Power distribution network fault early warning method, device, equipment, readable storage medium and program product

Abstract

The invention relates to a power distribution network fault early warning method, device and equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: acquiring and crossly transmitting real-time information of a power distribution network based on two sets of equipment to obtain two groups of real-time data corresponding to the two sets of equipment; respectively extracting fault features of the power distribution network from the two groups of real-time data to obtain a first fault feature and a second fault feature; inputting the first fault feature into a preset analysis model; the analysis model is used for analyzing the first fault feature and outputting corresponding first fault information; performing analogue simulation analysis on the second fault feature to obtain second fault information; determining the consistency of the first fault information and the second fault information; and if the consistency reaches a preset standard, determining fault information of the power distribution network and outputting an early warning signal of a corresponding level. According to the method, automation of power distribution network troubleshooting processing can be realized, and the accuracy of power distribution network fault early warning is improved.

Description

Technical Field

[0001] The present application relates to the field of electric power technology, and in particular to a distribution network fault early warning method, device, equipment, computer-readable storage medium and computer program product. Background Art

[0002] A distribution network refers to an electric power grid that receives electrical energy from a transmission network or a regional power plant and distributes it locally or step by step according to voltage to various types of users through distribution facilities. It consists of overhead lines, cables, poles, towers, distribution transformers, disconnectors, reactive power compensators and some ancillary facilities.

[0003] For distribution network faults, regular manual maintenance is generally used to troubleshoot and handle them, which is inefficient. At the same time, due to the diversity of distribution network faults, this requires high professional skills of operators, and the results of distribution network fault troubleshooting vary depending on the operator. Therefore, how to improve the efficiency and accuracy of distribution network fault troubleshooting is a problem that needs to be solved. Summary of the invention

[0004] Based on this, it is necessary to provide a distribution network fault warning method, device, computer equipment, computer readable storage medium and computer program product to address the above technical problems, so as to improve the efficiency and accuracy of distribution network fault troubleshooting.

[0005] In a first aspect, the present application provides a distribution network fault early warning method, comprising:

[0006] Based on the collection and cross-transmission of real-time information of the distribution network by two sets of equipment, two sets of real-time data corresponding to the two sets of equipment are obtained;

[0007] Extracting fault characteristics of the distribution network from the two sets of real-time data respectively to obtain a first fault characteristic and a second fault characteristic; the fault characteristic includes at least one of the following: a power parameter characteristic, a topological structure characteristic, a time series data characteristic, a fault signal characteristic, and a data statistical characteristic;

[0008] Inputting the first fault feature into a preset analysis model; the analysis model is used to analyze the first fault feature and output corresponding first fault information; the analysis model is obtained by training based on historical data of the distribution network;

[0009] Based on the simulation analysis of the second fault feature, second fault information is obtained;

[0010] Determining consistency between the first fault information and the second fault information;

[0011] When the consistency reaches the preset standard, the fault information of the distribution network is determined and a warning signal of the corresponding level is output.

[0012] In one embodiment, obtaining the second fault information based on the simulation analysis of the second fault feature includes:

[0013] According to the second fault feature, determine the corresponding feature point value and the feature overall mean value;

[0014] Determine the degree of deviation of the feature point value compared to the overall mean of the feature;

[0015] According to the degree of deviation, second fault information is determined.

[0016] In one embodiment, determining the deviation of the feature point value from the feature overall mean value includes:

[0017] Determine the correlation coefficient between the second fault feature and the historical data according to the feature point value;

[0018] According to the characteristic point value, historical data and correlation coefficient, the degree of deviation of the characteristic point value in the second fault feature compared to the overall mean value of the feature is determined.

[0019] In one embodiment, the power parameter features are extracted in the following manner:

[0020] Obtaining the original power signal of the distribution network;

[0021] Determine at least one local extreme point of the original signal;

[0022] Determine the average and envelope values ​​of adjacent local extreme points;

[0023] The power parameter characteristics are determined based on the average value and the envelope value.

[0024] In one embodiment, the method further comprises:

[0025] According to the first fault feature, a preliminary analysis model is determined from a plurality of preset basic analysis models;

[0026] Determine a fault type corresponding to the first fault feature;

[0027] Based on the fault type and the corresponding simulation, the preliminary analysis model is optimized to obtain the analysis model.

[0028] In one embodiment, the real-time data includes: current in the circuit, voltage, device and line temperature, ambient humidity, power usage, circuit switch status, transformer operating status;

[0029] The method also includes: preprocessing the real-time data to improve data quality; the preprocessing includes at least one of the following: cleaning, removing abnormal data, and filling missing values.

