Distribution network equipment risk hidden danger studying and judging method and system based on short-time ground loss signal

Abstract

The invention relates to a distribution network equipment risk hidden danger studying and judging method and system based on a short-time ground loss signal, and the method comprises the steps: obtaining a historical short-time ground loss signal, extracting a key field of the historical short-time ground loss signal as a sample attribute through a regularization expression, and enabling the key field to comprise a distribution network equipment risk hidden danger and a ground loss index; extracting land loss index features, drawing a distribution network equipment risk hidden danger-land loss index feature sample distribution map, and calculating a preliminary probability value of the distribution network equipment risk hidden danger by using a large number theorem; constructing a land loss index feature matrix, performing standardization processing, calculating a probability matrix and information entropy, and calculating an initial weight according to the information entropy; obtaining a judgment weight by using the AHP, and obtaining a comprehensive weight in combination with the initial weight and the judgment weight; combining the preliminary probability value and the comprehensive weight to obtain a distribution network equipment risk hidden danger comprehensive probability, and performing analysis to obtain a research and judgment result; the invention provides a method for studying and judging the hidden danger of the specific distribution network equipment so as to assist on-site repair personnel to accurately and quickly process faults.

Description

Technical Field

[0001] The present invention relates to the field of power system detection, and mainly to a method and system for analyzing and judging risk hazards of distribution network equipment based on short-term ground loss signals. Background Art

[0002] In recent years, with the rapid development of the power industry, the power network has become more and more complex, especially the distribution network. The distribution network covers a wide area, the wiring is intricate, the power supply environment is uneven, and the quality and loss of distribution equipment and lines in different regions are different. After being disturbed by the outside world, power outages frequently occur, among which 10kV line short circuit faults are a common type of fault in the distribution network. With the rapid development of the urban economy and the continuous improvement of people's living standards, the requirements for power quality are becoming increasingly stringent. Once a power outage occurs due to a short circuit fault, it will bring inconvenience to people's normal production and life, and will also affect the development of the national economy.

[0003] A short-term ground loss signal refers to a signal generated when a single-phase ground fault is caused by some reason in the distribution network and is restored in a short time. These reasons are often branches or foreign objects that are short-circuited on the line for a short time under the action of wind, or short-term surface discharge grounding caused by defects in the equipment itself. According to statistics, more than 70% of distribution lines will send multiple ground loss signals before a single-phase ground fault occurs. Therefore, how to deeply explore the ground loss fault signal has great application potential in judging the potential risks and hidden dangers of distribution network equipment and improving the level of distribution network operation and maintenance.

[0004] For example, CN117347788A "A method for predicting the probability of single-phase grounding fault category in a distribution network" discloses "a method for predicting the probability of single-phase grounding fault category in a distribution network, comprising: S1. According to the mainstream method, constructing a number of fault features reflecting the grounding fault to form a fault feature set X; S2. Combining the multi-layer perceptron method, constructing a single-phase grounding fault category identification learning model with a mapping relationship between fault features and multiple fault categories, and introducing multiple types of activation functions in the output layer of the single-phase grounding fault category identification learning model, and performing a probabilistic conversion on the results of the output layer obtained by the activation function; S3. Introducing the cross entropy method to perform reverse weight and bias correction on the single-phase grounding fault category identification learning model, so that the corrected single-phase grounding fault category identification learning model can The probability of predicting the fault category is predicted based on the fault characteristics. However, the prediction of the single-phase grounding fault category by the method is based on the in-depth analysis of the ground loss signal of a long-term single-phase fault; the multiple short-term ground loss warning signals that may be issued before the ground loss fault occurs are not fully explored and analyzed to achieve an effective prediction of the probability of risk hazards of distribution network equipment before the fault; in addition, the method relies on the probability conversion of the multi-layer perceptron model and its activation function, resulting in poor robustness of the model when processing complex fault characteristics; in addition, the model of the method needs to be trained based on a large number of fault characteristics, involving a relatively complex back propagation and weight correction process, resulting in a large computational overhead, unable to adapt to the dynamically changing distribution network environment in a timely manner, and unable to accurately predict its probability distribution, affecting the accuracy and reliability of fault identification. Summary of the invention

[0005] In order to solve the above-mentioned problems existing in the prior art, the present application provides a method and system for analyzing and assessing the risk hazards of distribution network equipment based on short-term ground loss signals.

