Power distribution network fault early warning and positioning system

Abstract

The invention discloses a power distribution network fault early warning and positioning system, and belongs to the technical field of power distribution networks, and the system comprises a data collection module which is used for collecting power distribution network traveling wave monitoring information and power distribution network traveling wave real-time data; the data processing module is used for processing the traveling wave real-time data of the power distribution network; the model training module is used for constructing a power distribution network fault positioning model; the test optimization module is used for determining an optimal power distribution network fault positioning model; and the early warning and positioning module is used for carrying out early warning and positioning on the power distribution network fault in time. According to the invention, the problem of poor operation safety and reliability of the power distribution network caused by incapability of timely early warning and positioning the fault of the power distribution network based on traveling wave positioning in the prior art is solved. According to the method, the power distribution network fault can be early warned and positioned in time based on traveling wave positioning, the position of the power distribution network fault can be accurately determined in time, timely and accurate data support can be provided for subsequent maintenance work, and the operation safety and reliability of the power distribution network can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power distribution networks, and in particular to a power distribution network fault early warning and positioning system. Background Art

[0002] The distribution network is at the end of the power system and has distinct characteristics such as wide geographical distribution, large grid scale, many types of equipment, diverse network connections, and changeable operating modes. With the growth of urbanization and electricity demand, the distribution network has been continuously transformed and expanded, and its scale has continued to expand. The scale of most distribution networks at or above the county level within the State Grid Corporation system has reached more than 100 feeders, and the number of medium-voltage feeders in some medium and large cities has reached or exceeded 1,000.

[0003] Driven by the development of the national economy, power outages and power cuts have occurred; the composition ratio of loads has changed, and residential and commercial electricity consumption has increased significantly, which has increased the load on some lines and transformers, and even caused local power outages due to overload; local faults caused by defects in the power system itself have expanded, resulting in large-scale and long-term power outages; because the configuration and adjustment of reactive power sources are not adapted to the development of loads, the voltage level of the power grid cannot meet the power quality requirements, causing damage to electrical equipment and even power outages caused by voltage collapse; there are problems with excessive network losses in the operation of the distribution network, etc. Therefore, it is necessary to monitor the operating status of the distribution network in real time, and ensuring the stable operation of electrical equipment is the direction that needs to be studied and developed at present.

[0004] The Chinese patent with publication number CN109738766A discloses a distribution network fault warning system, which solves the problems of the existing distribution network fault warning system that it is inconvenient to change the working location during use and the limited warning effect during use, and generates a large amount of heat, resulting in excessive temperature and affecting use, by providing a heat dissipation fan, ventilation holes, warning data calculation module, comprehensive analysis and processing module, data storage module, data collection module, support rod, balance base, adjustment nut and rolling wheel structure; however, the patent has the following defects:

[0005] The existing ones cannot provide timely warning and positioning of distribution network faults based on traveling wave positioning, cannot timely and accurately determine the location of distribution network faults, and cannot provide timely and accurate data support for subsequent maintenance work, resulting in poor safety and reliability of distribution network operation. Summary of the invention

[0006] The purpose of the present invention is to provide a distribution network fault warning and positioning system, which can timely warn and locate distribution network faults based on traveling wave positioning, can timely and accurately determine the location of distribution network faults, can provide timely and accurate data support for subsequent maintenance work, can improve the safety and reliability of distribution network operation, and solve the problems raised in the above-mentioned background technology.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] Distribution network fault warning and positioning system, including:

[0009] The data acquisition module is used to collect the traveling wave signals and traveling wave conditions of the distribution network in real time during operation, determine the distribution network traveling wave monitoring information and the distribution network traveling wave real-time data based on traveling wave positioning;

[0010] A data processing module is used to process the real-time data of the traveling waves of the distribution network based on the traveling wave positioning, and determine the characteristic data of the traveling waves of the distribution network based on the traveling wave positioning;

[0011] Model training module, used to build a distribution network fault location model based on traveling wave positioning according to the distribution network fault warning and location requirements;

[0012] A test optimization module is used to test and optimize the distribution network fault location model based on traveling wave location to determine the optimal distribution network fault location model;

[0013] The early warning and positioning module is used to provide timely early warning and positioning of distribution network faults.

[0014] Preferably, the data acquisition module includes:

[0015] The traveling wave monitoring unit is used to monitor and capture the traveling wave signals of the distribution network in real time during operation, and determine the traveling wave monitoring information of the distribution network;

[0016] The traveling wave acquisition unit is used to monitor and collect the traveling wave conditions of the distribution network in real time during operation, and determine the real-time data of the distribution network traveling waves based on traveling wave positioning.

