Rural power distribution network problem treatment measure decision-making method and system

Abstract

The invention discloses a rural power distribution network problem treatment measure decision-making method. The method comprises the steps of obtaining working data information and problem data information of a target rural power distribution network; determining a plurality of typical application scenes and correlation indexes of the typical application scenes; performing clustering analysis on the data information of the typical application scene to obtain a subdivided scene of the typical application scene; determining a boundary condition corresponding to the typical application scene; and making a decision of rural power distribution network problem treatment measures according to the boundary conditions. The invention also discloses a system for realizing the rural power distribution network problem treatment measure decision-making method. Through selection, subdivision, clustering and decision making of typical scenes related to rural power distribution network problems, automatic decision making of rural power distribution network problem treatment measures is achieved, reliability is higher, accuracy is better, and the method is more scientific and reasonable.

Description

Technical Field

[0001] The present invention belongs to the field of electrical automation, and in particular relates to a rural distribution network problem management measure decision-making method and system. Background Art

[0002] With the development of economy and technology and the improvement of people's living standards, electricity has become an indispensable secondary energy source in people's production and life, bringing endless convenience to people's production and life. Therefore, ensuring the stable and reliable supply of electricity has become one of the most important tasks of the power system.

[0003] Rural distribution networks are different from urban distribution networks; rural distribution networks have the characteristics of varying load densities, rapid load development, large power supply radius, small power supply line cross-section, and large seasonal load fluctuations. These characteristics lead to a wide variety of problems in rural distribution lines, and the causes of these problems are varied; and the interaction of various factors also causes low voltage, heavy overload and other problems to frequently occur in rural distribution networks, posing a great threat to the safe and stable operation of the lines. Therefore, the decision-making on the governance measures for rural distribution network problems is of great significance to the power system.

[0004] At present, the traditional decision-making process for rural distribution network problem governance measures is still a manual decision-making process, that is, the power system operators make single, isolated and manual governance decisions based on the problems in the rural distribution network. However, this manual decision-making solution has poor reliability and accuracy. Moreover, rural distribution network problems may have key and non-key problems, or there may be correlation characteristics between multiple problems. It is also obviously unreasonable to make governance decisions based on a single problem. Summary of the invention

[0005] One of the purposes of the present invention is to provide a method for making decisions on rural distribution network problem management measures that is highly reliable, accurate and scientifically reasonable.

[0006] A second objective of the present invention is to provide a system for implementing the rural distribution network problem management measures decision-making method.

[0007] The method for decision-making on rural distribution network problem governance measures provided by the present invention comprises the following steps:

[0008] S1. Obtain working data information and problem data information of the target rural distribution network;

[0009] S2. According to the data information obtained in step S1, determine several typical application scenarios and correlation indicators of each typical application scenario;

[0010] S3. Perform cluster analysis on the data information of typical application scenarios according to the correlation index determined in step S2 to obtain the subdivision scenarios of typical application scenarios;

[0011] S4. Determine the boundary conditions corresponding to the typical application scenario according to the segmented scenario obtained in step S3;

[0012] S5. Based on the boundary conditions obtained in step S4, decisions are made on measures to address rural distribution network issues.

[0013] Step S2, according to the data information obtained in step S1, determines a number of typical application scenarios and correlation indicators of each typical application scenario, specifically including the following steps:

[0014] Identify several typical application scenarios;

[0015] For each typical application scenario, the corresponding equipment operation data and equipment parameter data are screened to form the preset influencing factors of each typical application scenario;

[0016] According to the preset influencing factors of each typical application scenario, historical data information of the preset influencing factors is obtained; the historical data information is divided into several intervals, and the number of devices in each interval is counted;

[0017] Calculate the corresponding proportion value according to the number of devices in each interval; and select the influencing factors whose difference between the proportion values ​​is greater than the set value as the correlation indicator.

[0018] Step S3, based on the correlation index determined in step S2, performs cluster analysis on the data information of the typical application scenario to obtain the subdivided scenarios of the typical application scenario, specifically includes the following steps:

[0019] A. Divide each correlation index determined in step S2 into several intervals;

[0020] B. Arrange and combine each interval to form a preliminary subdivision interval, and count the number of correlation indicators of each preliminary subdivision interval;

[0021] C. Screen several subdivision intervals with the largest number of correlation indicators, and calculate the average value of the correlation indicators of each subdivision interval obtained by screening as the initial cluster center;

[0022] D. Use the following formula to calculate the distance between each data and the current cluster center:

[0023]