[0006] The technical solution of this application is as follows:

[0007] A method for analyzing and judging the risk of distribution network equipment based on short-term ground loss signals, the method comprising:

[0008] Obtain historical short-term land loss signals, and use regularized expressions to extract key fields of the historical short-term land loss signals as sample attributes, wherein the sample attributes include risk hazards of distribution network equipment and land loss indicators, and the land loss indicators include the number of land losses, the duration of land loss, and the area of ​​land loss;

[0009] Extracting features of the land loss indicator to obtain land loss indicator features; drawing a sample distribution map of distribution network equipment risk hazards-land loss indicator features, and calculating and obtaining preliminary probability values ​​of distribution network equipment risk hazards in each land loss indicator feature based on the sample distribution map and Bernoulli's law of large numbers;

[0010] Constructing an N×m land loss indicator feature matrix, performing standardization processing on the land loss indicator feature matrix to obtain a standardized matrix; calculating a probability matrix corresponding to the standardized matrix, obtaining information entropy of the land loss indicator based on the probability matrix, and calculating an initial weight corresponding to the land loss indicator according to the information entropy;

[0011] The analytic hierarchy process (AHP) was used to analyze the characteristics of land loss indicators to obtain the judgment weights; the initial weights and judgment weights were combined to obtain the comprehensive weights of the characteristics of land loss indicators;

[0012] By combining the preliminary probability values ​​of distribution network equipment risk hazards in each land loss indicator feature and the comprehensive weight of the land loss indicator feature, the comprehensive probability of distribution network equipment risk hazards is obtained, and the comprehensive probability is analyzed to obtain the research and judgment results of distribution network equipment risk hazards.

[0013] Preferably, the method further comprises performing data cleaning on sample attributes, wherein the data cleaning comprises processing missing values, outliers and standardizing data formats.

[0014] Preferably, a sample distribution diagram corresponding to the risk hidden danger of distribution network equipment and the land loss indicator characteristics is drawn, and the preliminary probability value of the risk hidden danger of distribution network equipment in each land loss indicator characteristic is calculated based on the sample distribution diagram and Bernoulli's law of large numbers, specifically:

[0015] Calculate the proportion of each distribution network equipment risk under each land loss indicator characteristic, and express it in the formula:

[0016]

[0017] In the formula, P(C i F j ) indicates the risk of distribution network equipment C i In the land loss indicator feature F j The proportion of Indicates the feature of the land loss indicator F j Risks of Downstream Distribution Network Equipment C i The number of samples, Indicates the characteristics of the land loss indicator F j The number of samples;

[0018] The proportion of each distribution network equipment risk hazard under the corresponding land loss indicator characteristic is used as the ordinate of the sample distribution diagram, and each land loss indicator characteristic is used as the abscissa, and the corresponding line graph is drawn as the distribution network equipment risk hazard-land loss indicator characteristic sample distribution diagram corresponding to each distribution network equipment risk hazard; the land loss indicator characteristic in the sample distribution diagram is divided into intervals, and the distribution of each distribution network equipment risk hazard in each interval is counted, specifically, the number of occurrences of each distribution network equipment risk hazard in each interval; the probability p of each distribution network equipment risk hazard in each interval is calculated according to the distribution.ij , expressed as:

[0019]

[0020] In the formula, k ij Indicates the number of samples of risk hazards of each distribution network equipment in each interval, n i represents the number of samples in each interval;

[0021] According to the probability p ij Calculate the overall preliminary probability value p of the risk hazards of each distribution network equipment j , expressed as:

[0022]

[0023] Where N represents the sum of all interval samples.

