correlation between the operating states of power distribution equipment based on the operating state matrix includes:

[0022] An operation status matrix is ​​constructed based on the monitoring index data of power distribution equipment, and cluster analysis is performed on the operation status matrix to obtain the cluster labels of the operation status matrix;

[0023] Construct the corresponding topology diagram based on the connection relationships of power distribution equipment;

[0024] The association relationships between the operating states of power distribution equipment are determined by combining cluster labeling based on the operating state matrix and topology graph with a frequent itemset algorithm that mines association rules.

[0025] Optionally, the load status analysis of the power distribution system based on correlation and preprocessed monitoring data to obtain load status data includes:

[0026] Load fluctuation feature data is obtained by extracting load fluctuation features from the preprocessed monitoring data.

[0027] The initial load factor of the power distribution system is determined based on the preprocessed monitoring data, and the mutual influence factor between power distribution equipment is determined based on the correlation between the operating states of the power distribution equipment.

[0028] The target load factor is determined based on the mutual influence coefficient and the initial load factor. Then, the load state analysis of the power distribution system is performed based on the target load factor and load fluctuation characteristic data, combined with the load state analysis model, to obtain load state data.

[0029] Optionally, the uncertainty analysis of power distribution system anomalies based on operating loss data and load state data, to obtain anomaly uncertainty data, includes:

[0030] The state transition rate is determined based on the number of transitions from normal operating state to abnormal state and the number of transitions from abnormal state to normal operating state in the power distribution system.

[0031] Based on the historical operating parameters of the distribution network system, the time-period parameter probability distribution is determined, and based on the time-period parameter probability distribution and state transition rate, an uncertainty analysis model for anomaly occurrence based on long short-term memory network is constructed.

[0032] Based on operating loss data and load status data, an anomaly occurrence uncertainty analysis model based on long short-term memory network is used to perform anomaly occurrence uncertainty analysis on the power distribution system to obtain anomaly occurrence uncertainty data.

[0033] Optionally, the step of performing anomaly risk assessment of the power distribution system based on current phase vector data, operating loss data, load status data, and anomaly occurrence uncertainty data to obtain anomaly risk assessment results, and determining whether to issue alarm information based on the anomaly risk assessment results, includes:

[0034] An anomaly risk coefficient for the power distribution system is generated using an anomaly risk assessment model based on current phase vector data, operating loss data, load status data, and anomaly occurrence uncertainty data.

[0035] If the abnormal risk coefficient of the power distribution system is greater than or equal to the preset abnormal risk threshold, an alarm message will be issued.

[0036] In addition, the present invention also provides an abnormal alarm device for a power distribution system, the device comprising:

[0037] Current phase vector module: used to preprocess the monitoring data of the power distribution equipment in the power distribution system, obtain the preprocessed monitoring data, and perform current phase vector analysis based on the preprocessed monitoring data to obtain current phase vector data;

[0038] Operational Loss Analysis Module: Used to perform operational loss analysis based on preprocessed monitoring data and environmental parameters to obtain operational loss data;

[0039] The correlation analysis module is used to analyze the correlation between the operating states of power distribution equipment based on the operating state matrix.

[0040] Load status analysis module: used to perform load status analysis of the power distribution system based on correlation and preprocessed monitoring data to obtain load status data;

[0041] Uncertainty Analysis Module: Used to perform uncertainty analysis on the occurrence of anomalies in the power distribution system based on operating loss data and load status data, and to obtain uncertainty data on the occurrence of anomalies;

[0042] Anomaly Alarm Judgment Module: This module is used to perform anomaly risk assessment of the power distribution system based on current phase vector data, operating loss data, load status data, and anomaly occurrence uncertainty data. It obtains the anomaly risk assessment results and determines whether an alarm message needs to be issued based on these results.

[0043] In addition, the present invention also provides an electronic device, which includes a processor and a memory. The memory is used to store instructions, and the processor is used to call the instructions in the memory to cause the electronic device to execute the above-mentioned abnormal alarm method for the power distribution system.

[0044] In addition, the present invention also provides a computer-readable storage medium that stores computer instructions that, when executed on an electronic device, cause the electronic device to perform the above-described abnormal alarm method for a power distribution system.

[0045] In this embodiment of the invention, operational loss analysis is performed based on preprocessed monitoring data combined with environmental parameters, improving the reliability of the operational loss analysis. Load status analysis of the power distribution system is performed based on correlations and preprocessed monitoring data, incorporating the correlations between the operating states of power distribution equipment, enabling a more specific and comprehensive analysis of the power distribution system's load status. Uncertainty analysis of power distribution system anomalies is performed based on operational loss data and load status data. Considering the inherent uncertainty in the occurrence of power distribution system anomalies, incorporating uncertainty analysis provides a basis for assessing anomaly risks. Anomaly risk assessment of the power distribution system is performed based on current phase vector data, operational loss data, load status data, and anomaly occurrence uncertainty data. By considering sufficiently comprehensive factors, the accuracy of anomaly risk assessment is effectively improved, avoiding false alarms and missed alarms. Attached Figure Description

[0046] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0047] Figure 1 This is a flowchart illustrating the abnormal alarm method for a power distribution system in an embodiment of the present invention.

