Power distribution network fault risk level prediction method and system in extreme weather

Abstract

The invention is suitable for the technical field of fault risk prediction, and provides a power distribution network fault risk grade prediction method and system in extreme weather, and the method comprises the steps: collecting the extreme weather information data of the operation of a power distribution network within a preset time; screening an environment feature vector based on the extreme weather information data, and converting a one-dimensional array corresponding to the environment feature vector into a two-dimensional matrix array; calculating a mahalanobis distance between the environment feature vector in the current time period and a power distribution network environment feature vector in a pre-constructed association rule base based on the two-dimensional matrix array; judging whether a mahalanobis distance between the environment feature vector and a power distribution network environment feature vector in a pre-constructed association rule base is within a threshold range or not; and if yes, determining a fault risk level corresponding to the environment feature vector based on the association rule base. According to the invention, the calculation process is simple, the fault risk under the influence of extreme weather is considered, and the comprehensiveness and accuracy of fault risk prediction are effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of fault risk prediction, and in particular to a method and system for predicting the fault risk level of a distribution network in extreme weather. Background Art

[0002] The distribution network has a complex topology, a large number of equipment types, and the faults caused by external environmental factors are highly random and have complex mechanisms. Conventional electrical mechanism analysis and probability statistics are difficult to play an effective role. Distribution network failures are prone to occur under severe weather conditions. Therefore, weather factors are increasingly becoming an important factor affecting distribution network failures. However, due to the irregularity, suddenness, and diversity of weather changes, distribution network failures caused by weather factors are not easy to be effectively monitored.

[0003] In the prior art, there are several common methods for predicting fault risks in extreme weather. One is to use historical power outage data and weather data to establish an empirical model, and evaluate the effectiveness of the prediction model through actual extreme weather data; another method is to use linear regression, exponential regression, linear multivariate regression and artificial neural network models based on characteristics such as temperature, wind speed and the average annual number of pruning times of vegetation near the feeder to predict distribution network failures caused by vegetation. The above methods are more comprehensive in predicting distribution network failures, but the prediction accuracy for risk levels is relatively low, which easily leads to the problem that the prediction results are not comprehensive enough. Summary of the invention

[0004] The embodiments of the present application provide a method and system for predicting the risk level of distribution network failures in extreme weather, which are used to solve the problem that the prediction results are not comprehensive enough.

[0005] A first aspect of an embodiment of the present application provides a method for predicting the risk level of distribution network failure in extreme weather, comprising:

[0006] Collect extreme weather information data of distribution network operation within a preset time;

[0007] Filtering environmental feature vectors based on the extreme weather information data, and converting the one-dimensional array corresponding to the environmental feature vectors into a two-dimensional matrix array;

[0008] Calculating the Mahalanobis distance between the environmental feature vector in the current time period and the distribution network environmental feature vector in the pre-built association rule library based on the two-dimensional matrix array;

[0009] Determine whether the Mahalanobis distance between the environmental feature vector and the distribution network environmental feature vector in the pre-built association rule library is within a threshold range;

[0010] If so, the fault risk level corresponding to the environmental feature vector is determined based on the association rule base.

[0011] Furthermore, the calculating, based on the two-dimensional matrix array, the Mahalanobis distance between the environmental feature vector in the current time period and the distribution network environmental feature vector in the pre-built association rule library includes:

[0012] Constructing an initial association rule base, wherein the initial association rule base includes a distribution network fault feature vector and an environment feature vector of the distribution network under extreme weather conditions;

[0013] Setting effective rules and invalid rules, wherein the effective rules are the ratio of the average annual occurrence probability of distribution network failures in extreme weather environments to the average annual occurrence probability of distribution network failures in normal weather environments;

[0014] The enhancement index and the leverage ratio index are introduced to further screen the effective rules to obtain the target association rule base.

[0015] Furthermore, the expression of the effective rule base is as follows:

[0016]

[0017]

[0018] Where: μ x is the average annual probability of occurrence of fault condition Y under environmental condition X, μ is the average annual probability of occurrence of fault condition Y, count(X∪Y) is the number of times the environmental condition X is met and the fault condition is Y, count(Y) is the number of times the fault condition Y occurs, D is the total number of statistical days, p(x i ) is the term x i The probability of appearing in its dimension.

[0019] Furthermore, the expressions of the lift and leverage are as follows:

[0020]

[0021] Leverage(X→Y)=support(X→Y)-support(X)·support(Y);

[0022] Where: Lifi(X→Y) is the degree of improvement, Leverage(X→Y) is the leverage ratio.