[0030] In a second aspect, the present application also provides a distribution network fault early warning device, the device comprising:

[0031] A feature acquisition module is used to obtain two sets of real-time data corresponding to the two sets of equipment based on the collection and cross-transmission of real-time information of the distribution network; extract the fault characteristics of the distribution network from the two sets of real-time data respectively to obtain a first fault characteristic and a second fault characteristic; the fault characteristic includes at least one of the following: power parameter characteristics, topological structure characteristics, time series data characteristics, fault signal characteristics, and data statistical characteristics;

[0032] A first analysis module is used to input the first fault feature into a preset analysis model; the analysis model is used to analyze the first fault feature and output corresponding first fault information; the analysis model is obtained by training based on historical data of the distribution network;

[0033] A second analysis module, configured to obtain second fault information based on a simulation analysis of a second fault feature;

[0034] The fault warning module is used to determine the consistency between the first fault information and the second fault information; when the consistency reaches a preset standard, the fault information of the distribution network is determined and a warning signal of the corresponding level is output.

[0035] In a third aspect, the present application further provides a distribution network fault warning device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method in the first aspect when executing the computer program.

[0036] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the method in the first aspect above.

[0037] In a fifth aspect, the present application further provides a computer program product, comprising a computer program, which implements the steps of the method in the above-mentioned first aspect when executed by a processor.

[0038] The above-mentioned distribution network fault early warning method, device, equipment, computer-readable storage medium and computer program product collect and cross-transmit the real-time information of the distribution network through two sets of equipment to obtain two sets of real-time data, so as to perform distribution network fault early warning through two sets of real-time data, which can effectively prevent data anomalies from affecting the accuracy of distribution network fault early warning. By extracting fault characteristics from real-time data, a first fault characteristic and a second fault characteristic are obtained. By using an analysis model to analyze the first fault characteristic, a first fault information is obtained; by simulating and analyzing the second fault characteristic, a second fault information is obtained. Finally, the fault information of the distribution network is determined according to the consistency between the first fault information and the second fault information, and a warning signal of the corresponding level is output. That is, the real-time data of the distribution network is analyzed from two aspects of the analysis model and the simulation, and according to the analysis results, it is comprehensively determined whether there is a fault in the distribution network, and the specific fault situation corresponding to the fault when there is a fault, and the corresponding distribution network is given at the same time. This realizes the automation of distribution network fault troubleshooting and makes the distribution network fault early warning more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0040] Figure 1 A schematic diagram of a flow chart of a distribution network fault early warning method in one embodiment;

[0041] Figure 2 It is another schematic diagram of a flow chart of a method for early warning of a power distribution network failure in an embodiment;

[0042] Figure 3 It is another flow chart of a distribution network fault early warning method in one embodiment;

[0043] Figure 4 A schematic diagram of the structure of a power distribution network fault early warning system in one embodiment;

[0044] Figure 5 It is a structural block diagram of a distribution network fault early warning device in one embodiment;

[0045] Figure 6 The figure is a diagram of the internal structure of a distribution network fault warning device in one embodiment. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. 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.

[0047] In one embodiment, Figure 1 As shown, a distribution network fault warning method is provided. This embodiment uses the method applied to the distribution network warning system as an example for explanation. It can be understood that the method can also be applied to other systems, such as the distribution network management system, the management system of the entire power grid, etc. The method can be applied to a terminal or a server, in which the above system is pre-installed. The server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. It can also be applied to a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. In this embodiment, the method includes steps S101 to S105:

[0048] Step S101: Based on the collection and cross-transmission of real-time information of the distribution network by two sets of equipment, two groups of real-time data corresponding to the two sets of equipment are obtained.

[0049] The equipment may include but is not limited to sensors and smart meters. The two sets of equipment may operate independently of each other, and the two sets of equipment may operate alternately to transmit data. Real-time data refers to data used to reflect the real-time operation of the distribution network.

[0050] Exemplarily, by utilizing two sets of equipment to collect and cross-transmit real-time information of the power distribution network, two sets of real-time data can be obtained, thereby facilitating subsequent use of the two sets of real-time data.

[0051] Step S102: extracting fault characteristics of the distribution network from the two sets of real-time data respectively to obtain a first fault characteristic and a second fault characteristic; the fault characteristic includes at least one of the following: power parameter characteristics, topology structure characteristics, time series data characteristics, fault signal characteristics, and data statistical characteristics.

[0052] Among them, power parameter features refer to the features of power-related parameters, such as voltage, current, etc. Topological structure features refer to the features related to the topology of the distribution network structure, such as the components of the distribution network, such as overhead lines, cables, towers, distribution transformers, disconnectors, etc., as well as the relationship between the components. Time series data features refer to the features of time series data related to the distribution network. Fault signal features refer to the features of the distribution network related to fault signals. Data statistical features refer to the features of the relevant statistical data of the distribution network.