[0048] Figure 2 This is a flowchart illustrating an abnormal alarm method for a power distribution system according to another embodiment of the present invention.

[0049] Figure 3 This is a schematic diagram of the structural composition of the abnormal alarm device of the power distribution system in an embodiment of the present invention;

[0050] Figure 4 This is a schematic diagram of the structural composition of the electronic device in an embodiment of the present invention. Detailed Implementation

[0051] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0052] Example 1

[0053] Please see Figure 1 , Figure 1 This is a flowchart illustrating an abnormal alarm method for a power distribution system according to an embodiment of the present invention. The method includes:

[0054] S11: Preprocess the monitoring data of the power distribution equipment in the power distribution system to obtain preprocessed monitoring data, and perform current phase vector analysis based on the preprocessed monitoring data to obtain current phase vector data;

[0055] In the specific implementation of this invention, the preprocessing of monitoring data of power distribution equipment in the power distribution system to obtain preprocessed monitoring data, and the performance of current phase vector analysis based on the preprocessed monitoring data to obtain current phase vector data, includes: performing data cleaning on the monitoring data to obtain cleaned monitoring data; performing data integration on the cleaned monitoring data to obtain preprocessed monitoring data; extracting current vector data and current values ​​from the preprocessed monitoring data, and performing amplitude correction on the current vector data based on an amplitude correction coefficient table to obtain corrected current vector data; constructing a current flow direction matrix based on the corrected current vector data, and performing current phase vector analysis based on the current flow direction matrix, the corrected current vector data, and the current values ​​to obtain current phase vector data.

[0056] Specifically, monitoring data from various power distribution devices in the power distribution system are collected through a combination of sensors. These devices include distribution cabinets, generators, transformers, power lines, circuit breakers, low-voltage switchgear, distribution panels, switch boxes, and control boxes. The monitoring data includes temperature, current values, current vector data, voltage, operating time, and load fluctuation values ​​at various times. Data cleaning is performed on the monitoring data to remove outliers, spaces, and garbled characters, resulting in cleaned monitoring data. This cleaned data is then integrated into a unified data structure, and data verification is performed to confirm its integrity and consistency. This completes the data preprocessing, resulting in preprocessed monitoring data. Current vector data and current values ​​are extracted from the preprocessed monitoring data. Since the collected current vector values ​​do not always exhibit a linear transformation within the full-scale range of the instrument transformer, there is a significant error in the collected current vector values, which needs to be corrected. The amplitude correction coefficient table is used to correct the amplitude of the current vector data to obtain corrected current vector data. An amplitude correction experiment is conducted using a high-precision current meter and an experimental instrument transformer. The experimental instrument transformer is of the same type as the instrument transformer in the power distribution equipment. An amplitude correction coefficient table is constructed based on the ratio of the current amplitude output by the high-precision current meter to the sampling current amplitude collected by the experimental instrument transformer. Correction coefficients are matched in the amplitude correction coefficient table according to the current vector data. The current vector data is then corrected according to the correction coefficients to obtain corrected current vector data, thus avoiding significant errors in subsequent data analysis. A current flow direction matrix is ​​constructed based on the corrected current vector data. The power distribution system is divided into several distribution areas. The flow direction relationship between the current and the distribution area is determined according to the corrected current vector data. For example, if a current flows into a distribution area, the matrix element is set to 1; if a current flows out of a distribution area, the matrix element is set to zero. The current flow direction matrix is ​​constructed, and current phase vector analysis is performed based on the current flow direction matrix, the corrected current vector data, and the current value. The corrected current vector data is multiplied by the current flow direction matrix to obtain the product value. The product value is multiplied by the current value, and the vector is taken from the product to obtain the current phase vector at each time moment, that is, the current phase vector data. The current phase vector data can help analyze the degree of anomaly in the system.

[0057] S12: Based on the preprocessed monitoring data and environmental parameters, perform operational loss analysis to obtain operational loss data;

[0058] In the specific implementation of this invention, the step of performing operational loss analysis based on preprocessed monitoring data and environmental parameters to obtain operational loss data includes: determining the environmental climate characteristics of the power distribution equipment, and performing an environmental climate state impact analysis on the power distribution equipment based on the environmental climate characteristics to obtain environmental climate state impact data; determining the environmental parameters of the environment in which the power distribution equipment is located and the safe parameter range for normal operation of the power distribution equipment, and calculating the target difference between the preprocessed monitoring data and the endpoints of the safe parameter range; determining the loss coefficient based on the environmental parameters and the actual usage time of the power distribution equipment; and performing operational loss analysis based on the target difference, the loss coefficient, and the environmental climate state impact data to obtain operational loss data.