[0024] Where dist(x,c i ) is the data point x and the i-th cluster center c idistance; d is the dimension value; x j is the value of the jth dimension of the data point x; c ij is the cluster center c i The value of the jth dimension of ;

[0025] E. Assign each data to the nearest cluster center;

[0026] F. After the allocation is completed, the mean of all data points in each cluster is calculated using the following formula:

[0027]

[0028] Where c i ' is the mean of the data points of the i-th cluster center; |S i | is the number of data points in the i-th cluster center;

[0029] G. Use the mean as the new cluster center;

[0030] H. Repeat steps D to G until the distance between the new cluster center and the cluster center of the previous round is less than the set threshold, or the number of repetitions is reached;

[0031] I. The final cluster centers and the corresponding cluster data sets are tested for mean square value, degree of freedom, F value and significance level, and the cluster centers that pass the test are used as segmentation scenarios.

[0032] Step S4, according to the subdivided scenarios obtained in step S3, determines the boundary conditions corresponding to the typical application scenarios, specifically including the following steps:

[0033] According to the segmented scenarios obtained in step S3, the data range of the correlation index in the segmented scenarios is obtained, and the value range of the correlation index whose data difference degree is greater than the set value is selected as the boundary condition of the typical application scenario.

[0034] Step S5, according to the boundary conditions obtained in step S4, makes a decision on the treatment measures for the rural distribution network problems, which specifically includes the following steps:

[0035] According to the boundary conditions obtained in step S4, the expert method is used to make decisions on the management measures for the rural distribution network problems.

[0036] The present invention also provides a system for implementing the rural distribution network problem governance measure decision-making method, comprising a data acquisition module, a scenario determination module, a cluster analysis module, a boundary determination module and a measure decision module; the data acquisition module, the scenario determination module, the cluster analysis module, the boundary determination module and the measure decision module are connected in series in sequence; the data acquisition module is used to acquire the working data information and problem data information of the target rural distribution network, and upload the data information to the scenario determination module; the scenario determination module is used to determine a number of typical application scenarios and correlation indicators of each typical application scenario according to the received data information and the acquired data information, and upload the data information to the cluster analysis module; the cluster analysis module is used to perform cluster analysis on the data information of the typical application scenario according to the received data information and the determined correlation indicators, obtain the subdivided scenarios of the typical application scenario, and upload the data information to the boundary determination module; the boundary determination module is used to determine the boundary conditions corresponding to the typical application scenario according to the received data information and the obtained subdivided scenarios, and upload the data information to the measure decision module; the measure decision module is used to make decisions on rural distribution network problem governance measures according to the received data information and the obtained boundary conditions.

[0037] The method and system for making decisions on measures to address rural distribution network problems provided by the present invention not only realize automatic decision-making on measures to address rural distribution network problems through the selection, segmentation, clustering and decision-making of typical scenarios involved in rural distribution network problems, but also have higher reliability, better accuracy, and are more scientific and reasonable. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 The figure is a schematic diagram of the method flow of the present invention.

[0039] Figure 2 This is a schematic diagram of actual data of preset influencing factors when selecting correlation indicators in an embodiment of the method of the present invention.

[0040] Figure 3 Schematic diagram of the functional modules of the system of the present invention. DETAILED DESCRIPTION

[0041] like Figure 1 The method flow diagram of the method of the present invention is shown as follows: The rural distribution network problem governance measure decision-making method disclosed in the present invention comprises the following steps:

[0042] S1. Obtain working data information and problem data information of the target rural distribution network;

[0043] S2. According to the data information obtained in step S1, determine several typical application scenarios and correlation indicators of each typical application scenario; specifically including the following steps:

[0044] Identify several typical application scenarios; in specific implementation, the typical application scenarios determined include heavy overload of lines, heavy overload of distribution transformers, low voltage at the outlet of the substation area, and low voltage at users;

[0045] For each typical application scenario, the corresponding equipment operation data and equipment parameter data are screened to form the preset influencing factors of each typical application scenario;

[0046] According to the preset influencing factors of each typical application scenario, historical data information of the preset influencing factors is obtained; the historical data information is divided into several intervals, and the number of devices in each interval is counted;

[0047] Calculate the corresponding proportion value according to the number of devices in each interval; and select the influencing factors whose difference between the proportion values ​​is greater than the set value as the correlation index;

[0048] S3. According to the correlation index determined in step S2, cluster analysis is performed on the data information of the typical application scenario to obtain the subdivided scenarios of the typical application scenario; specifically, the steps include:

[0049] A. Divide each correlation index determined in step S2 into several intervals;