[0017] Preferably, the data processing module includes:

[0018] A data cleaning unit, used for cleaning the real-time data of the traveling wave of the distribution network based on the traveling wave positioning;

[0019] Obtain real-time data of traveling waves in the distribution network based on traveling wave positioning;

[0020] Clean the real-time data of the distribution network traveling wave based on traveling wave positioning, including:

[0021] Conduct consistency check on the real-time data of the distribution network traveling wave based on traveling wave positioning;

[0022] According to the reasonable value range and mutual relationship of each parameter in the distribution network traveling wave real-time data based on traveling wave positioning, check whether the distribution network traveling wave real-time data based on traveling wave positioning meets the requirements;

[0023] Remove inconsistent data that is beyond the normal range, logically unreasonable or contradictory in the real-time data of the distribution network based on traveling wave positioning;

[0024] Process invalid and missing values ​​of real-time data of traveling waves in distribution network based on traveling wave positioning;

[0025] Remove invalid data and missing data that are of no value to distribution network fault warning and positioning in the real-time data of distribution network traveling waves based on traveling wave positioning;

[0026] Determine the real-time data of distribution network traveling waves that are valuable for distribution network fault warning and location.

[0027] Preferably, the data processing module further includes:

[0028] A feature extraction unit, used for extracting features from the cleaned distribution network traveling wave real-time data;

[0029] Obtain real-time data of distribution network traveling waves that are valuable for early warning and location of distribution network faults after cleaning;

[0030] Extract features from real-time data of distribution network traveling waves that are valuable for early warning and location of distribution network faults;

[0031] Extract features that can reflect the early warning and location of distribution network faults;

[0032] Determine the traveling wave characteristic data of the distribution network based on traveling wave positioning.

[0033] Preferably, the model training module includes:

[0034] A data collection unit, used for collecting historical data of traveling waves of the distribution network based on traveling wave positioning;

[0035] According to the distribution network fault warning and positioning requirements, the distribution network traveling wave historical data based on traveling wave positioning is collected;

[0036] A model building unit, used for building a distribution network fault location model based on traveling wave location;

[0037] Obtain the historical data of traveling waves of the distribution network based on traveling wave positioning;

[0038] Divide the historical data of traveling waves of distribution network based on traveling wave positioning;

[0039] Determine the training data set and the test data set;

[0040] Select a neural network model framework suitable for distribution network fault location;

[0041] Based on the training data set, train the selected neural network model framework;

[0042] A distribution network fault location model based on traveling wave positioning is determined.

[0043] Preferably, the test optimization module includes:

[0044] A performance test unit, used to test the distribution network fault location model based on traveling wave location;

[0045] Obtain a distribution network fault location model based on traveling wave positioning;

[0046] Based on the test data set, the performance of the distribution network fault location model based on traveling wave positioning is tested;

[0047] Determine the performance test results based on the distribution network fault location model;

[0048] An adjustment and optimization unit is used to optimize a distribution network fault location model based on traveling wave location;

[0049] Obtain performance test results based on distribution network fault location model;

[0050] Mining and analyzing the distribution network fault location model based on traveling wave positioning;

[0051] Determine the parameter adjustment optimization scheme based on the distribution network fault location model;

[0052] Optimizing the parameters of the distribution network fault location model according to a parameter adjustment optimization scheme based on the distribution network fault location model;

[0053] Determine the optimal distribution network fault location model.

[0054] Preferably, the early warning positioning module includes:

[0055] Fault warning unit, used to provide timely warning of distribution network faults;

[0056] Obtaining distribution network traveling wave monitoring information;

[0057] Mining and analyzing the traveling wave monitoring information of the distribution network;

[0058] When the distribution network traveling wave monitoring information contains traveling wave signals, it indicates that there is a fault in the distribution network, and a timely fault warning is given to the distribution network;

[0059] When the distribution network traveling wave monitoring information does not contain a traveling wave signal, it indicates that there is no fault behavior in the distribution network and no fault warning is issued for the distribution network.

[0060] Preferably, the early warning positioning module also includes:

[0061] Fault location unit, used to locate distribution network faults in a timely manner;

[0062] Among them, the optimal distribution network fault location model is obtained;

[0063] Acquire the traveling wave characteristic data of the distribution network based on traveling wave positioning;

[0064] Input the distribution network traveling wave characteristic data based on traveling wave positioning into the optimal distribution network fault location model;

[0065] Based on the optimal distribution network fault location model, the distribution network traveling wave characteristic data based on traveling wave location is predicted and evaluated, and fault location is performed;

[0066] Determine the distribution network fault location result based on traveling wave location;

[0067] Among them, the distribution network fault location result based on traveling wave positioning includes the distribution network fault location coordinates.