[0053] For example, since two sets of real-time data are obtained, the fault features can be extracted from the two sets of real-time data to obtain the first fault feature and the second fault feature. For example, the first fault feature can be obtained by integrating the fault features of the above-mentioned specific types, and similarly, the second fault feature can also be obtained in this way.

[0054] Step S103: input the first fault feature into a preset analysis model; the analysis model is used to analyze the first fault feature and output corresponding first fault information; the analysis model is obtained by training based on historical data of the distribution network.

[0055] The preset analysis model is an artificial intelligence model pre-set for distribution network fault analysis. The historical data of the distribution network is the relevant data of the distribution network in the past time range. The historical data is relative to the aforementioned real-time data. In essence, both are relevant data reflecting the operation of the distribution network. The core difference lies in the time difference.

[0056] Exemplarily, the historical data of the distribution network can be used in advance to train an existing or customized artificial intelligence model to obtain an analysis model. Of course, the analysis model can also be optimized. The optimization can be based on all types of fault features, or it can be based on a certain type or several types of fault features. In addition, different types of analysis models can be trained in a targeted manner according to different types of fault features, and then the corresponding type of analysis model can be selected as needed. Then, the analysis model can be used to analyze the first fault feature to obtain the first fault information. The first fault information can be a characterization of the existence of a fault in the distribution network and specific fault information, or it can also be a characterization of the absence of a fault in the distribution network.

[0057] Step S104: obtaining second fault information based on simulation analysis of the second fault feature.

[0058] For example, the second fault feature obtained by the above steps can be simulated and analyzed to analyze the relevant information about the distribution network fault contained in the second fault feature, thereby obtaining the second fault information. The second fault information can be a characterization of the existence of a fault in the distribution network and specific fault information, or it can also be a characterization of the absence of a fault in the distribution network.

[0059] Step S105: Determine the consistency between the first fault information and the second fault information; when the consistency reaches a preset standard, determine the fault information of the distribution network and output a warning signal of a corresponding level.

[0060] Exemplarily, after obtaining the first fault information and the second fault information, the two can be compared to determine the consistency between the two. If the consistency reaches the preset standard, the fault information of the distribution network can be determined and the warning signal of the corresponding level can be output. For example, the consistency can include three levels: low, medium, and high, and the corresponding preset standard can be a requirement to reach medium or high. Further, for example, if the first fault information indicates that there is no fault in the distribution network and the second fault information indicates that there is a fault in the distribution network, it can be considered that the consistency is low and the consistency does not reach the preset standard. For another example, if the first fault information and the second fault information both indicate that there is a fault in the distribution network and the specific fault is in the distribution transformer, the consistency can be high and reach the preset standard. For another example, if the first fault information and the second fault information both indicate that there is a fault in the distribution network but the specific faults corresponding to each are different, the consistency can be medium and reach the preset standard. Of course, the consistency can include two levels of inconsistency and consistency, and the preset standard can be a requirement for consistency.

[0061] In this embodiment, two sets of equipment are used to collect and cross-transmit the real-time information of the distribution network to obtain two sets of real-time data, so that the distribution network fault warning is performed through the two sets of real-time data, which can effectively prevent data anomalies from affecting the accuracy of the distribution network fault warning. By extracting fault features from real-time data, the first fault feature and the second fault feature are obtained. By using the analysis model to analyze the first fault feature, the first fault information is obtained; by simulating and analyzing the second fault feature, the second fault information is obtained. Finally, the fault information of the distribution network is determined according to the consistency between the first fault information and the second fault information, and the corresponding level of warning signal is output. That is, the real-time data of the distribution network is analyzed from the two aspects of the analysis model and the simulation, and according to the analysis results, it is comprehensively determined whether there is a fault in the distribution network, and the specific fault situation corresponding to the fault when there is a fault, and the corresponding distribution network is given at the same time. This realizes the automation of the distribution network fault troubleshooting and processing, making the distribution network fault warning more accurate.

[0062] In one embodiment, if Figure 2 As shown, the "obtaining second fault information based on the simulation analysis of the second fault feature" in the above embodiment may include steps S201 to S203:

[0063] Step S201: According to the second fault feature, determine the corresponding feature point value and the feature overall mean value.

[0064] Among them, the fault feature may include multiple feature points, and corresponding feature point values, that is, the corresponding values ​​of specific points in the second fault feature; the overall feature mean, that is, the mean of the corresponding values ​​of all feature points.

[0065] Exemplarily, the second fault feature may include different types of fault features, and thus corresponding feature point values ​​and feature overall mean values ​​may be determined for different types of fault features, respectively. For example, for a temperature type fault feature, the corresponding feature point value may be a feature value corresponding to a single point temperature, and the feature overall mean value may be a feature mean value corresponding to the overall temperature.