[0050] B. Arrange and combine each interval to form a preliminary subdivision interval, and count the number of correlation indicators of each preliminary subdivision interval;

[0051] C. Screen several subdivision intervals with the largest number of correlation indicators, and calculate the average value of the correlation indicators of each subdivision interval obtained by screening as the initial cluster center;

[0052] D. Use the following formula to calculate the distance between each data and the current cluster center:

[0053]

[0054] Where dist(x,c i ) is the data point x and the i-th cluster center c i distance; d is the dimension value; x j is the value of the jth dimension of the data point x; c ij is the cluster center c i The value of the jth dimension of ;

[0055] E. Assign each data to the nearest cluster center;

[0056] F. After the allocation is completed, the mean of all data points in each cluster is calculated using the following formula:

[0057]

[0058] Where ci ' is the mean of the data points of the i-th cluster center; |S i | is the number of data points in the i-th cluster center;

[0059] G. Use the mean as the new cluster center;

[0060] H. Repeat steps D to G until the distance between the new cluster center and the cluster center of the previous round is less than the set threshold, or the number of repetitions is reached;

[0061] I. Test the mean square value, degree of freedom, F value and significance level for the final cluster centers and the corresponding cluster data sets, and use the cluster centers that pass the test as the segmentation scenarios;

[0062] In specific implementation, the mean square value is used to express the ratio of the sum of squared deviations to the degrees of freedom. Since the size of each sum of squared error is related to the number of observations, in order to eliminate the influence of the number of observations on the size of the sum of squared error, it is necessary to average them, that is, to divide each sum of squares by their corresponding degrees of freedom. This result is called the mean square value.

[0063] Degrees of freedom, which is the amount of information in the data, is determined by the number of observations in the sample. Increasing the sample size provides more information about the population, thereby increasing the total degrees of freedom. Increasing the number of terms in the model uses more information, which reduces the degrees of freedom available to estimate the variability of parameter estimates.

[0064] The F value is one of the most important statistics in the ANOVA table. It represents the ratio between the variation between groups and the variation within groups. The larger the F value, the greater the difference between groups. Usually, if the F value is greater than the F critical value (usually 2.70), it means that at least one group of data is significantly different from the other groups of data.

[0065] The significance level is used to indicate the significance level of hypothesis testing, usually with 0.05 as the standard; if the value is less than 0.05, it means that the statistical result is significant when the null hypothesis is rejected;

[0066] S4. According to the subdivided scenarios obtained in step S3, the boundary conditions corresponding to the typical application scenarios are determined; specifically, the following steps are included:

[0067] According to the segmented scenarios obtained in step S3, the data range of the correlation index in the segmented scenarios is obtained, and the value range of the correlation index whose data difference degree is greater than the set value is selected as the boundary condition of the typical application scenario;

[0068] S5. Based on the boundary conditions obtained in step S4, decision-making on the treatment measures for the rural distribution network problems is made; specifically, the steps include:

[0069] According to the boundary conditions obtained in step S4, the expert method is used to make decisions on the management measures for the rural distribution network problems.

[0070] The method of the present invention is further described below in conjunction with an embodiment:

[0071] The preset typical application scenario is heavy line overload;

[0072] In view of the existing problems of 10kV line heavy overload and voltage over-limit, 9 influencing indicators are screened and shown in Table 1. These influencing indicators can reflect the existing problems of 10kV line and their causes.

[0073] Table 1 Schematic table of impact indicators

[0074] Serial number Indicator name Indicator Description 1 Lead wire model Reflects the current carrying capacity of 10kV line 2 Power supply radius The maximum power supply distance from 10kV line to end user 3 Power supply load Respond to the load of 10kV line 4 Average load factor Analysis of the average load rate of 10kV lines 5 Light load time Reflects the total light load time of 10kV line 6 Reload time Total duration of 10kV line overload response 7 Overload duration Total time of reaction to 10kV line overload 8 Number of low voltage areas Reflect the serious situation of low voltage at the end of 10kV line 9 Low voltage days Reflect the serious situation of low voltage at the end of 10kV line

[0075] The provincial 10kV line is selected as the data sample, and the actual range of various indicators is considered. The label attributes are divided as the basis for cluster analysis of typical problem scenarios, and the distribution characteristics of different indicator labels are statistically analyzed, that is, significance analysis is performed; the labels of influencing factors of problem scenarios are shown in Table 2:

[0076] Table 2 Schematic diagram of labels of factors affecting problem scenarios

[0077]

[0078] The results of statistical analysis based on the label attributes in Table 2 are as follows Figure 2 (a)~(h) Figure 2 From (a) to (h), we can see that the distribution characteristics of 10kV line labels are as follows:

[0079] (1) In terms of power supply radius, lines with a radius less than 5 km account for the highest proportion, followed by lines in other intervals.