[0068] Preferably, the early warning positioning module includes:

[0069] A comparison module is used to calculate the optimal distribution network fault location model to determine the location error of the distribution network fault node after locating the distribution network fault based on the optimal distribution network fault location model, and compare it with the error threshold;

[0070] A determination module, used to generate abnormal information about the current fault location when it is determined that the location error is greater than a preset error threshold, and update the model parameters of the distribution network fault location model;

[0071] The comparison module comprises:

[0072] The first calculation module is used to calculate the probability p of the i-th location abnormality occurring when locating the fault node of the distribution network i :

[0073]

[0074] Wherein, C is a combination symbol; m is the number of positioning times when locating a faulty node in the distribution network; λ is the maximum classification error rate when distinguishing the faulty node category when locating a faulty node in the distribution network;

[0075] The second calculation module is used to calculate the probability p of the i-th location abnormality occurring when locating the fault node of the distribution network. i , calculate and determine the location error W of the fault node in the distribution network:

[0076]

[0077] Among them, Z is the total number of fault node categories in the distribution network; U is the working area of ​​the distribution network; and D is the deviation of the positioning path when the i-th positioning anomaly occurs.

[0078] Preferably, the system further comprises: a component positioning module, for:

[0079] After determining the faulty equipment of the distribution network based on the early warning positioning module, obtain the working signals of several working components constituting the faulty equipment under the current working condition;

[0080] Extracting features from the plurality of working signals respectively to obtain time domain features corresponding to each working signal; the time domain features include mean, variance, standard deviation, peak value, and valley value;

[0081] Combining a feature vector according to the time domain features corresponding to each working signal; performing cluster analysis on a plurality of feature vectors based on a K-means clustering algorithm to obtain a plurality of cluster sets, and obtaining cluster centers of the cluster sets; constructing a distance matrix according to the plurality of cluster centers; wherein the rows and columns of the distance matrix correspond to different cluster centers respectively, and each element in the matrix represents the distance between the corresponding row and column cluster centers;

[0082] Determine the distance matrix of all working conditions of the faulty equipment in the faulty state, and generate a first distance matrix group;

[0083] Determine the distance matrix of all working conditions of the faulty equipment in the normal state, and generate a second distance matrix group;

[0084] Compare the distance matrices of the first distance matrix group and the second distance matrix group under corresponding working conditions to determine normal working conditions and abnormal working conditions;

[0085] Determine the newly added work component information under normal working conditions and abnormal working conditions as the work components to be screened;

[0086] Perform electromagnetic interference test and anti-electromagnetic interference test on the working components to be screened, eliminate the working components to be screened that pass the electromagnetic interference test and anti-electromagnetic interference test, and obtain abnormal working components.

[0087] Compared with the prior art, the present invention has the following beneficial effects:

[0088] 1. The present invention determines the distribution network traveling wave monitoring information by real-time monitoring and capturing the traveling wave signals of the distribution network during operation. According to the distribution network fault warning requirements, the distribution network traveling wave monitoring information is mined and analyzed. When the distribution network traveling wave monitoring information contains traveling wave signals, it indicates that there is a fault behavior in the distribution network, and a timely fault warning is given to the distribution network.

[0089] 2. The present invention determines the real-time data of the distribution network traveling waves based on traveling wave positioning by real-time monitoring and collection of the traveling wave conditions of the distribution network during operation, determines the traveling wave characteristic data of the distribution network based on traveling wave positioning by processing the real-time data of the distribution network traveling waves based on traveling wave positioning, and predicts and evaluates the traveling wave characteristic data of the distribution network based on traveling wave positioning and locates the fault based on the optimal distribution network fault location model according to the distribution network fault location requirement, determines the distribution network fault location result based on traveling wave positioning, can timely warn and locate the distribution network fault based on traveling wave positioning, can timely and accurately determine the distribution network fault location, can provide timely and accurate data support for subsequent maintenance work, and can improve the operation safety and reliability of the distribution network. BRIEF DESCRIPTION OF THE DRAWINGS

[0090] Figure 1 It is a structural schematic diagram of the distribution network fault early warning and positioning system of the present invention;

[0091] Figure 2 This is an algorithm flow chart of the distribution network fault early warning and positioning system of the present invention. DETAILED DESCRIPTION

[0092] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0093] In order to solve the existing problems of not being able to timely warn and locate distribution network faults based on traveling wave positioning, not being able to timely and accurately determine the location of distribution network faults, and not being able to provide timely and accurate data support for subsequent maintenance work, resulting in poor distribution network operation safety and reliability, please refer to Figure 1-Figure 2 , this embodiment provides the following technical solutions:

[0094] The distribution network fault early warning and positioning system includes: a data acquisition module, a data processing module, a model training module, a test optimization module and a early warning positioning module.