[0066] Step S202: Determine the degree of deviation of the feature point value compared to the overall feature mean.

[0067] Exemplarily, the degree of deviation may be determined by using a range method, a standard deviation method, etc., which is not limited here.

[0068] Step S203: Determine the second fault information according to the degree of deviation.

[0069] Exemplarily, a deviation threshold may be set. If the obtained deviation value does not exceed the deviation threshold, it may be considered that there is no fault. Otherwise, it may be considered that there is a fault.

[0070] In this embodiment, the second fault information is determined by determining the characteristic point value and the overall mean value of the characteristic corresponding to the second fault characteristic, and based on the degree of deviation of the characteristic point value from the overall mean value of the characteristic. This enables the obtained second fault information to more accurately determine whether there is a fault in the distribution network, and the corresponding fault information when there is a fault.

[0071] In one embodiment, "determining the degree of deviation of the characteristic point value compared to the overall mean of the characteristic" in the aforementioned embodiment may include: determining the correlation coefficient between the second fault characteristic and the historical data based on the characteristic point value; determining the degree of deviation of the characteristic point value in the second fault characteristic compared to the overall mean of the characteristic based on the characteristic point value, the historical data and the correlation coefficient.

[0072] The correlation coefficient is a coefficient indicating the correlation degree between the second fault feature and the historical data.

[0073] Exemplarily, the correlation coefficient can be calculated by the following formula:

[0074] ,in,

[0075] x represents the second fault characteristic;

[0076] represents the second fault feature mean;

[0077] r represents the correlation coefficient;

[0078] y represents a data point in historical data;

[0079] Represents the mean value of a data point in the historical data of the distribution network.

[0080] Exemplarily, the degree of deviation can be calculated by the following formula:

[0081] ,in,

[0082] Z represents the degree of deviation;

[0083] Represents the standard deviation of x.

[0084] In this embodiment, by calculating the correlation coefficient and determining the degree of deviation based on the characteristic point value, historical data and the correlation coefficient, a more accurate degree of deviation can be obtained, which helps to obtain more accurate second fault information, and further helps to achieve more accurate distribution network fault warning.

[0085] In one embodiment, the power parameter characteristics in the above embodiment can be as follows: Figure 3 The method shown in the figure specifically includes steps S301 to S304:

[0086] Step S301: Acquire the original power signal of the distribution network.

[0087] The original power signal of the distribution network is an unprocessed signal related to power of the distribution network, for example, a voltage signal.

[0088] Step S302: Determine at least one local extreme point of the original signal.

[0089] The local extreme value point may be a point corresponding to a local maximum value and / or a local minimum value in the original signal.

[0090] Step S303: Determine the average value and envelope value of adjacent local extreme value points.

[0091] Step S304: Determine the power parameter characteristics according to the average value and the envelope value.

[0092] A possible specific method for determining the characteristics of power parameters is given below:

[0093] Determine all local extreme points n of the original signal x(t) i (i=1,2,…,M); Calculate the average value m of all adjacent local extreme points i and envelope value a i , the specific calculation formula is as follows:

[0094] , , where i=1,2,…,M-1.

[0095] According to the average value m i and envelope value a i , the power parameter characteristics can be obtained. For example, the average value m i and envelope value a i By comparing with previous data, we can find out whether the data is faulty.

[0096] In this embodiment, by decomposing the original power signal to obtain the average value and envelope value, more accurate power parameter characteristics can be obtained, that is, the power parameter characteristics can more accurately reflect the actual operation of the distribution network and be more representative, which helps to achieve more accurate distribution network fault warning.

[0097] In one of the embodiments, the distribution network fault warning method in the aforementioned embodiment may also include: determining a preliminary analysis model from a plurality of preset basic analysis models according to the first fault feature; determining the fault type corresponding to the first fault feature; and optimizing the preliminary analysis model based on the fault type and the corresponding simulation to obtain an analysis model.

[0098] Exemplarily, multiple basic analysis models can be pre-set, and the basic analysis models can be existing or customized artificial intelligence models. Then, based on the first fault feature, a suitable basic analysis model is selected from the multiple basic analysis models as the preliminary analysis model. The selection here can be manual selection or automatic selection by setting selection conditions. For example, it can be selected based on the amount of computing required for the first fault feature and the specific computing task type.

[0099] Exemplarily, after determining the preliminary analysis model, the fault type corresponding to the first fault feature is determined, that is, the fault type that the model needs to analyze can be clarified, so the preliminary analysis model can be optimized in a targeted manner. For example, the data corresponding to the fault type can be determined from the historical data of the distribution network and the data can be used to perform simulation analysis, and then the results of the simulation analysis can be used to optimize the preliminary analysis model to obtain an analysis model.