[0080] (2) The maximum load is mainly concentrated in the range of 0-2MW and 2-4MW.

[0081] (3) In terms of average load factor, the lines with a load factor less than 10% account for more than 50% and the rest of the lines decrease gradually.

[0082] (4) In terms of light load duration, lines with light load duration exceeding 5000 h account for the highest proportion, while other intervals are distributed relatively evenly.

[0083] (5) The lines with the longest overload duration in the range of 10 to 50 hours and greater than 100 hours are the most numerous, and the overall distribution is relatively even;

[0084] (6) The lines with overload duration greater than 100 h are the most numerous, and the rest are evenly distributed.

[0085] (7) The indicators of the number of low-voltage areas and the number of days with low voltage are concentrated in the range of 1 to 5 areas / day, and the number decreases significantly as the indicators increase;

[0086] Therefore, it can be seen that the indicators with significant differences in distribution characteristics are the average load rate (%), heavy load duration (H), and light load duration (H). Therefore, these three influencing indicators are regarded as correlation indicators.

[0087] Then cluster analysis was performed, and the following three sub-scenarios were obtained after screening for line overload scenarios:

[0088] Table 3 Schematic diagram of subdivided scenarios

[0089]

[0090] Then, the boundary conditions of scenario 1 can be determined as the heavy load duration is more than 50 hours and the light load duration is less than 1000 hours;

[0091] The boundary conditions of scenario 2 can determine that the heavy load duration does not exceed 50 hours, and the light load duration is 1000-2000 hours;

[0092] The boundary conditions of scenario three can be determined as heavy load duration not exceeding 10 hours and light load duration 1000-2000 hours;

[0093] Finally, the expert method is used to obtain the corresponding treatment measures decision results:

[0094] The boundary conditions of scenario 1 are that the heavy load duration exceeds 50 hours and the light load duration is less than 1000 hours. The strategy of building or renovating a 10 kV line can be selected.

[0095] The boundary conditions of scenario 2 can be determined as heavy load duration not exceeding 50 hours and light load duration of 1000-2000 hours; load transfer or installation of reactive power compensation is preferred for governance. If the expected effect is not good, a distributed energy storage governance strategy can also be selected;

[0096] The boundary conditions of scenario three can be determined as heavy load duration not exceeding 10 hours and light load duration of 3000-4000 hours. The wiring mode can be flexibly modified or a distributed energy storage management strategy can be configured.

[0097] The scheme of the present invention sets typical problem scenarios according to typical problems existing in rural distribution networks, and determines indicators with differentiated characteristics according to the significance of each indicator in each typical application scenario, so that each typical problem scenario can be subdivided, and then the boundary conditions of each typical problem scenario are determined according to the distribution characteristics of the correlation indicators in each subdivided scenario, so that the boundary conditions of each typical problem scenario can be accurately obtained, so that in the process of configuring the governance strategy, the boundary conditions can be accurately configured, the rationality and accuracy of the governance strategy can be improved, so as to effectively improve the effectiveness of the governance of various problems in the rural distribution network and ensure the safe and stable operation of the rural distribution network.

[0098] like Figure 2 The functional module schematic diagram of the system of the present invention is shown as follows: the system for realizing the rural distribution network problem governance measure decision method disclosed by the present invention comprises a data acquisition module, a scenario determination module, a cluster analysis module, a boundary determination module and a measure decision module; the data acquisition module, the scenario determination module, the cluster analysis module, the boundary determination module and the measure decision module are connected in series in sequence; the data acquisition module is used to acquire the working data information and the problem data information of the target rural distribution network, and upload the data information to the scenario determination module; the scenario determination module is used to determine a number of typical application scenarios and the correlation index of each typical application scenario according to the received data information and the acquired data information, and upload the data information to the cluster analysis module; the cluster analysis module is used to perform cluster analysis on the data information of the typical application scenario according to the received data information and the determined correlation index, obtain the subdivided scenario of the typical application scenario, and upload the data information to the boundary determination module; the boundary determination module is used to determine the boundary conditions of the corresponding typical application scenario according to the received data information and the obtained subdivided scenario, and upload the data information to the measure decision module; the measure decision module is used to make decisions on the rural distribution network problem governance measures according to the received data information and the obtained boundary conditions.