[0095] It should be noted that the data acquisition module is used to collect the traveling wave signals and traveling wave conditions of the distribution network in real time during operation, and the distribution network traveling wave monitoring information and the distribution network traveling wave real-time data based on traveling wave positioning are determined; the distribution network traveling wave real-time data based on traveling wave positioning is processed by the data processing module to determine the distribution network traveling wave characteristic data based on traveling wave positioning; the distribution network fault location model based on traveling wave positioning is constructed by the model training module; the distribution network fault location model based on traveling wave positioning is tested and optimized by the test optimization module to determine the optimal distribution network fault location model; the early warning and positioning module is used to timely warn and locate the distribution network fault; the distribution network fault can be timely warned and located based on traveling wave positioning, the distribution network fault location can be timely and accurately determined, and timely and accurate data support can be provided for subsequent maintenance work, which can improve the safety and reliability of distribution network operation.

[0096] Among them, the data acquisition module is used to collect the traveling wave signals and traveling wave conditions of the distribution network in real time during operation, determine the distribution network traveling wave monitoring information and the distribution network traveling wave real-time data based on traveling wave positioning;

[0097] In this embodiment, the data acquisition module includes:

[0098] The traveling wave monitoring unit is used to monitor and capture the traveling wave signals of the distribution network in real time during operation, and determine the traveling wave monitoring information of the distribution network;

[0099] The traveling wave acquisition unit is used to monitor and collect the traveling wave conditions of the distribution network in real time during operation, and determine the real-time data of the distribution network traveling waves based on traveling wave positioning.

[0100] It should be noted that traveling wave positioning technology is a technology that locates faults based on traveling wave phenomena in electromagnetic transient processes. When a fault occurs in the power system, such as short circuit, grounding or other abnormalities, voltage and current traveling waves will be propagated on the power cables. These two traveling waves propagate along the lines to both ends at a speed close to the speed of light. By arranging distributed traveling wave sensors or monitoring terminals on the distribution network lines, the traveling wave signals generated by distribution network faults can be captured in real time. These signals contain the time information of the traveling waves arriving at each sensor. By comparing the time difference of traveling wave signals received by sensors at different positions, the specific location of the fault point can be calculated using traveling wave ranging technology.

[0101] The data processing module is used to process the real-time data of the traveling wave of the distribution network based on the traveling wave positioning, and determine the characteristic data of the traveling wave of the distribution network based on the traveling wave positioning;

[0102] In this embodiment, the data processing module includes:

[0103] A data cleaning unit, used for cleaning the real-time data of the traveling wave of the distribution network based on the traveling wave positioning;

[0104] Obtain real-time data of traveling waves in the distribution network based on traveling wave positioning;

[0105] Clean the real-time data of the distribution network traveling wave based on traveling wave positioning, including:

[0106] Conduct consistency check on the real-time data of the distribution network traveling wave based on traveling wave positioning;

[0107] According to the reasonable value range and mutual relationship of each parameter in the distribution network traveling wave real-time data based on traveling wave positioning, check whether the distribution network traveling wave real-time data based on traveling wave positioning meets the requirements;

[0108] Remove inconsistent data that is beyond the normal range, logically unreasonable or contradictory in the real-time data of the distribution network based on traveling wave positioning;

[0109] Process invalid and missing values ​​of real-time data of traveling waves in distribution network based on traveling wave positioning;

[0110] Remove invalid data and missing data that are of no value to distribution network fault warning and positioning in the real-time data of distribution network traveling waves based on traveling wave positioning;

[0111] Determine the real-time data of distribution network traveling waves that are valuable for early warning and location of distribution network faults;

[0112] A feature extraction unit, used for extracting features from the cleaned distribution network traveling wave real-time data;

[0113] Obtain real-time data of distribution network traveling waves that are valuable for early warning and location of distribution network faults after cleaning;

[0114] Extract features from real-time data of distribution network traveling waves that are valuable for early warning and location of distribution network faults;

[0115] Extract features that can reflect the early warning and location of distribution network faults;

[0116] Determine the traveling wave characteristic data of the distribution network based on traveling wave positioning.

[0117] Among them, the model training module is used to build a distribution network fault location model based on traveling wave positioning according to the distribution network fault warning and location requirements;

[0118] In this embodiment, the model training module includes:

[0119] A data collection unit, used for collecting historical data of traveling waves of the distribution network based on traveling wave positioning;

[0120] According to the distribution network fault warning and positioning requirements, the distribution network traveling wave historical data based on traveling wave positioning is collected;

[0121] A model building unit, used for building a distribution network fault location model based on traveling wave location;

[0122] Obtain the historical data of traveling waves of the distribution network based on traveling wave positioning;

[0123] Divide the historical data of traveling waves of distribution network based on traveling wave positioning;

[0124] Determine the training data set and the test data set;

[0125] Select a neural network model framework suitable for distribution network fault location;

[0126] Based on the training data set, train the selected neural network model framework;

[0127] A distribution network fault location model based on traveling wave positioning is determined.