[0059] Specifically, the environmental climate characteristics of the power distribution equipment are determined. These environmental climate characteristics include the cumulative operating time of the power distribution equipment under different climates, the proportion of power distribution equipment failures in the total number of failures under different climates, and the data on the changes in the operating status of the power distribution equipment under different climates. Based on these environmental climate characteristics, an environmental climate state impact analysis of the power distribution equipment is conducted. A preset neural network model is trained based on the environmental climate characteristics and historical environmental climate data to obtain an analysis model. Current environmental climate data, such as wind speed and rainfall, is acquired and input into the analysis model to conduct an environmental climate state impact analysis of the power distribution equipment, thereby obtaining the influence coefficient of environmental climate on the state of the power distribution equipment, i.e., obtaining the environmental climate state impact data. The environmental parameters of the environment where the power distribution equipment is located and the safe parameter range for normal operation of the equipment are determined. Environmental parameters include the temperature and humidity of the environment where the equipment is located. The safe parameter range is the range of operating parameters during normal operation of the equipment. The target difference between the pre-processed monitoring data and the endpoints of the safe parameter range is calculated. Specifically, the difference between the pre-processed monitoring data outside the safe parameter range and the corresponding endpoint value within the safe parameter range is calculated. For example, if the pre-processed monitoring data is 0.9 and the safe parameter range is (1.2, 1.5), the difference is 0.3. The loss coefficient is determined based on the environmental parameters and the actual usage time of the power distribution equipment. An initial loss coefficient is matched based on the actual usage time, and a proportional coefficient is matched based on the environmental parameters. The final loss coefficient is determined based on the proportional coefficient and the initial loss coefficient. Operational loss analysis is performed based on the target difference, loss coefficient, and environmental climate state impact data. The degree of operational loss of the power distribution equipment is determined based on the target difference, loss coefficient, impact coefficient, and their corresponding weighting coefficients, thus obtaining operational loss data.

[0060] S13: Analyze the correlation between the operating states of power distribution equipment based on the operating state matrix;

[0061] In the specific implementation of this invention, the step of analyzing the correlation between the operating states of power distribution equipment based on the operating state matrix includes: constructing an operating state matrix based on the monitoring index data of the power distribution equipment, and performing cluster analysis on the operating state matrix to obtain the cluster labels of the operating state matrix; constructing a corresponding topology diagram based on the connection relationship of the power distribution equipment; and determining the correlation between the operating states of the power distribution equipment based on the cluster labels of the operating state matrix and the topology diagram combined with a frequent itemset algorithm for mining association rules.

[0062] Specifically, an operational status matrix is ​​constructed based on monitoring index data of power distribution equipment. The monitoring index data consists of parameters representing the uses of different power distribution equipment. For example, monitoring index data for substations includes power supply capacity and total outage power supply capacity. Common and differential index data among the power distribution equipment are extracted from the monitoring index data. Specifically, common index data and differential index data are extracted, such as shared index data (e.g., interconnection rate with adjacent equipment) between generators and transformers, and differential index data (e.g., voltage deviation between generators and transformers). Pearson correlation coefficients are calculated for the common index data, and a Pearson coefficient of 100% is selected. Using key indicators, data reaching a preset threshold are used as critical indicators. A regression algorithm is employed to establish a fitting model between the differential indicator data and the critical indicators. The differential indicator data serves as input to the fitting model, outputting the fitted values ​​of the critical indicators. An operating state matrix for the power distribution equipment is constructed from the common indicator data and the fitted values ​​of the critical indicators. Cluster analysis is then performed on the operating state matrices of all power distribution equipment, yielding the number of clusters and cluster sets. The operating state matrices within each cluster set are labeled with preset values, obtaining cluster labels for the operating state matrices. For example, the operating state matrix of the first cluster set is labeled with the first value, and the operating state matrix of the second cluster set is labeled with the second value, and so on. A corresponding topology diagram is constructed based on the connection relationships between the power distribution equipment, obtaining the connection relationships between the equipment, such as their location and method. This topology diagram is then built within a preset topology tree structure diagram, incorporating the connection relationships between the power distribution equipment. The frequent itemset algorithm, which combines cluster labeling of the operating state matrix and topology graph to mine association rules, determines the association relationships between the operating states of power distribution equipment. This algorithm performs association analysis on the cluster labels of the operating state matrix, thereby mining association rules for the operating states of power distribution equipment. It utilizes an iterative, layer-by-layer search method to find relationships between itemsets in the database to form rules. The process consists of joins and pruning. In this algorithm, an itemset is defined as a set of items. A set containing x items is called an x-itemset. The frequency of an itemset is the number of transactions containing that itemset. If an itemset satisfies the minimum support condition, it is called a frequent itemset. After mining the association rules, the confidence level is used to determine the association relationships between the operating states of the equipment. For example, in the association rule X→Y, the higher the confidence level, the greater the probability that Y is included in the X transaction. The overall connection state of the power distribution system is determined through the topology graph, which represents the association relationships between the operating states of power distribution equipment. This provides a more intuitive understanding of the association relationships and a reliable basis for subsequent state analysis of power distribution.