[0024] Preferably, an N×m land loss indicator feature matrix is ​​constructed, and the land loss indicator feature matrix is ​​standardized to obtain a standardized matrix, specifically:

[0025] The land loss characteristic index is used as a column vector, and the risk hidden danger of distribution network equipment is constructed in the land loss indicator feature to form an N×m matrix X={x ab}, where m represents the number of land loss indicator features, a = 1, 2, 3, ..., N, b = 1, 2, 3, ..., m, and the matrix X is expressed as:

[0026]

[0027] Each column vector is normalized to obtain a normalized matrix Y = {y ab}, expressed as:

[0028]

[0029] in, In the formula, max(X b ) represents the maximum value of the current column vector; min(X b ) represents the minimum value of the current column vector;

[0030] Calculate the probability matrix YP corresponding to the standardized matrix = {yp ab}, expressed as:

[0031]

[0032] Preferably, the information entropy of the land loss indicator is obtained based on the probability matrix calculation, and the initial weight corresponding to the land loss indicator is calculated according to the information entropy, specifically:

[0033] The information entropy of the land loss indicator is expressed as follows:

[0034]

[0035] In the formula, S b The information entropy representing the characteristics of the b-th land loss indicator;

[0036] The initial weight corresponding to the land loss indicator is expressed by the formula:

[0037]

[0038] In the formula, α b Represents the indicator weight value of the b-th land loss indicator.

[0039] Preferably, the characteristics of the land loss indicators are analyzed using the hierarchical analysis method (AHP) to obtain the judgment weights, and the importance of the attributes of each two data value samples is scored according to the expert scoring table to obtain the judgment matrix Q = {q ub}, specifically expressed as:

[0040]

[0041] In the formula, q ub Indicates the score value of the importance of each two data sample attributes, where q ub and q bu The two are reciprocal;

[0042] Solve the mth power of the product of each row of the matrix to obtain an m-dimensional weight vector, which can be expressed as:

[0043]

[0044] Perform normalization to obtain the judgment weight It is expressed as:

[0045]

[0046] Preferably, the comprehensive weight of the land loss indicator characteristics is obtained by combining the initial weight and the judgment weight, which is expressed as:

[0047]

[0048] k 1 +k 2 =1;

[0049] In the formula, k 1 Indicates the degree to which the pre-set comprehensive weights tend to the entropy weight method; k 2 It indicates the degree to which the pre-set comprehensive weights tend to favor the analytic hierarchy process.

[0050] A distribution network equipment risk hazard analysis system based on short-term ground loss signals, the system includes a data acquisition module, a probability calculation module, a weight calculation module and a cause analysis module, specifically:

[0051] The data acquisition module is used to acquire historical short-term land loss signals, and use regularized expressions to extract key fields of the historical short-term land loss signals as sample attributes, wherein the sample attributes include distribution network equipment risk hazards and land loss indicators, and the land loss indicators include the number of land loss times, the duration of land loss, and the area of ​​land loss;

[0052] The probability calculation module is used to extract the characteristics of the land loss indicator to obtain the land loss indicator characteristics; draw a sample distribution map corresponding to the distribution network equipment risk hidden danger-land loss indicator characteristics, and calculate the preliminary probability value of the distribution network equipment risk hidden danger in each land loss indicator characteristic based on the sample distribution map and Bernoulli's law of large numbers;

[0053] The weight calculation module is used to construct an N×m land loss indicator feature matrix, standardize the land loss indicator feature matrix to obtain a standardized matrix; calculate the probability matrix corresponding to the standardized matrix, calculate the information entropy of the land loss indicator based on the probability matrix, and calculate the initial weight corresponding to the land loss indicator according to the information entropy; use the hierarchical analysis method AHP to analyze the land loss indicator characteristics to obtain the judgment weight; combine the initial weight and the judgment weight to obtain the comprehensive weight of the land loss indicator characteristics;

[0054] The cause analysis module is used to combine the preliminary probability value of the distribution network equipment risk hazards in each land loss indicator feature and the comprehensive weight of the land loss indicator feature to obtain the comprehensive probability of the distribution network equipment risk hazards, analyze the comprehensive probability, and obtain the analysis result of the distribution network equipment risk hazards.

[0055] An electronic device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method for analyzing and determining the risk of distribution network equipment based on short-term ground loss signals as described in any one of claims 1 to 7 is implemented.

[0056] A computer-readable storage medium having a computer program stored thereon, characterized in that when the program is executed by a processor, the method for analyzing and determining the risk hazards of distribution network equipment based on short-term ground loss signals as described in any one of claims 1 to 7 is implemented.