[0128] Among them, the test optimization module is used to test and optimize the distribution network fault location model based on traveling wave positioning to determine the optimal distribution network fault location model;

[0129] In this embodiment, the test optimization module includes:

[0130] A performance test unit, used to test the distribution network fault location model based on traveling wave location;

[0131] Obtain a distribution network fault location model based on traveling wave positioning;

[0132] Based on the test data set, the performance of the distribution network fault location model based on traveling wave positioning is tested;

[0133] Determine the performance test results based on the distribution network fault location model;

[0134] An adjustment and optimization unit is used to optimize a distribution network fault location model based on traveling wave location;

[0135] Obtain performance test results based on distribution network fault location model;

[0136] Mining and analyzing the distribution network fault location model based on traveling wave positioning;

[0137] Determine the parameter adjustment optimization scheme based on the distribution network fault location model;

[0138] Optimizing the parameters of the distribution network fault location model according to a parameter adjustment optimization scheme based on the distribution network fault location model;

[0139] Determine the optimal distribution network fault location model.

[0140] Among them, the early warning and positioning module is used to provide timely early warning and positioning of distribution network faults.

[0141] In this embodiment, the early warning positioning module includes:

[0142] Fault warning unit, used to provide timely warning of distribution network faults;

[0143] Obtaining distribution network traveling wave monitoring information;

[0144] Mining and analyzing the traveling wave monitoring information of the distribution network;

[0145] When the distribution network traveling wave monitoring information contains traveling wave signals, it indicates that there is a fault in the distribution network, and a timely fault warning is given to the distribution network;

[0146] When the distribution network traveling wave monitoring information does not contain traveling wave signals, it indicates that there is no fault behavior in the distribution network, and no fault warning is issued for the distribution network;

[0147] Fault location unit, used to locate distribution network faults in a timely manner;

[0148] Among them, the optimal distribution network fault location model is obtained;

[0149] Acquire the traveling wave characteristic data of the distribution network based on traveling wave positioning;

[0150] Input the distribution network traveling wave characteristic data based on traveling wave positioning into the optimal distribution network fault location model;

[0151] Based on the optimal distribution network fault location model, the distribution network traveling wave characteristic data based on traveling wave location is predicted and evaluated, and fault location is performed;

[0152] Determine the distribution network fault location result based on traveling wave location;

[0153] Among them, the distribution network fault location result based on traveling wave positioning includes the distribution network fault location coordinates.

[0154] Among them, the coordinates of the distribution network fault location are displayed to the distribution network maintenance personnel in a visual form, which is convenient for the distribution network maintenance personnel to carry out timely maintenance of the distribution network fault and improve the safety and reliability of distribution network operation.

[0155] Therefore, by real-time monitoring and capturing of traveling wave signals during the operation of the distribution network, the traveling wave monitoring information of the distribution network is determined. According to the fault warning needs of the distribution network, the traveling wave monitoring information of the distribution network is mined and analyzed. When the traveling wave monitoring information of the distribution network contains traveling wave signals, it indicates that there is a fault behavior in the distribution network, and a timely fault warning is given to the distribution network.

[0156] Therefore, by real-time monitoring and collection of traveling wave conditions of the distribution network during operation, the real-time data of distribution network traveling waves based on traveling wave positioning is determined, and by processing the real-time data of distribution network traveling waves based on traveling wave positioning, the traveling wave characteristic data of the distribution network based on traveling wave positioning is determined. According to the demand for distribution network fault location, the traveling wave characteristic data of the distribution network based on traveling wave positioning is predicted, evaluated and faults are located based on the optimal distribution network fault location model, and the distribution network fault location result based on traveling wave positioning is determined. Based on traveling wave positioning, timely warning and positioning of distribution network faults can be carried out, the distribution network fault location can be determined timely and accurately, and timely and accurate data support can be provided for subsequent maintenance work, which can improve the safety and reliability of distribution network operation.

[0157] Preferably, the early warning positioning module includes:

[0158] A comparison module is used to calculate the optimal distribution network fault location model to determine the location error of the distribution network fault node after locating the distribution network fault based on the optimal distribution network fault location model, and compare it with the error threshold;

[0159] A determination module, used to generate abnormal information about the current fault location when it is determined that the location error is greater than a preset error threshold, and update the model parameters of the distribution network fault location model;

[0160] The comparison module comprises:

[0161] The first calculation module is used to calculate the probability p of the i-th location abnormality occurring when locating the fault node of the distribution network i :

[0162]

[0163] Wherein, C is a combination symbol; m is the number of positioning times when locating a faulty node in the distribution network; λ is the maximum classification error rate when distinguishing the faulty node category when locating a faulty node in the distribution network;

[0164] The second calculation module is used to calculate the probability p of the i-th location abnormality occurring when locating the fault node of the distribution network. i , calculate and determine the location error W of the fault node in the distribution network:

[0165]

[0166] Among them, Z is the total number of fault node categories in the distribution network; U is the working area of ​​the distribution network; and D is the deviation of the positioning path when the i-th positioning anomaly occurs.