[0063] S14: Based on the correlation relationship and combined with the preprocessed monitoring data, perform load status analysis of the power distribution system to obtain load status data;

[0064] In the specific implementation of this invention, the step of analyzing the load status of the power distribution system based on correlation and preprocessed monitoring data to obtain load status data includes: extracting load fluctuation features based on preprocessed monitoring data to obtain load fluctuation feature data; determining the initial load factor of the power distribution system based on the preprocessed monitoring data, and determining the mutual influence coefficient between power distribution equipment based on the correlation between the operating states of the power distribution equipment; determining the target load factor based on the mutual influence coefficient and the initial load factor, and performing load status analysis of the power distribution system based on the target load factor and load fluctuation feature data combined with the load status analysis model to obtain load status data.

[0065] Specifically, load fluctuation features are extracted based on preprocessed monitoring data. Historical load data of power distribution equipment is obtained, including historical load fluctuation values. The average value of the historical load data is calculated and used as a benchmark value. Load fluctuation values ​​are extracted from the preprocessed monitoring data. The load fluctuation amplitude at each time point is calculated based on the load fluctuation value and the benchmark value. The absolute value of the difference between the load fluctuation value and the benchmark value is divided by the benchmark value to obtain the load fluctuation amplitude. The time period and fluctuation frequency when the load fluctuation amplitude is greater than or equal to a preset amplitude threshold are determined. The fluctuation frequency is the number of times the load fluctuation amplitude is greater than or equal to the preset amplitude threshold per unit time. The load fluctuation amplitude, time period, and fluctuation frequency are used as feature data to obtain load fluctuation feature data. The initial load factor of the power distribution system is determined based on the preprocessed monitoring data. This involves analyzing the maximum and minimum line temperatures, line currents, and the difference between the maximum and minimum line currents within the monitoring period. The initial load factor is then calculated based on these values ​​and their corresponding proportionality coefficients. Furthermore, the mutual influence coefficients between the power distribution equipment are determined based on the correlation between their operating states. Based on this correlation, a load fluctuation simulation of the power distribution system is performed in simulation software to analyze the impact of the set load fluctuations on the operating states of the equipment, i.e., analyzing the changes in their operating parameters. The influence coefficients between each power distribution device are analyzed based on the simulation data. Finally, the overall mutual influence coefficients between the power distribution equipment in the power distribution system are determined based on these influence coefficients. The target load factor is determined based on the mutual influence coefficient and the initial load factor. The target load factor is determined according to the mutual influence coefficient, the initial load factor and their corresponding weight coefficients. The load state analysis of the power distribution system is then performed based on the target load factor and load fluctuation characteristic data combined with the load state analysis model. The load state analysis model is a convergent model obtained by inputting the sample dataset into a deep neural network for training. The target load factor and load fluctuation characteristic data are input into the load state analysis model, and the load state coefficient of the power distribution system is output, thus obtaining the load state data. This load state coefficient can reflect the overall load state of the power distribution system.

[0066] S15: Perform uncertainty analysis on the occurrence of anomalies in the power distribution system based on operating loss data and load status data to obtain uncertainty data on the occurrence of anomalies;

[0067] In the specific implementation of this invention, the step of performing uncertainty analysis on the occurrence of anomalies in the power distribution system based on operating loss data and load status data to obtain uncertainty data on the occurrence of anomalies includes: determining the state transition rate based on the number of transitions from normal operating state to abnormal state and the number of transitions from abnormal state to normal operating state in the power distribution system; determining the time-segmented parameter probability distribution based on the historical operating parameters of the power distribution network system, and constructing an uncertainty analysis model on the occurrence of anomalies based on a long short-term memory network based on the time-segmented parameter probability distribution and the state transition rate; and performing uncertainty analysis on the occurrence of anomalies in the power distribution system using the uncertainty analysis model on the occurrence of anomalies based on a long short-term memory network based on operating loss data and load status data to obtain uncertainty data on the occurrence of anomalies.

[0068] Specifically, the state transition rate is determined based on the number of transitions from normal operation to abnormal operation and the number of transitions from abnormal operation to normal operation of the power distribution system. The state transition rate includes a first state transition rate and a second state transition rate. The first state transition rate is obtained by dividing the number of transitions from normal operation to abnormal operation by the time the power distribution system is in the normal operation state, and the second state transition rate is obtained by dividing the number of transitions from abnormal operation to normal operation by the time the power distribution system is in the abnormal operation state. Based on the historical operating parameters of the distribution network system, the time-period parameter probability distribution is determined. Historical operating parameters include the voltage, phase difference, and power of each branch of the distribution system. The historical operating parameters are divided into preset time periods. The probability distribution of each value of the historical operating parameters in each time period is analyzed by the kernel density estimation function. The probability distributions of each time period are combined to generate the time-period parameter probability distribution. Based on the time-period parameter probability distribution and the state transition rate, an anomaly occurrence uncertainty analysis model based on a long short-term memory network is constructed. A training parameter set is generated according to the time-period parameter probability distribution and the state transition rate combined with historical loss and load data. The long short-term memory network is used as the initial anomaly occurrence uncertainty analysis model. The initial anomaly occurrence uncertainty analysis model is trained according to the training parameter set to obtain the final anomaly occurrence uncertainty analysis model. Uncertainty analysis of anomalies in power distribution systems is performed using an anomaly occurrence uncertainty analysis model based on long short-term memory (LSTM) networks, based on operating loss data and load status data. This involves inputting the operating loss data and load status data into the LSM network-based model to analyze the probability of anomalies occurring in the power distribution system, thus obtaining anomaly occurrence uncertainty data. Because of the losses and load variations in power distribution systems, the occurrence of anomalies carries a certain probability of uncertainty. Therefore, performing anomaly occurrence uncertainty analysis on power distribution systems can provide more comprehensive data for risk assessment, reflecting the impact of anomaly occurrence uncertainty on the safety of the power distribution system.