[0100] In this embodiment, the preliminary analysis model is determined and optimized from the perspective of the fault type to obtain an analysis model, which makes the analysis model more accurate in analyzing the first fault feature, that is, more accurate first fault information is obtained.

[0101] In one of the embodiments, the real-time data in the aforementioned embodiment may include: current in the circuit, voltage, device and line temperature, ambient humidity, usage of electric energy, circuit switch status, transformer operation status;

[0102] The distribution network fault warning method in the aforementioned embodiment may further include: preprocessing the real-time data to improve data quality; the preprocessing includes at least one of the following: cleaning, removing abnormal data, and filling missing values.

[0103] In this embodiment, since the real-time data may include the current, voltage, etc. in the circuit, this helps the real-time data to more comprehensively and accurately reflect the actual situation of the distribution network; since the real-time data is pre-processed by cleaning, removing abnormal data, filling missing values, etc., this can improve the data quality, facilitate the processing of the real-time data in subsequent steps, and also help the real-time data to more comprehensively and accurately reflect the actual situation of the distribution network.

[0104] In one embodiment, Figure 4 As shown, a distribution network fault warning system is provided, which includes a simulation data acquisition module, a data processing and analysis module, a fault feature extraction module, a warning model construction module, a warning module and a data recording module. The details are as follows:

[0105] About the simulation data acquisition module:

[0106] The module collects the voltage, current, frequency and power factor data of each node of the distribution network in real time through sensors and intelligent instrument equipment. The measurement equipment used in the simulation data acquisition module is set up with two sets, and the data is transmitted through cross operation; the simulation data acquisition module includes a sensor module, an intelligent instrument module and a data acquisition device module; the sensor module is used to monitor the current, voltage, equipment and line temperature and environmental humidity in the circuit; the intelligent instrument module is used to monitor the use of electric energy, the switch state of the circuit and the operating state of the transformer; the data acquisition device module includes a data collector and a data transmitter. The data collector is responsible for collecting the data obtained by the sensor module and the intelligent instrument module, and the data transmitter is responsible for transmitting the collected data to the data processing and analysis module. The data collected by each module in the simulation data acquisition module can better analyze the relevant problems in the distribution network and improve the accuracy of the results. Specifically, the simulation data acquisition module collects data through various devices, and at the same time, each type of equipment is set up with two sets. The two sets of equipment alternately operate to transmit data. By comparing the data measured by the two sets of equipment, it can be determined whether the equipment is faulty, thereby preventing unnecessary emergency measures caused by equipment failure.

[0107] About data processing and analysis module:

[0108] This module performs preprocessing operations such as cleaning, denoising and correction on the collected data to improve the data quality. Exemplarily, the data processing and analysis module may include a data cleaning module, a feature extraction module, a data analysis module, a fault diagnosis module, a prediction module and a visualization module. Among them, the data cleaning module is responsible for cleaning the collected raw data, removing abnormal data and filling missing values ​​to ensure the quality and accuracy of the data; the feature extraction module performs feature extraction based on the preprocessing of the data by the data cleaning module, extracts feature variables reflecting the system state and characteristics from the data, so that the extracted data is representative, and the results of subsequent secondary processing are more typical, thereby improving the accuracy of the system's fault prediction; the data analysis module analyzes the data after feature extraction based on the data extracted by the feature extraction data module, and identifies potential fault modes, laws and abnormal conditions; the fault diagnosis module performs fault diagnosis and positioning on the data analysis results based on the data analysis by the data analysis module, and identifies the fault types and locations in the distribution network; the prediction module performs fault prediction and trend prediction based on historical data and analysis results; the visualization module is responsible for presenting the results of data processing and analysis in a visual way, thereby helping users understand the data and discover potential laws. The data processing and analysis module is used to pre-process the data to improve the quality and effectiveness of the data and reduce the interference of useless data. At the same time, the processed data is used for initial fault diagnosis and visual output.

[0109] About the fault feature extraction module:

[0110] Based on the stored data, this module uses data mining and machine learning techniques to extract fault features, making the extracted fault features representative. The fault feature extraction module algorithm is:

[0111] Assume a set of monitoring data x1, x1...xn, and calculate the mean μ and standard deviation σ as features.

[0112] The calculation formula corresponding to the mean is: ;

[0113] The calculation formula for the standard deviation is: ;

[0114] The fault feature extraction module includes a power parameter extraction module, a topology feature extraction module, a timing feature extraction module, a fault signal feature extraction module, a statistical feature extraction module and a multimodal feature fusion module; among them, the power parameter extraction module is responsible for extracting features related to power parameters. A possible method for extracting features related to power parameters is given below:

[0115] For any signal x(t), the decomposition process is as follows: determine all local extreme points n of the original signal x(t)i (i=1,2,…,M); Calculate the average value m of all adjacent local extreme points i and envelope value a i ;

[0116] ,i=1,2,…,M-1 (1)

[0117] , i=1,2,…,M-1 (2)

[0118] The calculated average value m i and envelope value a i By comparing with previous data, we can find out whether the data is faulty.