[0023] Furthermore, the calculating, based on the two-dimensional matrix array, the Mahalanobis distance between the environmental feature vector in the current time period and the distribution network environmental feature vector in the pre-built association rule library includes:

[0024] The calculation formula of Mahalanobis distance is as follows:

[0025]

[0026] Where: S mn is the mnth element in the diagonal matrix in the two-dimensional matrix array, c is the mean vector of the population A, and k is the coefficient.

[0027] Furthermore, the determining whether the Mahalanobis distance between the environmental feature vector and the distribution network environmental feature vector in the pre-built association rule library is within a threshold range includes:

[0028] If not, the Mahalanobis distance between the environmental feature vector corresponding to the extreme weather information data in the next time period and the distribution network environmental feature vector in the pre-built association rule library is calculated.

[0029] Furthermore, the determining the fault risk level corresponding to the environmental feature vector based on the association rule base includes:

[0030] Calculate the risk weight of each influencing factor according to the set environmental characteristic vector influencing factor, and determine the extreme weather risk level based on the calculated influencing factor weight;

[0031] The distribution network failure risk level is determined according to the extreme weather risk level.

[0032] A second aspect of an embodiment of the present application provides a system for predicting the risk level of distribution network failures in extreme weather, including:

[0033] An extreme weather information data collection unit is used to collect extreme weather information data of the distribution network operation within a preset time;

[0034] An array conversion unit, used for screening environmental feature vectors based on the extreme weather information data, and converting the one-dimensional array corresponding to the environmental feature vector into a two-dimensional matrix array;

[0035] A Mahalanobis distance calculation unit, used for calculating the Mahalanobis distance between the environmental feature vector in the current time period and the distribution network environmental feature vector in the pre-built association rule library based on the two-dimensional matrix array;

[0036] A Mahalanobis distance judgment unit, used to judge whether the Mahalanobis distance between the environmental feature vector and the distribution network environmental feature vector in the pre-built association rule library is within a threshold range;

[0037] The fault risk level determination unit is used to determine the fault risk level corresponding to the environmental feature vector based on the association rule base when the Mahalanobis distance between the environmental feature vector and the distribution network environmental feature vector in the pre-built association rule base is within a threshold range.

[0038] A third aspect of an embodiment of the present application provides a computer device, including a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, wherein when the processor executes the computer-readable instructions, the steps of the method for predicting the risk level of distribution network failure in extreme weather as described in any one of the above items are implemented.

[0039] A fourth aspect of an embodiment of the present application provides a readable storage medium, which stores computer-readable instructions. When the computer-readable instructions are executed by a processor, the steps of the method for predicting the risk level of distribution network failure in extreme weather as described in any one of the above are implemented.

[0040] It can be seen from the above technical solutions that the embodiments of the present application have the following advantages:

[0041] This application first collects extreme weather data of power grid operation within a preset time, screens environmental feature vectors, and converts the one-dimensional array corresponding to the environmental feature vector into a two-dimensional matrix array, and then calculates the Mahalanobis distance between the environmental feature vector at the current moment and the environmental feature vector in the pre-built association rule library, and determines whether the calculated Mahalanobis distance is within the threshold range. If so, the fault risk level corresponding to the environmental feature vector is determined based on the association rule library. By calculating the value of the Mahalanobis distance to compare environmental characteristics, it is determined whether a distribution network fault occurs based on the acquired weather data, and the risk level of the fault is further determined. The calculation process is simple and takes into account the fault risk under the influence of extreme weather, effectively improving the comprehensiveness and accuracy of fault risk prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a schematic diagram of an embodiment of a method for predicting the risk level of distribution network failure in extreme weather in the present invention;

[0043] Figure 2 It is a structural block diagram of an embodiment of an extreme weather distribution network fault risk level prediction system in the present invention;

[0044] Figure 3 The figure is a structural block diagram of a computer device in the present invention. DETAILED DESCRIPTION

[0045] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0046] In this embodiment, the method for predicting the risk level of distribution network failure in extreme weather is used to improve the comprehensiveness and accuracy of failure risk prediction. In this embodiment, the implementation method can be implemented in the system, can be implemented in the server, and can also be implemented in the terminal, without specific limitation.

[0047] Embodiment 1

[0048] See also Figure 1 , an embodiment of a method for predicting the risk level of distribution network failure in extreme weather in the present invention comprises the following steps:

[0049] S11. Collect extreme weather information data of distribution network operation within a preset time;

[0050] In this embodiment, a certain time period is set, such as hours or days, and the weather forecast related system is connected to collect extreme weather information data within the set time period, such as strong winds, precipitation, lightning, heavy snow and other extreme weather information.