[0069] S16: Based on current phase vector data, operating loss data, load status data and anomaly occurrence uncertainty data, perform anomaly risk assessment of the power distribution system, obtain anomaly risk assessment results, and determine whether alarm information needs to be issued based on the anomaly risk assessment results.

[0070] In the specific implementation of this invention, the step of performing anomaly risk assessment of the power distribution system based on current phase vector data, operating loss data, load status data, and anomaly occurrence uncertainty data to obtain anomaly risk assessment results, and determining whether to issue alarm information based on the anomaly risk assessment results, includes: generating anomaly risk coefficient of the power distribution system using anomaly risk assessment model based on current phase vector data, operating loss data, load status data, and anomaly occurrence uncertainty data; if the anomaly risk coefficient of the power distribution system is greater than or equal to a preset anomaly risk threshold, then an alarm information is issued.

[0071] Specifically, based on current phase vector data, operating loss data, load status data, and anomaly occurrence uncertainty data, an anomaly risk assessment model is used to generate an anomaly risk coefficient for the power distribution system. The absolute value of the difference between current phase vector data at adjacent time points is calculated; this absolute value reflects the stability of the circuit frequency. The absolute value of the difference, operating loss data, load status data, and anomaly occurrence uncertainty data are input into the anomaly risk assessment model to obtain the anomaly risk coefficient of the power distribution system. This anomaly risk coefficient reflects the overall anomaly risk of the power distribution system. If the anomaly risk coefficient of the power distribution system is greater than or equal to a preset anomaly risk threshold, an alarm message is issued. If the anomaly risk coefficient of the power distribution system is less than the preset anomaly risk threshold, monitoring data continues to be collected for anomaly risk analysis of the power distribution system.

[0072] In this embodiment of the invention, operational loss analysis is performed based on preprocessed monitoring data combined with environmental parameters, improving the reliability of the operational loss analysis. Load status analysis of the power distribution system is performed based on correlations and preprocessed monitoring data, incorporating the correlations between the operating states of power distribution equipment, enabling a more specific and comprehensive analysis of the power distribution system's load status. Uncertainty analysis of power distribution system anomalies is performed based on operational loss data and load status data. Considering the inherent uncertainty in the occurrence of power distribution system anomalies, incorporating uncertainty analysis provides a basis for assessing anomaly risks. Anomaly risk assessment of the power distribution system is performed based on current phase vector data, operational loss data, load status data, and anomaly occurrence uncertainty data. By considering sufficiently comprehensive factors, the accuracy of anomaly risk assessment is effectively improved, avoiding false alarms and missed alarms.

[0073] Example 2

[0074] Please see Figure 2 , Figure 2 This is a flowchart illustrating an abnormal alarm method for a power distribution system according to another embodiment of the present invention, the method comprising:

[0075] S201: Preprocess the monitoring data of the power distribution equipment in the power distribution system to obtain the preprocessed monitoring data, and perform current phase vector analysis based on the preprocessed monitoring data to obtain current phase vector data.

[0076] S202: Based on the preprocessed monitoring data and environmental parameters, perform operational loss analysis to obtain operational loss data;

[0077] S203: Analyze the correlation between the operating states of power distribution equipment based on the operating state matrix;

[0078] S204: Extract load fluctuation features from preprocessed monitoring data to obtain load fluctuation feature data;

[0079] S205: Determine the initial load factor of the power distribution system based on the preprocessed monitoring data, and determine the mutual influence factor between power distribution equipment based on the correlation between the operating states of the power distribution equipment;

[0080] S206: Determine the target load factor based on the mutual influence coefficient and the initial load factor, and perform load state analysis of the power distribution system based on the target load factor and load fluctuation characteristic data combined with the load state analysis model to obtain load state data;

[0081] S207: Based on operating loss data and load status data, perform uncertainty analysis on the occurrence of anomalies in the power distribution system to obtain uncertainty data on the occurrence of anomalies;

[0082] S208: Based on current phase vector data, operating loss data, load status data and anomaly occurrence uncertainty data, perform anomaly risk assessment of the power distribution system, obtain anomaly risk assessment results, and determine whether alarm information needs to be issued based on the anomaly risk assessment results.