[0167] The working principle of the above technical solution is as follows: in the comparison module, after locating the distribution network fault based on the optimal distribution network fault location model, the optimal distribution network fault location model is calculated to determine the location error of the distribution network fault node, and the location error includes the location error of the abnormal node category. The error threshold represents the threshold of the distance value between the preset categories. When it is determined that the location error is greater than the preset error threshold, abnormal information of the current fault location is generated, that is, an abnormal report containing fault location details, error size, timestamp and other information is generated. At the same time, the model parameters of the distribution network fault location model are updated, such as formulating an update strategy for the model parameters, including adjusting the value of specific parameters, introducing new parameters, or changing the optimization method of parameters.

[0168] The maximum classification error rate ranges from (0,1). First, the probability of the i-th location abnormality when locating the distribution network fault node is calculated based on the first calculation module. Secondly, according to the probability p of the i-th location abnormality when locating the distribution network fault node, i , calculate and determine the location error of the faulty node in the distribution network. The working area of ​​the distribution network represents the area formed by each node of the distribution network. The deviation of the positioning path when the i-th positioning anomaly occurs represents the distance between the node with the positioning anomaly and the actual abnormal node.

[0169] The beneficial effects of the above technical solution are as follows: after locating the distribution network fault based on the early warning positioning module, positioning detection is performed to determine the positioning error of the distribution network fault node. When it is determined that the positioning error is greater than the preset error threshold, abnormal information about the current fault positioning is generated, and the model parameters of the distribution network fault positioning model are updated at the same time, so as to improve the accuracy of the distribution network fault positioning. Based on the above formula, the positioning error of the distribution network fault node is accurately calculated, which improves the accuracy of judging the size of the positioning error and the error threshold, thereby improving the reliability of the system.

[0170] Preferably, the system further comprises: a component positioning module, for:

[0171] After determining the faulty equipment of the distribution network based on the early warning positioning module, obtain the working signals of several working components constituting the faulty equipment under the current working condition;

[0172] Extracting features from the plurality of working signals respectively to obtain time domain features corresponding to each working signal; the time domain features include mean, variance, standard deviation, peak value, and valley value;

[0173] Combining a feature vector according to the time domain features corresponding to each working signal; performing cluster analysis on a plurality of feature vectors based on a K-means clustering algorithm to obtain a plurality of cluster sets, and obtaining cluster centers of the cluster sets; constructing a distance matrix according to the plurality of cluster centers; wherein the rows and columns of the distance matrix correspond to different cluster centers respectively, and each element in the matrix represents the distance between the corresponding row and column cluster centers;

[0174] Determine the distance matrix of all working conditions of the faulty equipment in the faulty state, and generate a first distance matrix group;

[0175] Determine the distance matrix of all working conditions of the faulty equipment in the normal state, and generate a second distance matrix group;

[0176] Compare the distance matrices of the first distance matrix group and the second distance matrix group under corresponding working conditions to determine normal working conditions and abnormal working conditions;

[0177] Determine the newly added work component information under normal working conditions and abnormal working conditions as the work components to be screened;

[0178] Perform electromagnetic interference test and anti-electromagnetic interference test on the working components to be screened, eliminate the working components to be screened that pass the electromagnetic interference test and anti-electromagnetic interference test, and obtain abnormal working components.

[0179] Working principle of the above technical solution: In this embodiment, first, it is determined which working components the faulty device is composed of. These components include sensors, actuators, and circuit boards. The working signals of each working component are obtained in real time or from historical records. These signals include various types of data such as voltage, current, temperature, and vibration.

[0180] Feature extraction is performed on the several working signals to obtain the time domain features corresponding to each working signal; Mean: the average value of the signal, reflecting the central trend of the signal. Variance: the average value of the square of the deviation of the signal from its mean, measuring the discreteness of the signal. Standard deviation: the square root of the variance, which is also an indicator to measure the discreteness of the signal. Peak: the maximum value of the signal within a given time range, reflecting the strength or extreme situation of the signal. Valley: the minimum value of the signal within a given time range, corresponding to the peak, providing another extreme of the signal change.

[0181] Combine the time domain features (mean, variance, standard deviation, peak, valley) corresponding to each working signal into a feature vector. Each feature vector represents a specific state or behavior mode of the faulty equipment under the current working condition. Set the number of clusters K and initialize the cluster centers: randomly select K feature vectors as the initial cluster centers; for each feature vector, calculate its distance (Euclidean distance) to all cluster centers and assign it to the nearest cluster center. For each cluster, recalculate its cluster center, that is, the mean of all feature vectors in the cluster. Repeat the iteration until the cluster center no longer changes significantly or the preset number of iterations is reached. K cluster centers obtained from the K-means clustering algorithm. Create a K×K matrix in which rows and columns correspond to different cluster centers. For each element (i, j) in the matrix, calculate the distance between the i-th cluster center and the j-th cluster center. Fill the calculated distance value into the corresponding position of the distance matrix. The distance matrix is ​​used to assess the degree of separation between clusters and the compactness within clusters by examining the distances between cluster centers in the distance matrix.