[0057] Compared with the prior art, the present invention has the following beneficial effects:

[0058] 1) The present invention provides a method and system for analyzing and judging the risk hazards of distribution network equipment based on short-term ground loss signals, which can effectively check the risk hazards of distribution network equipment before ground loss failures occur, and avoid the occurrence of ground loss failures. The key fields of historical short-term ground loss signals, including text fields and data fields, are extracted using regular expressions, which improves the accuracy and efficiency of data extraction and ensures the relevance and integrity of the acquired information;

[0059] 2) The present invention provides a method and system for analyzing the risk of distribution network equipment based on short-term ground loss signals. By applying Bernoulli's law of large numbers, the probability values ​​of the risk of different distribution network equipment are calculated, which improves the scientific nature of the risk analysis of distribution network equipment and provides more reliable data support.

[0060] 3) The present invention provides a method and system for analyzing and assessing the risks of distribution network equipment based on short-term ground loss signals. The matrix constructed based on data sample attributes is combined with the entropy weight method and the hierarchical analysis method to enhance the comprehensive weight calculation capability of data sample attributes. Through weighted calculation, various factors are comprehensively considered and probability values ​​are integrated to provide a scientific and objective basis for analyzing and assessing the risks of distribution network equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 is a method flow chart of an embodiment of the present invention;

[0062] Figure 2 is the sample distribution of the tree obstacle-land loss times in the embodiment of the present invention;

[0063] Figure 3 is a sample distribution of the number of times of losing land within a user in an embodiment of the present invention;

[0064] Figure 4 is the sample distribution of the number of times of bird's nest-land loss in an embodiment of the present invention;

[0065] Figure 5 is the sample distribution of the number of short circuits and ground loss of small animals in the embodiment of the present invention;

[0066] Figure 6 This is the sample distribution of the number of early discharge-ground loss of the equipment type according to the embodiment of the present invention. DETAILED DESCRIPTION

[0067] The specific implementation modes of the present invention are described below so that those skilled in the art can understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific implementation modes. For those of ordinary skill in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the attached claims, these changes are obvious, and all inventions and creations utilizing the concept of the present invention are protected.

[0068] The present invention provides the following technical solution: a method and system for analyzing and determining the risk hazards of distribution network equipment based on short-term ground loss signals.

[0069] Example 1

[0070] This embodiment provides a method for analyzing and determining the risk of distribution network equipment based on short-term ground loss signals, and the specific steps include:

[0071] S1. Obtain historical short-term land loss signals, and use regularized expressions to extract key fields of the historical short-term land loss signals as sample attributes, wherein the sample attributes include distribution network equipment risk hazards and land loss indicators, and the land loss indicators include the number of land loss times, the duration of land loss, and the area of ​​land loss;

[0072] The method further includes performing data cleaning on sample attributes, wherein the data cleaning includes processing missing values, outliers and unifying data formats;

[0073] S2. Extracting features from the land loss indicator to obtain land loss indicator features. In this embodiment, extracting features from the land loss duration sequence to calculate the maximum value, variance and sum of the land loss duration sequence.

[0074] S3, draw a sample distribution map corresponding to the risk hidden danger of distribution network equipment-land loss indicator characteristics, and calculate the preliminary probability value of the risk hidden danger of distribution network equipment in each land loss indicator characteristic based on the sample distribution map and Bernoulli's law of large numbers;

[0075] Calculate the proportion of each distribution network equipment risk under each land loss indicator characteristic, and express it in the formula:

[0076]

[0077] In the formula, P(C i F j ) indicates the risk of distribution network equipment C i In the land loss indicator feature F j The proportion of Indicates the feature of the land loss indicator F j Risks of Downstream Distribution Network Equipment C i The number of samples, Indicates the characteristics of the land loss indicator F j The number of samples;

[0078] The proportion of each distribution network equipment risk hazard under the corresponding land loss indicator characteristic is used as the ordinate of the sample distribution diagram, and each land loss indicator characteristic is used as the abscissa, and the corresponding line graph is drawn as the distribution network equipment risk hazard-land loss indicator characteristic sample distribution diagram corresponding to each distribution network equipment risk hazard; the land loss indicator characteristic in the sample distribution diagram is divided into intervals, and the distribution of each distribution network equipment risk hazard in each interval is counted, specifically, the number of occurrences of each distribution network equipment risk hazard in each interval; the probability p of each distribution network equipment risk hazard in each interval is calculated according to the distribution. ij , expressed as:

[0079]

[0080] In the formula, k ij Indicates the number of samples of risk hazards of each distribution network equipment in each interval, n i represents the number of samples in each interval;

[0081] According to the probability p ij Calculate the overall preliminary probability value p of the risk hazards of each distribution network equipment j , expressed as:

[0082]

[0083] In the formula, N represents the sum of all interval samples;

[0084] In this embodiment, the distribution network equipment risk hazards include: "tree barriers", "user interior", "bird nests", "small animal short circuits" and "early discharge of equipment"; draw a sample distribution map, tree barriers - the number of times of loss of land sample distribution, such as Figure 2 As described above; the sample distribution of the number of times the user loses land is as follows: Figure 3 As described above; the sample distribution of the number of bird nests and land loss is as follows: Figure 4 As described above; the sample distribution of the number of short circuits and ground loss of small animals, such as Figure 5 As described above; the sample distribution of the number of early discharge-ground loss of equipment, such as Figure 6 as stated;

[0085] When the risk hazard of distribution network equipment is "tree obstacle", the initial probability value p of the risk hazard of distribution network equipment corresponding to the number of ground loss, the maximum duration of ground loss, the variance of the duration of ground loss and the sum of the duration of ground loss is obtained. 1 、p 2 、p 3 and p 4 ; and the initial probability value p of the lost area 5 ;

[0086] S4, constructing an N×m land loss indicator characteristic matrix, and performing standardization processing on the land loss indicator characteristic matrix to obtain a standardized matrix;

[0087] S41. Take the land loss characteristic index as a column vector and construct an N×m matrix X={x ab}, where m represents the number of land loss indicator features, a = 1, 2, 3, ..., N, b = 1, 2, 3, ..., m, and the matrix X is expressed as:

[0088]

[0089] Each column vector is normalized to obtain a normalized matrix Y = {y ab}, expressed as:

[0090]

[0091] in, In the formula, max(X b ) represents the maximum value of the current column vector; min(X b ) represents the minimum value of the current column vector;

[0092] Calculate the probability matrix YP corresponding to the standardized matrix = {yp ab}, expressed as:

[0093]

[0094] In this embodiment, the number of times of losing land, the maximum value of the duration of losing land, the variance of the duration of losing land and the sum of the duration of losing land are taken as column vectors to construct an N×4 matrix X={x ab}, where a=1,2,3,...,N, b=1,2,3,4, the matrix X is expressed as:

[0095]

[0096] Each column vector is normalized as follows:

[0097]

[0098] In the formula, max(X b ) represents the maximum value of the current column vector; min(X b ) represents the minimum value of the current column vector;

[0099] Get the standardized matrix Y = {y ab}, expressed as:

[0100]

[0101] By formula The probability matrix YP is calculated to be ab}, expressed as:

[0102]

[0103] S42, calculating a probability matrix corresponding to the standardized matrix, obtaining information entropy of the land loss indicator based on the probability matrix, and calculating an initial weight corresponding to the land loss indicator according to the information entropy;

[0104] The information entropy of the land loss indicator is expressed as follows:

[0105]

[0106] In the formula, S b The information entropy representing the characteristics of the b-th land loss indicator;

[0107] The initial weight corresponding to the land loss indicator is expressed by the formula:

[0108]

[0109] In the formula, α b represents the indicator weight value of the b-th land loss indicator;

[0110] In this embodiment, the information entropy of the land loss indicator is expressed as follows:

[0111]

[0112] In the formula, S b The information entropy representing the characteristics of the b-th land loss indicator;

[0113] The initial weight corresponding to the land loss indicator is expressed by the formula:

[0114]

[0115] In the formula, α b represents the indicator weight value of the b-th land loss indicator;

[0116] S5. Use the analytic hierarchy process (AHP) to analyze the characteristics of the land loss indicators and obtain the judgment weights. According to the expert scoring table, please refer to Table 1 for the scoring rules, score the importance of each two data value sample attributes to obtain the judgment matrix Q = {q ub}, specifically expressed as:

[0117]

[0118] In the formula, q ub Indicates the score value of the importance of each two data sample attributes, where qub and q bu The two are reciprocal;

[0119] Table 1 Expert scoring table

[0120]

[0121] Solve the mth power of the product of each row of the matrix to obtain an m-dimensional weight vector, which can be expressed as:

[0122]

[0123] Perform normalization to obtain the judgment weight It is expressed as:

[0124]

[0125] In this embodiment, since m=4, the judgment matrix Q={q ub}, expressed as:

[0126]

[0127] In the formula, q ub Indicates the score value of the importance of each two data sample attributes, where q ub and q bu The two are reciprocal;

[0128] Solve the fourth power of the product of each row of the matrix to obtain a 4-dimensional weight vector, which can be expressed as:

[0129]

[0130] Normalize the data to get the index weight value of the hierarchical analysis method It is expressed as:

[0131]

[0132] S6. Combine the initial weight and the judgment weight to obtain the comprehensive weight of the land loss indicator characteristics; the formula is:

[0133]

[0134] k 1 +k 2 =1;

[0135] In the formula, k 1 Indicates the degree to which the pre-set comprehensive weights tend to the entropy weight method; k 2 It indicates the degree to which the pre-set comprehensive weights tend to favor the analytic hierarchy process.

[0136] In this embodiment, the comprehensive weight of the number of times of losing land is ω 1 , the comprehensive weight of the maximum duration of land loss is ω 2 The comprehensive weight of the variance of the duration of land loss is ω 3 The combined weight of the sum of the duration of the loss of land is ω 4 ; and set the weight value ω according to the importance of the lost area 5 ;

[0137] S7. Combining the preliminary probability values ​​of the distribution network equipment risk hazards in each land loss indicator feature and the comprehensive weight of the land loss indicator feature, obtaining the comprehensive probability of the distribution network equipment risk hazards, analyzing the comprehensive probability, and obtaining the research and judgment results of the distribution network equipment risk hazards;

[0138] In this embodiment, the final probability P when the risk hidden danger of the distribution network equipment is "tree obstacle" is calculated and expressed as:

[0139] P=ω 1 p 1 +ω 2 p 2 +ω 3 p 3 +ω 4 p 4 +ω 5 p 5 .

[0140] Example 2

[0141] This embodiment provides a distribution network equipment risk hazard analysis system based on short-term ground loss signals, the system includes a data acquisition module, a probability calculation module, a weight calculation module and a cause analysis module, specifically:

[0142] The data acquisition module is used to acquire historical short-term land loss signals, and use regularized expressions to extract key fields of the historical short-term land loss signals as sample attributes, wherein the sample attributes include distribution network equipment risk hazards and land loss indicators, and the land loss indicators include the number of land loss times, the duration of land loss, and the area of ​​land loss;

[0143] The probability calculation module is used to extract the characteristics of the land loss indicator to obtain the land loss indicator characteristics; draw a sample distribution map corresponding to the distribution network equipment risk hidden danger-land loss indicator characteristics, and calculate the preliminary probability value of the distribution network equipment risk hidden danger in each land loss indicator characteristic based on the sample distribution map and Bernoulli's law of large numbers;

[0144] The weight calculation module is used to construct an N×m land loss indicator feature matrix, standardize the land loss indicator feature matrix to obtain a standardized matrix; calculate the probability matrix corresponding to the standardized matrix, calculate the information entropy of the land loss indicator based on the probability matrix, and calculate the initial weight corresponding to the land loss indicator according to the information entropy; use the hierarchical analysis method AHP to analyze the land loss indicator characteristics to obtain the judgment weight; combine the initial weight and the judgment weight to obtain the comprehensive weight of the land loss indicator characteristics;

[0145] The cause analysis module is used to combine the preliminary probability value of the distribution network equipment risk hazards in each land loss indicator feature and the comprehensive weight of the land loss indicator feature to obtain the comprehensive probability of the distribution network equipment risk hazards, analyze the comprehensive probability, and obtain the analysis result of the distribution network equipment risk hazards.

[0146] Example 3

[0147] This embodiment provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, a method for analyzing and determining risk hazards of distribution network equipment based on short-term ground loss signals as described in any one of Embodiment 1 is implemented.