[0119] The topological structure feature extraction module is responsible for extracting the topological structure features of the distribution network; the time series feature extraction module is responsible for extracting the features of the time series data; the fault signal feature extraction module is responsible for extracting the features related to the fault signal; the statistical feature extraction module is responsible for extracting the statistical features of the data; and the multimodal feature fusion module is responsible for fusing different types of features. The fault feature extraction module further extracts various types of data, making the data more helpful to maintenance personnel and facilitating subsequent maintenance.

[0120] About the early warning model building module:

[0121] This module is used to build an early warning model, compare and analyze the extracted fault features with historical data, and identify potential fault modes. The early warning model building module includes a model selection module, a model training module, a model evaluation module, a model optimization module, and a real-time monitoring module.

[0122] Among them, the model selection module selects a suitable prediction model based on the data extracted by the fault feature extraction module. Through the prediction model, the development trend of the distribution network under the data can be more intuitively felt; the model training module is responsible for training the selected model using historical data so that the model can accurately predict possible faults in the future; the model evaluation module is responsible for evaluating the performance of the trained model and verifying the effectiveness and reliability of the model; the model optimization module is responsible for optimizing and adjusting the parameters of the model according to the evaluation results to improve the prediction performance and generalization ability of the model. The real-time monitoring module includes a data adjustment module, a secondary simulation module and a data comparison module; the data adjustment module is responsible for real-time monitoring of the distribution network operation status and fault warning conditions, timely updating the model and adjusting the warning strategy; the secondary simulation module performs a secondary simulation based on the prediction model selected by the model selection module combined with another set of data, thereby improving the accuracy of the results and preventing unnecessary emergency measures caused by equipment failures; the data comparison module compares the data provided by the model selection module based on the secondary simulation of the secondary simulation module, and the next step is carried out if the two results are the same.

[0123] Specifically, the early warning model construction module will first extract data in conjunction with the fault feature extraction module to select a suitable prediction model. The prediction model can more intuitively observe the development trend and provide technical support for subsequent review. After selecting a suitable prediction model, the prediction model will be trained so that the prediction model can be analyzed based on the extracted data. At the same time, the model evaluation module will evaluate the performance of the prediction model to verify the effectiveness and reliability of the model. Finally, the model optimization module will optimize the model and adjust the parameters to improve the prediction performance and generalization ability of the model. While optimizing the model and adjusting the parameters, another set of data can be analyzed to obtain the analysis results, and the analysis results can be compared with the analysis results obtained using the prediction model. If the results are consistent, the next step will be performed. If the results are inconsistent, the two sets of new data will be simulated and analyzed again. If the results of the two new sets of data are the same as the results of the previous two sets of data, there is a problem of equipment failure.

[0124] Here is a possible specific method for analyzing another set of data:

[0125] The mean is calculated using the following formula:

[0126] The variance is calculated as follows:

[0127] Among them, x represents the fault characteristics of the distribution network;

[0128] represents the mean value of power grid fault characteristics;

[0129] represents the variance of distribution network fault characteristics;

[0130] The correlation coefficient between power grid fault characteristics and historical data is calculated according to the following formula:

[0131] ,in,

[0132] r represents the correlation coefficient;

[0133] y represents a data point in historical data;

[0134] Represents the mean value of a data point in the historical data of the distribution network.

[0135] The degree to which a particular value deviates from the overall mean can be calculated using the following formula:

[0136] , where Z represents the degree of deviation. The threshold range of Z can be set, and a fault is considered when it exceeds the threshold.

[0137] About the early warning module: This module automatically triggers an early warning signal when a fault risk is detected based on the early warning model construction module, reminding the staff to take corresponding measures. At the same time, the data that is abnormal but the fault mode is not identified is marked and uploaded for manual processing. The early warning module includes an alarm module, an information matching module and a data output module. Among them, the alarm module releases the corresponding early warning signal according to the severity based on the model constructed by the early warning model construction module; the information matching module matches the most suitable maintenance personnel based on the early warning signal released by the alarm module and the fault information of the distribution network; the data output module gives the maintenance personnel the corresponding maintenance data, methods and real-time data based on the maintenance personnel matched by the information matching module.

[0138] Specifically, the early warning module will release different early warning signals according to the model results constructed by the early warning model construction module in combination with the severity. At the same time, it will match the most suitable maintenance personnel according to the information and skills of the current on-duty personnel, so that the maintenance personnel can focus on research in a certain technical field and reduce unnecessary waste of energy. In addition, the corresponding maintenance data, plans and real-time data will be provided to the maintenance personnel, and the actual operating methods of the maintenance personnel will be recorded to screen useful information for the next solution.