[0182] In this embodiment, the distance matrix of all working conditions of the faulty device in the faulty state and the distance matrix of all working conditions of the faulty device in the normal state are determined, and the distance matrices of the first distance matrix group and the second distance matrix group under the corresponding working conditions are compared to determine the normal working condition and the abnormal working condition; the newly added working component information under the normal working condition and the abnormal working condition is determined as the working component to be screened (possibly abnormal); the electromagnetic interference test and the anti-electromagnetic interference test are performed on the working components to be screened, and the working components to be screened that pass the electromagnetic interference test and the anti-electromagnetic interference test are eliminated to obtain the abnormal working components.

[0183] The beneficial effects of the above technical solution are as follows: after determining the faulty equipment in the distribution network based on the early warning positioning module, the abnormal working components in the faulty equipment are accurately determined based on the component positioning module, thereby further improving the accuracy of abnormal positioning, facilitating timely maintenance and improving the reliability of the equipment.

[0184] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0185] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. Distribution network fault warning and positioning system, characterized in that: include: The data acquisition module is used to collect the traveling wave signals and traveling wave conditions of the distribution network in real time during operation, determine the distribution network traveling wave monitoring information and the distribution network traveling wave real-time data based on traveling wave positioning; A data processing module is used to process the real-time data of the traveling waves of the distribution network based on the traveling wave positioning, and determine the characteristic data of the traveling waves of the distribution network based on the traveling wave positioning; Model training module, used to build a distribution network fault location model based on traveling wave positioning according to the distribution network fault warning and location requirements; A test optimization module is used to test and optimize the distribution network fault location model based on traveling wave location to determine the optimal distribution network fault location model; The early warning and positioning module is used to provide timely early warning and positioning of distribution network faults.

2. The distribution network fault early warning and positioning system according to claim 1, characterized in that: The data acquisition module comprises: The traveling wave monitoring unit is used to monitor and capture the traveling wave signals of the distribution network in real time during operation, and determine the traveling wave monitoring information of the distribution network; The traveling wave acquisition unit is used to monitor and collect the traveling wave conditions of the distribution network in real time during operation, and determine the real-time data of the distribution network traveling waves based on traveling wave positioning.

3. The distribution network fault early warning and positioning system according to claim 2, characterized in that: The data processing module comprises: A data cleaning unit, used for cleaning the real-time data of the traveling wave of the distribution network based on the traveling wave positioning; Obtain real-time data of traveling waves in the distribution network based on traveling wave positioning; Clean the real-time data of the distribution network traveling wave based on traveling wave positioning, including: Conduct consistency check on the real-time data of the distribution network traveling wave based on traveling wave positioning; According to the reasonable value range and mutual relationship of each parameter in the distribution network traveling wave real-time data based on traveling wave positioning, check whether the distribution network traveling wave real-time data based on traveling wave positioning meets the requirements; Remove inconsistent data that is beyond the normal range, logically unreasonable or contradictory in the real-time data of the distribution network based on traveling wave positioning; Process invalid and missing values ​​of real-time data of traveling waves in distribution network based on traveling wave positioning; Remove invalid data and missing data that are of no value to distribution network fault warning and positioning in the real-time data of distribution network traveling waves based on traveling wave positioning; Determine the real-time data of distribution network traveling waves that are valuable for distribution network fault warning and location.

4. The distribution network fault early warning and positioning system according to claim 3, characterized in that: The data processing module also includes: A feature extraction unit, used for extracting features from the cleaned distribution network traveling wave real-time data; Obtain real-time data of distribution network traveling waves that are valuable for early warning and location of distribution network faults after cleaning; Extract features from real-time data of distribution network traveling waves that are valuable for early warning and location of distribution network faults; Extract features that can reflect the early warning and location of distribution network faults; Determine the traveling wave characteristic data of the distribution network based on traveling wave positioning.

5. The distribution network fault early warning and positioning system according to claim 4, characterized in that: The model training module includes: A data collection unit, used for collecting historical data of traveling waves of the distribution network based on traveling wave positioning; According to the distribution network fault warning and positioning requirements, the distribution network traveling wave historical data based on traveling wave positioning is collected; A model building unit, used for building a distribution network fault location model based on traveling wave location; Obtain the historical data of traveling waves of the distribution network based on traveling wave positioning; Divide the historical data of distribution network traveling waves based on traveling wave positioning; Determine the training data set and the test data set; Select a neural network model framework suitable for distribution network fault location; Based on the training data set, train the selected neural network model framework; A distribution network fault location model based on traveling wave positioning is determined.