[0083] In this embodiment of the invention, operational loss analysis is performed based on preprocessed monitoring data combined with environmental parameters, improving the reliability of the operational loss analysis. Load status analysis of the power distribution system is performed based on correlations and preprocessed monitoring data, incorporating the correlations between the operating states of power distribution equipment, enabling a more specific and comprehensive analysis of the power distribution system's load status. Uncertainty analysis of power distribution system anomalies is performed based on operational loss data and load status data. Considering the inherent uncertainty in the occurrence of power distribution system anomalies, incorporating uncertainty analysis provides a basis for assessing anomaly risks. Anomaly risk assessment of the power distribution system is performed based on current phase vector data, operational loss data, load status data, and anomaly occurrence uncertainty data. By considering sufficiently comprehensive factors, the accuracy of anomaly risk assessment is effectively improved, avoiding false alarms and missed alarms.

[0084] Example 3

[0085] Please see Figure 3 , Figure 3 This is a schematic diagram of the structural composition of an abnormal alarm device for a power distribution system according to an embodiment of the present invention. The device includes:

[0086] Current phase vector module 31: used to preprocess the monitoring data of the power distribution equipment in the power distribution system, obtain the preprocessed monitoring data, and perform current phase vector analysis based on the preprocessed monitoring data to obtain current phase vector data;

[0087] Operational Loss Analysis Module 32: Used to perform operational loss analysis based on preprocessed monitoring data and environmental parameters to obtain operational loss data;

[0088] Module 33 for correlation analysis: used to analyze the correlation between the operating states of power distribution equipment based on the operating state matrix;

[0089] Load status analysis module 34: used to perform load status analysis of the power distribution system based on correlation and preprocessed monitoring data to obtain load status data;

[0090] Uncertainty Analysis Module 35: Used to perform uncertainty analysis on the occurrence of anomalies in the power distribution system based on operating loss data and load status data, and obtain uncertainty data on the occurrence of anomalies;

[0091] Anomaly alarm judgment module 36: It is used to perform anomaly risk assessment of the power distribution system based on current phase vector data, operating loss data, load status data and anomaly occurrence uncertainty data, obtain anomaly risk assessment results, and determine whether an alarm message needs to be issued based on the anomaly risk assessment results.

[0092] In the specific implementation of this invention, the specific implementation of the device item can be referred to the implementation of the method item above, and will not be repeated here.

[0093] In this embodiment of the invention, operational loss analysis is performed based on preprocessed monitoring data combined with environmental parameters, improving the reliability of the operational loss analysis. Load status analysis of the power distribution system is performed based on correlations and preprocessed monitoring data, incorporating the correlations between the operating states of power distribution equipment, enabling a more specific and comprehensive analysis of the power distribution system's load status. Uncertainty analysis of power distribution system anomalies is performed based on operational loss data and load status data. Considering the inherent uncertainty in the occurrence of power distribution system anomalies, incorporating uncertainty analysis provides a basis for assessing anomaly risks. Anomaly risk assessment of the power distribution system is performed based on current phase vector data, operational loss data, load status data, and anomaly occurrence uncertainty data. By considering sufficiently comprehensive factors, the accuracy of anomaly risk assessment is effectively improved, avoiding false alarms and missed alarms.

[0094] This invention provides a computer-readable storage medium storing a computer program. When executed by a processor, this program implements the abnormal alarm method for a power distribution system according to any of the above embodiments. The computer-readable storage medium includes, but is not limited to, any type of disk (including floppy disks, hard disks, optical disks, CD-ROMs, and magneto-optical disks), ROM (Read-Only Memory), RAM (Random Access Memory), EPROM (Erasable Programmable Read-Only Memory), EEPROM (Electrically Erasable Programmable Read-Only Memory), flash memory, magnetic cards, or optical cards. In other words, the storage device includes any medium that stores or transmits information in a readable form by a device (e.g., a computer, a mobile phone), and can be a read-only memory, a disk, or an optical disk, etc.

[0095] Example 4

[0096] Please see Figure 4 , Figure 4 This is a schematic diagram of the structural composition of the electronic device in an embodiment of the present invention.

[0097] This invention also provides an electronic device, such as... Figure 4As shown, the electronic device includes a memory 41, a processor 43, and a computer program 42 stored in the memory 41 and executable on the processor 43. Those skilled in the art will understand that... Figure 4 The illustrated electronic device does not constitute a limitation on all devices and may include more or fewer components than illustrated, or combine certain components. Memory 41 can be used to store computer program 42 and various functional modules. Processor 43 runs the computer program 42 stored in memory 41, thereby performing various functional applications and data processing of the device. Memory can be internal memory or external memory, or both. Internal memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory, or random access memory. External memory may include hard disks, floppy disks, ZIP disks, USB flash drives, magnetic tapes, etc. Processor 43 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor, a single-chip microcomputer, or a processor 43, or any conventional processor, etc. The processors and memories disclosed in this invention include, but are not limited to, these types of processors and memories. The processors and memories disclosed in this invention are merely examples and not intended to be limiting.