[0139] About the data recording module: This module records and stores the corresponding data after the staff completes the fault risk processing, and stores the marked data in a separate type to facilitate the next fault identification. The data recording module includes a data storage module, which is responsible for storing the collected historical data for subsequent analysis and processing.

[0140] In this embodiment, by establishing a fault feature extraction module, the fault feature extraction module can extract fault features based on stored data. At the same time, the fault feature extraction module will select corresponding feature extraction algorithms according to different types of faults, and determine the threshold of fault features, so that the extracted data is more representative, which is convenient for maintenance by staff; by establishing an early warning model construction module, the extracted fault features are compared and analyzed with historical data to identify potential fault modes. At the same time, the system will perform data simulation analysis again based on real-time data. If the results of the two analyses are the same, an early warning signal will be released. The category of the early warning signal release is also related to the severity of the simulation results, so that emergency measures can be initiated in time when an emergency occurs; by establishing a simulation data acquisition module, two sets of relevant measurement equipment used in the simulation data acquisition module are provided, and data is transmitted by cross-operation. Then, by comparing the two sets of data results, the phenomenon of data abnormality caused by measurement equipment failure can be effectively prevented, thereby improving the accuracy of the system's judgment on distribution network related problems. At the same time, automatic measurement of data by relevant measurement equipment can also reduce errors in manual measurement and improve data accuracy.

[0141] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0142] Based on the same inventive concept, the embodiment of the present application also provides a distribution network fault warning device for implementing the distribution network fault warning method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more distribution network fault warning device embodiments provided below can refer to the limitations of the distribution network fault warning method above, and will not be repeated here.

[0143] In an exemplary embodiment, Figure 5 As shown, a distribution network fault early warning device 500 is provided, comprising:

[0144] The feature acquisition module 501 is used to obtain two sets of real-time data corresponding to the two sets of equipment based on the collection and cross-transmission of real-time information of the distribution network by the two sets of equipment; respectively extract the fault characteristics of the distribution network from the two sets of real-time data to obtain a first fault characteristic and a second fault characteristic; the fault characteristic includes at least one of the following: power parameter characteristics, topology structure characteristics, time series data characteristics, fault signal characteristics, and data statistical characteristics;

[0145] A first analysis module 502 is used to input the first fault feature into a preset analysis model; the analysis model is used to analyze the first fault feature and output corresponding first fault information; the analysis model is obtained by training based on historical data of the distribution network;

[0146] A second analysis module 503, configured to obtain second fault information based on a simulation analysis of a second fault feature;

[0147] The fault warning module 504 is used to determine the consistency between the first fault information and the second fault information; when the consistency reaches a preset standard, the fault information of the distribution network is determined and a warning signal of a corresponding level is output.

[0148] In one embodiment, the second analysis module 503 is further used to determine the corresponding feature point value and the feature overall mean value according to the second fault feature; determine the degree of deviation of the feature point value compared to the feature overall mean value; and determine the second fault information according to the degree of deviation.

[0149] In one embodiment, the second analysis module 503 is also used to determine the correlation coefficient between the second fault feature and the historical data based on the feature point value; and determine the degree of deviation of the feature point value in the second fault feature compared to the overall mean of the feature based on the feature point value, the historical data and the correlation coefficient.

[0150] In one of the embodiments, the feature acquisition module 501 is also used to extract power parameter features in the following manner: obtain the original power signal of the distribution network; determine at least one local extreme point of the original signal; determine the average value and envelope value of adjacent local extreme points; and determine the power parameter features based on the average value and the envelope value.

[0151] In one embodiment, the first analysis module 502 is further used to determine a preliminary analysis model from a plurality of preset basic analysis models according to the first fault feature; determine the fault type corresponding to the first fault feature; and optimize the preliminary analysis model based on the fault type and the corresponding simulation to obtain an analysis model.

[0152] In one embodiment, the feature acquisition module 501 is also used for real-time data including: current, voltage, equipment and line temperature, ambient humidity, power usage, circuit switch status, transformer operating status in the circuit; preprocessing the real-time data to improve data quality; preprocessing includes at least one of the following: cleaning, removing abnormal data, and filling missing values.