[0051] S12. Filtering environmental feature vectors based on extreme weather data, and converting the one-dimensional array corresponding to the environmental feature vectors into a two-dimensional matrix array;

[0052] After the weather information data is collected as mentioned above, since most of the weather information data is of heterogeneous data types, it is necessary to convert the unstructured or semi-structured data into structured data, and then preprocess the data based on the structured data, such as data cleaning and elimination, and then use the model learning training method to learn and train based on the preprocessed data to extract the environmental feature vectors of the weather information data in the operation of the distribution network, where the environmental feature vectors are divided according to the type of extreme weather. In general, the feature vector corresponding to the data form obtained after model training is a one-dimensional array, so it is necessary to convert the one-dimensional array into a two-dimensional matrix array. Converting it into a two-dimensional matrix array can improve the data accuracy. So that the subsequent Mahalanobis distance calculation can adapt to the calculation logic of the algorithm and the preprocessing of the data.

[0053] S13. Calculate the Mahalanobis distance between the environmental feature vector in the current time period and the distribution network environmental feature vector in the pre-built association rule library based on the two-dimensional matrix array;

[0054] Specifically, the association rule base also includes the following steps:

[0055] 1. Construct an initial association rule base, which includes distribution network fault feature vectors and distribution network environment feature vectors under extreme weather conditions;

[0056] 2. Set effective rules and invalid rules. The effective rule is the ratio of the average annual probability of distribution network failure in extreme weather conditions to the average annual probability of distribution network failure in normal weather conditions.

[0057] 3. Introduce the improvement index and leverage ratio index to further screen the effective rules and obtain the target association rule base.

[0058] By setting up an association rule base, the environmental characteristic factors and fault risk characteristic information can be associated and quantified. The rule base reflects the association rules of the fault risk characteristics corresponding to the influencing factors in the environmental characteristic information. Here, the influencing factors in the environmental characteristic information only consider significant influencing factors. For example, in extreme weather typhoons, only influencing factors with strong correlations such as wind speed and precipitation are considered. The specific influencing factors can be set by calculating the correlation.

[0059] The expressions for valid rule bases are as follows:

[0060]

[0061]

[0062] Where: μ x is the average annual probability of occurrence of fault condition Y under environmental condition X, μ is the average annual probability of occurrence of fault condition Y, count(X∪Y) is the number of times the environmental condition X is met and the fault condition is Y, count(Y) is the number of times the fault condition Y occurs, D is the total number of statistical days, p(x i ) is the term x i The probability of appearing in its dimension.

[0063] The expressions for lift and leverage are as follows:

[0064]

[0065] Leverage(X→Y)=support(X→Y)-support(X)·support(Y);

[0066] Where: Lifi(X→Y) is the degree of improvement, Leverage(X→Y) is the leverage ratio.

[0067] In addition, calculating the Mahalanobis distance also includes the following steps:

[0068] 1. Transpose the two-dimensional matrix array and extract the eigenvectors after the matrix transposition to form a diagonal matrix;

[0069] 2. Based on the formed diagonal matrix, the Mahalanobis distance of the capacitance and voltage of each submodule at the current moment from the overall distribution is calculated.

[0070] Set the data matrix to an m×n matrix, transpose the matrix, and obtain the transposed matrix. For example, if A represents the matrix, then the transposed matrix is ​​A. T , if there is the following:

[0071] AA T =PΛ1P T ; A T A=QΛ2Q T ; then A=P∑Q T ; Where P = (p1, p2, p2, ..., p m ), and p1,p2,p2,…,p m AA T The characteristic vector of m ) is A T The eigenvector of A, ∑ is an m×n matrix, and the eigenvalues ​​of the elements on the diagonal arranged from large to small are taken as singular values. Based on the above, the expression of the improved Mahalanobis distance is as follows:

[0072]

[0073] Where: S mn is the mnth element in the diagonal matrix in the two-dimensional matrix array, c is the mean vector of the population A, and k is the coefficient.

[0074] By calculating the Mahalanobis distance between the environmental feature vector and the environmental feature vector in the set association rule library, it is possible to determine whether a distribution network fault occurs based on the extreme weather information data. The fault feature vector corresponding to the environmental feature vector is set in the association rule library, and the fault events that may occur due to the extreme weather information can also be inferred.