6. The distribution network fault early warning and positioning system according to claim 5, characterized in that: The test optimization module includes: A performance test unit, used to test the distribution network fault location model based on traveling wave location; Obtain a distribution network fault location model based on traveling wave positioning; Based on the test data set, the performance of the distribution network fault location model based on traveling wave positioning is tested; Determine the performance test results based on the distribution network fault location model; An adjustment and optimization unit is used to optimize a distribution network fault location model based on traveling wave location; Obtain performance test results based on distribution network fault location model; Mining and analyzing the distribution network fault location model based on traveling wave positioning; Determine the parameter adjustment optimization scheme based on the distribution network fault location model; Optimizing the parameters of the distribution network fault location model according to a parameter adjustment optimization scheme based on the distribution network fault location model; Determine the optimal distribution network fault location model.

7. The distribution network fault early warning and positioning system according to claim 6, characterized in that: The early warning positioning module includes: Fault warning unit, used to provide timely warning of distribution network faults; Obtaining distribution network traveling wave monitoring information; Mining and analyzing the traveling wave monitoring information of the distribution network; When the distribution network traveling wave monitoring information contains traveling wave signals, it indicates that there is a fault in the distribution network, and a timely fault warning is given to the distribution network; When the distribution network traveling wave monitoring information does not contain a traveling wave signal, it indicates that there is no fault behavior in the distribution network and no fault warning is issued for the distribution network.

8. The distribution network fault early warning and positioning system according to claim 7, characterized in that: The early warning positioning module also includes: Fault location unit, used to locate distribution network faults in a timely manner; Among them, the optimal distribution network fault location model is obtained; Acquire the traveling wave characteristic data of the distribution network based on traveling wave positioning; Input the distribution network traveling wave characteristic data based on traveling wave positioning into the optimal distribution network fault location model; Based on the optimal distribution network fault location model, the distribution network traveling wave characteristic data based on traveling wave location is predicted and evaluated, and fault location is performed; Determine the distribution network fault location result based on traveling wave location; Among them, the distribution network fault location result based on traveling wave positioning includes the distribution network fault location coordinates.

9. The distribution network fault early warning and positioning system according to claim 1, characterized in that: The early warning positioning module includes: A comparison module is used to calculate the optimal distribution network fault location model to determine the location error of the distribution network fault node after locating the distribution network fault based on the optimal distribution network fault location model, and compare it with the error threshold; A determination module, used to generate abnormal information about the current fault location when it is determined that the location error is greater than a preset error threshold, and update the model parameters of the distribution network fault location model; The comparison module comprises: The first calculation module is used to calculate the probability p of the i-th location abnormality occurring when locating the fault node of the distribution network i : Wherein, C is a combination symbol; m is the number of positioning times when locating a faulty node in the distribution network; λ is the maximum classification error rate when distinguishing the faulty node category when locating a faulty node in the distribution network; The second calculation module is used to calculate the probability p of the i-th location abnormality occurring when locating the fault node of the distribution network. i , calculate and determine the location error W of the fault node in the distribution network: Among them, Z is the total number of fault node categories in the distribution network; U is the working area of ​​the distribution network; and D is the deviation of the positioning path when the i-th positioning anomaly occurs.

10. The distribution network fault early warning and positioning system according to claim 1, characterized in that: The system also includes: a component positioning module, for: After determining the faulty equipment of the distribution network based on the early warning positioning module, obtain the working signals of several working components constituting the faulty equipment under the current working condition; Extracting features from the plurality of working signals respectively to obtain time domain features corresponding to each working signal; the time domain features include mean, variance, standard deviation, peak value, and valley value; Combining a feature vector according to the time domain features corresponding to each working signal; performing cluster analysis on a plurality of feature vectors based on a K-means clustering algorithm to obtain a plurality of cluster sets, and obtaining cluster centers of the cluster sets; constructing a distance matrix according to the plurality of cluster centers; wherein the rows and columns of the distance matrix correspond to different cluster centers respectively, and each element in the matrix represents the distance between the corresponding row and column cluster centers; Determine the distance matrix of all working conditions of the faulty equipment in the faulty state, and generate a first distance matrix group; Determine the distance matrix of all working conditions of the faulty equipment in the normal state, and generate a second distance matrix group; Compare the distance matrices of the first distance matrix group and the second distance matrix group under corresponding working conditions to determine normal working conditions and abnormal working conditions; Determine the newly added work component information under normal working conditions and abnormal working conditions as the work components to be screened; Perform electromagnetic interference test and anti-electromagnetic interference test on the working components to be screened, eliminate the working components to be screened that pass the electromagnetic interference test and anti-electromagnetic interference test, and obtain abnormal working components.

Images