[0148] Example 4

[0149] This embodiment provides a computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, the method for analyzing and determining the risk of distribution network equipment based on a short-term ground loss signal as described in any one of Embodiments 1 is implemented.

[0150] It is worth noting that the system, electronic device and computer-readable storage medium described in the present invention are based on the same principles as the method described in Example 1, and will not be described in detail here.

[0151] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for analyzing and judging the risk of distribution network equipment based on short-term ground loss signals, characterized in that: The method comprises: Obtain historical short-term land loss signals, and use regularized expressions to extract key fields of the historical short-term land loss signals as sample attributes, wherein the sample attributes include risk hazards of distribution network equipment and land loss indicators, and the land loss indicators include the number of land losses, the duration of land loss, and the area of ​​land loss; Extracting features of the land loss indicator to obtain land loss indicator features; drawing a sample distribution map of distribution network equipment risk hazards-land loss indicator features, and calculating and obtaining preliminary probability values ​​of distribution network equipment risk hazards in each land loss indicator feature based on the sample distribution map and Bernoulli's law of large numbers; Constructing an N×m land loss indicator feature matrix, performing standardization processing on the land loss indicator feature matrix to obtain a standardized matrix; calculating a probability matrix corresponding to the standardized matrix, obtaining information entropy of the land loss indicator based on the probability matrix, and calculating an initial weight corresponding to the land loss indicator according to the information entropy; The analytic hierarchy process (AHP) was used to analyze the characteristics of land loss indicators to obtain the judgment weights; the initial weights and judgment weights were combined to obtain the comprehensive weights of the characteristics of land loss indicators; By combining the preliminary probability values ​​of distribution network equipment risk hazards in each land loss indicator feature and the comprehensive weight of the land loss indicator feature, the comprehensive probability of distribution network equipment risk hazards is obtained, and the comprehensive probability is analyzed to obtain the research and judgment results of distribution network equipment risk hazards.

2. The method for analyzing and determining the risk of distribution network equipment based on short-term ground loss signals according to claim 1 is characterized in that: The method also includes performing data cleaning on sample attributes, wherein the data cleaning includes processing missing values, outliers and standardizing data formats.

3. The method for analyzing and determining the risk of distribution network equipment based on short-term ground loss signals according to claim 2 is characterized in that: Draw a sample distribution map corresponding to the risk hidden danger of distribution network equipment-land loss indicator characteristics, and calculate the preliminary probability value of the risk hidden danger of distribution network equipment in each land loss indicator characteristic based on the sample distribution map and Bernoulli's law of large numbers, specifically: Calculate the proportion of each distribution network equipment risk under each land loss indicator characteristic, and express it in the formula: In the formula, P(C i F j ) indicates the risk of distribution network equipment C i In the land loss indicator feature F j The proportion of Indicates the feature of the land loss indicator F j Risks of Downstream Distribution Network Equipment C i The number of samples, Indicates the characteristics of the land loss indicator F j The number of samples; The proportion of each distribution network equipment risk hazard under the corresponding land loss indicator characteristic is used as the ordinate of the sample distribution diagram, and each land loss indicator characteristic is used as the abscissa, and the corresponding line graph is drawn as the distribution network equipment risk hazard-land loss indicator characteristic sample distribution diagram corresponding to each distribution network equipment risk hazard; the land loss indicator characteristic in the sample distribution diagram is divided into intervals, and the distribution of each distribution network equipment risk hazard in each interval is counted, specifically, the number of occurrences of each distribution network equipment risk hazard in each interval; the probability p of each distribution network equipment risk hazard in each interval is calculated according to the distribution. ij , expressed as: In the formula, k ij Indicates the number of samples of risk hazards of each distribution network equipment in each interval, n i Indicates the number of samples in each interval; According to the probability p ij Calculate the overall preliminary probability value p of the risk hazards of each distribution network equipment j , expressed as: Where N represents the sum of all interval samples.