[0075] S14. Determine whether the Mahalanobis distance between the environmental feature vector and the distribution network environmental feature vector in the pre-built association rule library is within a threshold range;

[0076] S15. When the Mahalanobis distance between the environment feature vector and the distribution network environment feature vector in the pre-built association rule library is within a threshold range, the fault risk level corresponding to the environment feature vector is determined based on the association rule library.

[0077] S16. When the Mahalanobis distance between the environmental feature vector and the distribution network environmental feature vector in the pre-built association rule library is not within the threshold range, the Mahalanobis distance between the environmental feature vector corresponding to the extreme weather data in the next time period and the distribution network environmental feature vector in the pre-built association rule library is calculated.

[0078] When the Mahalanobis distance between the environment feature vector and the distribution network environment feature vector in the pre-built association rule library is within the threshold range, determining the fault risk level based on the set association rule library includes the following steps:

[0079] 1. Calculate the risk weight of each influencing factor according to the set environmental characteristic vector influencing factor, and determine the extreme weather risk level based on the calculated influencing factor weight;

[0080] 2. Determine the distribution network failure risk level based on the extreme weather risk level.

[0081] Specifically, the weight is calculated according to the distribution of the influencing factors in the environmental feature vector, and the weight score of each influencing factor is calculated by combining subjective scoring and objective weighting methods. Then, the confidence is calculated to quantify the correlation between each influencing factor and the fault risk level. The higher the confidence, the greater the possibility of the occurrence of the fault risk level event when the influencing factor appears. The fault risk level is generally set to five levels, and the level is assigned according to the weight score of each influencing factor. The level corresponding to the environmental feature vector is assigned according to the total weight of all influencing factors.

[0082] The above embodiment quantifies the correlation between the environmental feature vector and the fault feature information by setting up an association rule base, and then determines whether there is a fault risk based on the Mahalanobis distance between the environmental feature vector and the environmental feature vector in the rule base. Further analysis can determine the fault risk level and improve the comprehensiveness of fault risk prediction.

[0083] Embodiment 2

[0084] See also Figure 2 , an embodiment of an extreme weather distribution network fault risk level prediction system of the present invention comprises the following steps:

[0085] The extreme weather information data collection unit 101 is used to collect extreme weather information data of the distribution network operation within a preset time;

[0086] An array conversion unit 102 is used to filter the environmental feature vector based on the extreme weather information data, and convert the one-dimensional array corresponding to the environmental feature vector into a two-dimensional matrix array;

[0087] A Mahalanobis distance calculation unit 103, used for calculating the Mahalanobis distance between the environment feature vector in the current time period and the distribution network environment feature vector in the pre-built association rule library based on the two-dimensional matrix array;

[0088] A Mahalanobis distance judgment unit 104 is used to judge whether the Mahalanobis distance between the environment feature vector and the distribution network environment feature vector in the pre-built association rule library is within a threshold range;

[0089] The fault risk level determination unit 105 is used to determine the fault risk level corresponding to the environment feature vector based on the association rule base when the Mahalanobis distance between the environment feature vector and the distribution network environment feature vector in the pre-built association rule base is within a threshold range.

[0090] For the specific definition of the prediction system, please refer to the definition of the prediction method above, which will not be repeated here. Each module in the above prediction system can be implemented in whole or in part by software, hardware and a combination thereof. Each of the above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

[0091] Embodiment 3

[0092] See also Figure 3 The present invention provides a computer device, including a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor. When the processor executes the computer-readable instructions, the steps of the above method are implemented.

[0093] Those of ordinary skill in the art will appreciate that the units of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition of each example has been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0094] In the embodiments provided by the present invention, it should be understood that the division of units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units can be combined into one unit, one unit can be split into multiple units, or some features can be ignored. In addition, each functional unit in each embodiment of the present invention can be integrated into a processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units.

[0095] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-0nlyMemory), random access memory (RAM, RandomAccessMemory), mobile hard disk, magnetic disk or optical disk, etc., which can store program code.

[0096] It can be understood that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and specification of the present invention.

Claims

1. A method for predicting the risk level of distribution network failure in extreme weather, characterized in that: include: Collect extreme weather information data of distribution network operation within a preset time; Filtering environmental feature vectors based on the extreme weather information data, and converting the one-dimensional array corresponding to the environmental feature vectors into a two-dimensional matrix array; Calculating the Mahalanobis distance between the environmental feature vector in the current time period and the distribution network environmental feature vector in the pre-built association rule library based on the two-dimensional matrix array; Determine whether the Mahalanobis distance between the environmental feature vector and the distribution network environmental feature vector in the pre-built association rule library is within a threshold range; If so, the fault risk level corresponding to the environmental feature vector is determined based on the association rule base.