Claims

1. A decision-making method for solving rural distribution network problems, comprising the following steps: S1. Obtain working data information and problem data information of the target rural distribution network; S2. According to the data information obtained in step S1, determine several typical application scenarios and correlation indicators of each typical application scenario; S3. Perform cluster analysis on the data information of typical application scenarios according to the correlation index determined in step S2 to obtain the subdivision scenarios of typical application scenarios; S4. Determine the boundary conditions corresponding to the typical application scenario according to the segmented scenario obtained in step S3; S5. Based on the boundary conditions obtained in step S4, decisions are made on measures to address rural distribution network issues.

2. The rural distribution network problem governance decision-making method according to claim 1 is characterized in that Step S2, according to the data information obtained in step S1, determines a number of typical application scenarios and correlation indicators of each typical application scenario, specifically including the following steps: Identify several typical application scenarios; For each typical application scenario, the corresponding equipment operation data and equipment parameter data are screened to form the preset influencing factors of each typical application scenario; According to the preset influencing factors of each typical application scenario, historical data information of the preset influencing factors is obtained; the historical data information is divided into several intervals, and the number of devices in each interval is counted; Calculate the corresponding proportion value according to the number of devices in each interval; and select the influencing factors whose difference between the proportion values ​​is greater than the set value as the correlation indicator.

3. The rural distribution network problem governance measure decision-making method according to claim 2 is characterized in that Step S3, based on the correlation index determined in step S2, performs cluster analysis on the data information of the typical application scenario to obtain the subdivided scenarios of the typical application scenario, specifically includes the following steps: A. Divide each correlation index determined in step S2 into several intervals; B. Arrange and combine each interval to form a preliminary subdivision interval, and count the number of correlation indicators of each preliminary subdivision interval; C. Screen several subdivision intervals with the largest number of correlation indicators, and calculate the average value of the correlation indicators of each subdivision interval obtained by screening as the initial cluster center; D. Use the following formula to calculate the distance between each data and the current cluster center: Where dist(x,c i ) is the data point x and the i-th cluster center c i distance; d is the dimension value; x j is the value of the jth dimension of the data point x; c ij is the cluster center c i The value of the jth dimension of ; E. Assign each data to the nearest cluster center; F. After the allocation is completed, the mean of all data points in each cluster is calculated using the following formula: Where c i ' is the mean of the data points of the i-th cluster center; |S i | is the number of data points in the i-th cluster center; G. Use the mean as the new cluster center; H. Repeat steps D to G until the distance between the new cluster center and the cluster center of the previous round is less than the set threshold, or the number of repetitions is reached; I. The final cluster centers and the corresponding cluster data sets are tested for mean square value, degree of freedom, F value and significance level, and the cluster centers that pass the test are used as segmentation scenarios.

4. The rural distribution network problem governance decision-making method according to claim 3 is characterized in that Step S4, according to the subdivided scenarios obtained in step S3, determines the boundary conditions corresponding to the typical application scenarios, specifically including the following steps: According to the segmented scenarios obtained in step S3, the data range of the correlation index in the segmented scenarios is obtained, and the value range of the correlation index whose data difference degree is greater than the set value is selected as the boundary condition of the typical application scenario.

5. The rural distribution network problem governance decision-making method according to claim 4 is characterized in that Step S5, according to the boundary conditions obtained in step S4, makes a decision on the treatment measures for the rural distribution network problems, which specifically includes the following steps: According to the boundary conditions obtained in step S4, the expert method is used to make decisions on the management measures for the rural distribution network problems.

6. A system for implementing the rural distribution network problem governance measure decision-making method according to any one of claims 1 to 5, characterized in that It includes a data acquisition module, a scenario determination module, a cluster analysis module, a boundary determination module and a measure decision module; the data acquisition module, the scenario determination module, the cluster analysis module, the boundary determination module and the measure decision module are connected in series in sequence; the data acquisition module is used to obtain the working data information and problem data information of the target rural distribution network, and upload the data information to the scenario determination module; the scenario determination module is used to determine several typical application scenarios and the correlation indicators of each typical application scenario according to the received data information and the acquired data information, and upload the data information to the cluster analysis module; The cluster analysis module is used to perform cluster analysis on the data information of the typical application scenario according to the received data information and the determined correlation index, obtain the segmented scenarios of the typical application scenario, and upload the data information to the boundary determination module; the boundary determination module is used to determine the boundary conditions corresponding to the typical application scenario according to the received data information and the obtained segmented scenarios, and upload the data information to the measure decision module; The measure decision module is used to make decisions on rural distribution network problem management measures based on the received data information and the obtained boundary conditions.

Images