[0153] Each module in the above-mentioned distribution network fault warning device can be implemented in whole or in part by software, hardware and a combination thereof. Each of the above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory in the computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

[0154] In an exemplary embodiment, a distribution network fault warning device is provided. The distribution network fault warning device may be a terminal, and its internal structure diagram may be as follows: Figure 6 As shown. The distribution network fault warning device includes a processor, a memory, an input / output interface, a communication interface, and an early warning signal unit. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface and the early warning signal unit are connected to the system bus through the input / output interface. Among them, the processor of the distribution network fault warning device is used to provide computing and control capabilities. The memory of the distribution network fault warning device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the distribution network fault warning device is used to exchange information between the processor and the external device. The communication interface of the distribution network fault warning device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be realized through WIFI, a mobile cellular network, near field communication (Near Field Communication, NFC) or other technologies. When the computer program is executed by the processor, a distribution network fault warning method is realized. The early warning signal unit of the distribution network fault warning device is used to form a visually visible picture, which can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen.

[0155] Those skilled in the art will understand that Figure 6 The structure shown in the figure is only a block diagram of a partial structure related to the scheme of the present application, and does not constitute a limitation on the distribution network fault warning device to which the scheme of the present application is applied. The specific distribution network fault warning device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0156] In an exemplary embodiment, a distribution network fault warning device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the steps in the distribution network fault warning method in the above embodiment are implemented.

[0157] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the distribution network fault early warning method in the above embodiment are implemented.

[0158] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the distribution network fault early warning method in the above embodiment when executed by a processor.

[0159] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0160] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0161] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A distribution network fault early warning method, characterized in that: The method comprises: Based on the collection and cross-transmission of real-time information of the distribution network by two sets of equipment, two sets of real-time data corresponding to the two sets of equipment are obtained; Extracting fault characteristics of the distribution network from the two sets of real-time data respectively to obtain a first fault characteristic and a second fault characteristic; the fault characteristic comprises at least one of the following: a power parameter characteristic, a topological structure characteristic, a time series data characteristic, a fault signal characteristic, and a data statistical characteristic; Inputting the first fault feature into a preset analysis model; the analysis model is used to analyze the first fault feature and output corresponding first fault information; the analysis model is obtained by training based on historical data of the distribution network; Based on the simulation analysis of the second fault feature, second fault information is obtained; Determining a consistency between the first fault information and the second fault information; When the consistency reaches a preset standard, the fault information of the distribution network is determined and a warning signal of a corresponding level is output.

2. The method according to claim 1, characterized in that The obtaining of second fault information based on the simulation analysis of the second fault feature includes: According to the second fault feature, determining a corresponding feature point value and a feature overall mean value; Determining the degree of deviation of the feature point value compared to the overall mean of the feature; The second fault information is determined according to the deviation degree.

3. The method according to claim 2, characterized in that Determining the deviation of the feature point value compared to the overall mean of the feature includes: Determining, according to the characteristic point value, a correlation coefficient between the second fault characteristic and the historical data; The degree of deviation of the characteristic point value in the second fault feature compared to the overall mean value of the feature is determined according to the characteristic point value, the historical data and the correlation coefficient.

4. The method according to claim 1, characterized in that The power parameter features are extracted in the following manner: Acquiring an original power signal of the distribution network; Determine at least one local extreme point of the original signal; Determine the average value and envelope value of the adjacent local extreme value points; The power parameter characteristic is determined according to the average value and the envelope value.

5. The method according to claim 1, characterized in that The method further comprises: According to the first fault feature, determining a preliminary analysis model from a plurality of preset basic analysis models; Determine a fault type corresponding to the first fault feature; Based on the fault type and the corresponding simulation, the preliminary analysis model is optimized to obtain the analysis model.

6. The method according to claim 1, characterized in that The real-time data includes: current in the circuit, voltage, equipment and line temperature, ambient humidity, power usage, circuit switch status, transformer operation status; The method also includes: preprocessing the real-time data to improve data quality; the preprocessing includes at least one of the following: cleaning, removing abnormal data, and filling missing values.

7. A distribution network fault early warning device, characterized in that: The device comprises: A feature acquisition module is used to obtain two sets of real-time data corresponding to the two sets of equipment based on the collection and cross-transmission of real-time information of the distribution network by two sets of equipment; extract the fault characteristics of the distribution network from the two sets of real-time data respectively to obtain a first fault characteristic and a second fault characteristic; the fault characteristic includes at least one of the following: power parameter characteristics, topological structure characteristics, time series data characteristics, fault signal characteristics, and data statistical characteristics; A first analysis module, used for inputting the first fault feature into a preset analysis model; the analysis model is used for analyzing the first fault feature and outputting corresponding first fault information; the analysis model is obtained by training based on historical data of the distribution network; A second analysis module, configured to obtain second fault information based on a simulation analysis of the second fault feature; A fault warning module is used to determine the consistency between the first fault information and the second fault information; when the consistency reaches a preset standard, the fault information of the distribution network is determined and a warning signal of a corresponding level is output.

8. A distribution network fault warning device, comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images