[0098] As one embodiment, the electronic device includes: one or more processors 43, a memory 41, and one or more computer programs 42, wherein the one or more computer programs 42 are stored in the memory 41 and configured to be executed by the one or more processors 43, and the one or more computer programs 42 are configured to execute the abnormal alarm method of the power distribution system in any of the above embodiments. For the specific implementation process, please refer to the above embodiments, which will not be repeated here.

[0099] In this embodiment of the invention, operational loss analysis is performed based on preprocessed monitoring data combined with environmental parameters, improving the reliability of the operational loss analysis. Load status analysis of the power distribution system is performed based on correlations and preprocessed monitoring data, incorporating the correlations between the operating states of power distribution equipment, enabling a more specific and comprehensive analysis of the power distribution system's load status. Uncertainty analysis of power distribution system anomalies is performed based on operational loss data and load status data. Considering the inherent uncertainty in the occurrence of power distribution system anomalies, incorporating uncertainty analysis provides a basis for assessing anomaly risks. Anomaly risk assessment of the power distribution system is performed based on current phase vector data, operational loss data, load status data, and anomaly occurrence uncertainty data. By considering sufficiently comprehensive factors, the accuracy of anomaly risk assessment is effectively improved, avoiding false alarms and missed alarms.

[0100] Furthermore, the above provides a detailed description of an abnormal alarm method and related device for a power distribution system provided by the embodiments of the present invention. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. An abnormality alarm method of a power distribution system, characterized by, The method comprises: monitoring data of power distribution equipment of a power distribution system is preprocessed to obtain preprocessed monitoring data, and current phase vector analysis is performed based on the preprocessed monitoring data to obtain current phase vector data; based on the preprocessed monitoring data and environmental parameters, operation loss analysis is performed to obtain operation loss data; based on the correlation between the running states of the power distribution equipment, based on the correlation and the preprocessed monitoring data, load state analysis of the power distribution system is performed to obtain load state data; based on the operation loss data and the load state data, abnormal occurrence uncertainty analysis of the power distribution system is performed to obtain abnormal occurrence uncertainty data; based on the current phase vector data, the operation loss data, the load state data and the abnormal occurrence uncertainty data, abnormal risk assessment of the power distribution system is performed to obtain abnormal risk assessment results, and whether an alarm message needs to be sent is determined based on the abnormal risk assessment results; wherein the monitoring data of the power distribution equipment of the power distribution system is preprocessed to obtain preprocessed monitoring data, and current phase vector analysis is performed based on the preprocessed monitoring data to obtain current phase vector data, which comprises: data cleaning processing is performed on the monitoring data to obtain data cleaning processed monitoring data; data integration processing is performed on the data cleaning processed monitoring data to obtain preprocessed monitoring data; current vector data and current value are extracted from the preprocessed monitoring data, and the current vector data is amplitude corrected based on an amplitude correction coefficient table to obtain corrected current vector data; a current flow direction matrix is constructed based on the corrected current vector data, and current phase vector analysis is performed based on the current flow direction matrix, the corrected current vector data and the current value to obtain current phase vector data; the preprocessed monitoring data is combined with environmental parameters to perform operation loss analysis and obtain operation loss data, which comprises: environmental and climate characteristic quantities of the power distribution equipment are determined, and environmental and climate state influence analysis of the power distribution equipment is performed based on the environmental and climate characteristic quantities to obtain environmental and climate state influence data; environmental parameters of the environment in which the power distribution equipment is located and a safety parameter range of normal operation of the power distribution equipment are determined, and a target difference value between the preprocessed monitoring data and the end points of the safety parameter range is calculated; a loss coefficient is determined based on the environmental parameters and the actual use duration of the power distribution equipment; based on the target difference value, the loss coefficient and the environmental and climate state influence data, operation loss analysis is performed to obtain operation loss data; based on the correlation and the preprocessed monitoring data, load state analysis of the power distribution system is performed to obtain load state data, which comprises: load fluctuation feature data is obtained based on load fluctuation feature extraction of the preprocessed monitoring data; an initial load coefficient of the power distribution system is determined based on the preprocessed monitoring data, and a mutual influence coefficient between the power distribution equipment is determined based on the correlation between the running states of the power distribution equipment; a target load coefficient is determined based on the mutual influence coefficient and the initial load coefficient, and load state analysis of the power distribution system is performed based on the target load coefficient and the load fluctuation feature data in combination with a load state analysis model to obtain load state data.

2. The abnormality warning method of a power distribution system according to claim 1, characterized by, The correlation between the running states of the power distribution equipment is analyzed based on the running state matrix, including: An operation state matrix is constructed based on the monitoring index data of the power distribution equipment, and cluster analysis is performed on the operation state matrix to obtain cluster labels of the operation state matrix; A corresponding topological structure diagram is constructed based on the connection relationship of the power distribution equipment; The correlation between the running states of the power distribution equipment is determined based on the cluster labels of the operation state matrix and the topological structure diagram in combination with a frequent item set algorithm for mining association rules.