4. The method for analyzing and determining the risk of distribution network equipment based on short-term ground loss signals according to claim 3 is characterized in that: Construct an N×m land loss indicator feature matrix, and perform standardization on the land loss indicator feature matrix to obtain a standardized matrix, specifically: Taking the land loss characteristic index as a column vector, construct an N×m matrix X={x ab }, where m represents the number of land loss indicator features, a = 1, 2, 3, ..., N, b = 1, 2, 3, ..., m, and the matrix X is expressed as: Each column vector is normalized to obtain a normalized matrix Y = {y ab }, expressed as: in, In the formula, max(X b ) represents the maximum value of the current column vector; min(X b ) represents the minimum value of the current column vector; Calculate the probability matrix YP corresponding to the standardized matrix = {yp ab }, expressed as:

5. The method for analyzing and determining the risk of distribution network equipment based on short-term ground loss signals according to claim 4 is characterized in that: The information entropy of the land loss indicator is calculated based on the probability matrix, and the initial weight corresponding to the land loss indicator is calculated according to the information entropy, specifically: The information entropy of the land loss indicator is expressed as follows: In the formula, S b The information entropy representing the characteristics of the b-th land loss indicator; The initial weight corresponding to the land loss indicator is expressed by the formula: In the formula, α b Represents the indicator weight value of the b-th land loss indicator.

6. The method for analyzing and determining the risk of distribution network equipment based on short-term ground loss signals according to claim 5 is characterized in that: The analytic hierarchy process (AHP) is used to analyze the characteristics of land loss indicators to obtain judgment weights. The importance of each two data value sample attributes is scored according to the expert scoring table, and the judgment matrix Q = {q ub }, specifically expressed as: In the formula, q ub Indicates the score value of the importance of each two data sample attributes, where q ub and q bu The two are reciprocal; Solve the mth power of the product of each row of the matrix to obtain an m-dimensional weight vector, which can be expressed as: Perform normalization to obtain the judgment weight It is expressed as:

7. The method for analyzing and determining the risk of distribution network equipment based on short-term ground loss signals according to claim 6 is characterized in that: The comprehensive weight of the land loss indicator characteristics is obtained by combining the initial weight and the judgment weight, which is expressed as follows: k1+k2=1; Wherein, k1 indicates the degree to which the pre-set comprehensive weight tends to the entropy weight method; k2 indicates the degree to which the pre-set comprehensive weight tends to the analytic hierarchy process.

8. A distribution network equipment risk assessment system based on short-term ground loss signals, characterized in that: The system includes a data acquisition module, a probability calculation module, a weight calculation module and a cause analysis module, specifically: The data acquisition module is used to acquire historical short-term land loss signals, and use regularized expressions to extract key fields of the historical short-term land loss signals as sample attributes, wherein the sample attributes include distribution network equipment risk hazards and land loss indicators, and the land loss indicators include the number of land loss times, the duration of land loss, and the area of ​​land loss; The probability calculation module is used to extract features of the land loss indicator to obtain features of the land loss indicator; Draw a sample distribution map corresponding to the risk hidden danger of distribution network equipment-land loss indicator characteristics, and calculate the preliminary probability value of the risk hidden danger of distribution network equipment in each land loss indicator characteristic based on the sample distribution map and Bernoulli's law of large numbers; The weight calculation module is used to construct an N×m land loss indicator feature matrix, and perform standardization processing on the land loss indicator feature matrix to obtain a standardized matrix; Calculate the probability matrix corresponding to the standardized matrix, calculate the information entropy of the land loss indicator based on the probability matrix, and calculate the initial weight corresponding to the land loss indicator according to the information entropy; The analytic hierarchy process (AHP) was used to analyze the characteristics of land loss indicators to obtain the judgment weights; the initial weights and judgment weights were combined to obtain the comprehensive weights of the characteristics of land loss indicators; The cause analysis module is used to combine the preliminary probability value of the distribution network equipment risk hazards in each land loss indicator feature and the comprehensive weight of the land loss indicator feature to obtain the comprehensive probability of the distribution network equipment risk hazards, analyze the comprehensive probability, and obtain the analysis result of the distribution network equipment risk hazards.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the method for analyzing and determining the risk of distribution network equipment based on short-term ground loss signals as described in any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method for analyzing and determining the risk of distribution network equipment based on short-term ground loss signals as described in any one of claims 1 to 7 is implemented.

Images