2. The method for predicting the risk level of distribution network failure in extreme weather according to claim 1 is characterized in that: The calculating, based on the two-dimensional matrix array, the Mahalanobis distance between the environmental feature vector in the current time period and the distribution network environmental feature vector in the pre-built association rule library comprises: Constructing an initial association rule base, wherein the initial association rule base includes a distribution network fault feature vector and an environment feature vector of the distribution network under extreme weather conditions; Setting effective rules and invalid rules, wherein the effective rules are the ratio of the average annual occurrence probability of distribution network failures in extreme weather environments to the average annual occurrence probability of distribution network failures in normal weather environments; The enhancement index and the leverage ratio index are introduced to further screen the effective rules to obtain the target association rule base.

3. The method for predicting the risk level of distribution network failure in extreme weather according to claim 2 is characterized in that: The expression of the effective rule base is as follows: Where: μ x is the average annual probability of occurrence of fault condition Y under environmental condition X, μ is the average annual probability of occurrence of fault condition Y, count(X∪Y) is the number of times the environmental condition X is met and the fault condition is Y, count(Y) is the number of times the fault condition Y occurs, D is the total number of statistical days, p(x i ) is the term x i The probability of appearing in its dimension.

4. The method for predicting the risk level of distribution network failure in extreme weather according to claim 3 is characterized in that: The expressions for the lift and leverage are as follows: Leverage(X→Y)=support(X→Y)-support(X)·support(Y); Where: Lifi(X→Y) is the degree of improvement, Leverage(X→Y) is the leverage ratio.

5. The method for predicting the risk level of distribution network failure in extreme weather according to claim 4 is characterized in that: The calculating, based on the two-dimensional matrix array, the Mahalanobis distance between the environmental feature vector in the current time period and the distribution network environmental feature vector in the pre-built association rule library comprises: The calculation formula of Mahalanobis distance is as follows: Where: S mn is the mnth element in the diagonal matrix in the two-dimensional matrix array, c is the mean vector of the population A, and k is the coefficient.

6. The method for predicting the risk level of distribution network failure in extreme weather according to claim 1 is characterized in that: The determining whether the Mahalanobis distance between the environmental feature vector and the distribution network environmental feature vector in the pre-built association rule library is within a threshold range includes: If not, the Mahalanobis distance between the environmental feature vector corresponding to the extreme weather information data in the next time period and the distribution network environmental feature vector in the pre-built association rule library is calculated.

7. The method for predicting the risk level of distribution network failure in extreme weather according to claim 1 is characterized in that: The determining the fault risk level corresponding to the environmental feature vector based on the association rule base includes: Calculate the risk weight of each influencing factor according to the set environmental characteristic vector influencing factor, and determine the extreme weather risk level based on the calculated influencing factor weight; The distribution network failure risk level is determined according to the extreme weather risk level.

8. A system for predicting the risk level of distribution network failure in extreme weather, characterized in that: include: An extreme weather information data collection unit is used to collect extreme weather information data of the distribution network operation within a preset time; An array conversion unit, used for screening environmental feature vectors based on the extreme weather information data, and converting the one-dimensional array corresponding to the environmental feature vector into a two-dimensional matrix array; A Mahalanobis distance calculation unit, used for calculating the Mahalanobis distance between the environmental feature vector in the current time period and the distribution network environmental feature vector in the pre-built association rule library based on the two-dimensional matrix array; A Mahalanobis distance judgment unit, used to judge whether the Mahalanobis distance between the environmental feature vector and the distribution network environmental feature vector in the pre-built association rule library is within a threshold range; The fault risk level determination unit is used to determine the fault risk level corresponding to the environmental feature vector based on the association rule base when the Mahalanobis distance between the environmental feature vector and the distribution network environmental feature vector in the pre-built association rule base is within a threshold range.

9. A computer device comprising a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, characterized in that: When the processor executes the computer-readable instructions, the steps of the method for predicting the risk level of distribution network failure in extreme weather as described in any one of claims 1 to 7 are implemented.

10. A readable storage medium storing computer-readable instructions, characterized in that: When the computer-readable instructions are executed by a processor, the steps of the method for predicting the risk level of distribution network failure in extreme weather as described in any one of claims 1 to 7 are implemented.

Images