3. The abnormality warning method of a power distribution system according to Claim 1, characterized by, The abnormal occurrence uncertainty analysis of the power distribution system is performed based on the operation loss data and the load state data to obtain abnormal occurrence uncertainty data, including: The state transition rate is determined based on the number of times of transition from a normal running state to an abnormal state and the number of times of transition from an abnormal state to a normal running state of the power distribution system; The time period parameter probability distribution is determined based on the historical operation parameters of the power distribution network system, and an abnormal occurrence uncertainty analysis model based on a long short-term memory network is constructed based on the time period parameter probability distribution and the state transition rate; The abnormal occurrence uncertainty analysis of the power distribution system is performed based on the operation loss data and the load state data by using the abnormal occurrence uncertainty analysis model based on the long short-term memory network to obtain abnormal occurrence uncertainty data.

4. The abnormality warning method of a power distribution system according to Claim 1, characterized by, The abnormal risk assessment of the power distribution system is performed based on the current phase vector data, the operation loss data, the load state data, and the abnormal occurrence uncertainty data to obtain abnormal risk assessment results, and it is determined whether an alarm information needs to be sent based on the abnormal risk assessment results, including: The abnormal risk coefficient of the power distribution system is generated by using an abnormal risk assessment model based on the current phase vector data, the operation loss data, the load state data, and the abnormal occurrence uncertainty data; If the abnormal risk coefficient of the power distribution system is greater than or equal to a preset abnormal risk threshold, an alarm information is sent.

5. An abnormal alarm device for a power distribution system, characterized in that, The device includes: A current phase vector module: configured to preprocess the monitoring data of the power distribution equipment of the power distribution system to obtain preprocessed monitoring data, and perform current phase vector analysis based on the preprocessed monitoring data to obtain current phase vector data; An operation loss analysis module: configured to perform operation loss analysis based on the preprocessed monitoring data in combination with environmental parameters to obtain operation loss data; A correlation analysis module: configured to analyze the correlation between the running states of the power distribution equipment based on the running state matrix; A load state analysis module: configured to perform load state analysis of the power distribution system based on the correlation in combination with the preprocessed monitoring data to obtain load state data; An uncertainty analysis module: configured to perform abnormal occurrence uncertainty analysis of the power distribution system based on the operation loss data and the load state data to obtain abnormal occurrence uncertainty data; An abnormal alarm judgment module: configured to perform abnormal risk assessment of the power distribution system based on the current phase vector data, the operation loss data, the load state data, and the abnormal occurrence uncertainty data to obtain abnormal risk assessment results, and determine whether an alarm information needs to be sent based on the abnormal risk assessment results. The monitoring data of the power distribution equipment of the power distribution system is preprocessed to obtain preprocessed monitoring data, and current phase vector analysis is performed based on the preprocessed monitoring data to obtain current phase vector data, including: performing data cleaning processing on the monitoring data to obtain data cleaning processed monitoring data; performing data integration processing on the data cleaning processed monitoring data to obtain preprocessed monitoring data; extracting current vector data and current value in the preprocessed monitoring data, and correcting the amplitude of the current vector data based on an amplitude correction coefficient table to obtain corrected current vector data; constructing a current flow direction matrix based on the corrected current vector data, and performing current phase vector analysis based on the current flow direction matrix, the corrected current vector data and the current value to obtain current phase vector data; The running loss analysis is performed based on the preprocessed monitoring data combined with environmental parameters to obtain running loss data, including: determining the environmental climate characteristic quantity of the power distribution equipment, and performing environmental climate state influence analysis on the power distribution equipment based on the environmental climate characteristic quantity to obtain environmental climate state influence data; determining the environmental parameters of the environment where the power distribution equipment is located and the safety parameter range of the normal operation of the power distribution equipment, calculating the target difference value of the preprocessed monitoring data and the end point of the safety parameter range; determining the loss coefficient based on the environmental parameters and the actual use time length of the power distribution equipment; performing running loss analysis based on the target difference value, the loss coefficient and the environmental climate state influence data to obtain running loss data; The load state analysis of the power distribution system is performed based on the correlation combined with the preprocessed monitoring data to obtain load state data, including: performing load fluctuation feature extraction based on the preprocessed monitoring data to obtain load fluctuation feature data; determining the initial load coefficient of the power distribution system based on the preprocessed monitoring data, and determining the mutual influence coefficient between the power distribution equipment based on the correlation between the operating states of the power distribution equipment; determining the target load coefficient based on the mutual influence coefficient and the initial load coefficient, and performing load state analysis of the power distribution system based on the target load coefficient and the load fluctuation feature data combined with a load state analysis model to obtain load state data. 6.An electronic device comprising a processor and a memory, wherein, The memory is used to store instructions, and the processor is used to call the instructions in the memory, so that the electronic device executes the abnormal alarm method of the power distribution system as claimed in any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that, The computer readable storage medium stores computer instructions, when the computer instructions run on the electronic device, make the electronic device execute the abnormal alarm method of the power distribution system as claimed in any one of claims 1 to 4.

Images