CN104915681A - Transformer substation clustering method considering both voltage sag monitoring information and associated factor thereof - Google Patents

Abstract

The invention discloses a substation clustering method that takes into account voltage sag monitoring information and associated factors, including establishing voltage sag event monitoring information indicators and associated factor indicators; inputting voltage sag characteristic index data of power grid substations in areas to be analyzed , and carry out normalization processing; using the fuzzy C-means clustering method, clustering the substations of the power grid in the area to be analyzed according to the three voltage sag characteristic indicators respectively, and obtaining the clustering results in three dimensions; selecting key indicators, using fuzzy The C-means clustering method, according to the key indicators, clusters the substations of the power grid in the area to be analyzed to obtain a comprehensive clustering result; mainly based on the comprehensive clustering results of the substations, combined with the clustering results of three dimensions, the voltage sag of each category of substations is analyzed characteristics, and give preliminary suggestions for voltage sag control. The clustering method of the invention has the advantages of comprehensive index setting, good classification effect, rich extracted information and the like.

Description

Substation clustering method considering voltage sag monitoring information and its associated factors

technical field

The invention relates to the technical field of power systems, in particular to a substation clustering method that takes into account voltage sag monitoring information and associated factors.

technical background

At present, the problem of voltage sag has become increasingly prominent, and has gradually become the most serious power quality problem. In particular, the voltage sag caused by the short-circuit fault of the main network has a deep degree of sag, a wide range of influence, and very serious harm. Taking the substations in the regional power grid as the object, it is urgent and reasonable to study the prevention and control measures of voltage sag.

There are a large number of substations in regional power grids (such as provincial power grids and municipal power grids) in various places in my country. However, due to technical and financial constraints, it is difficult to prevent and control all substations in regional power grids, and analyze the large number of substations one by one. The workload is enormous and impractical. Therefore, it is of great practical significance to classify the substations in the regional power grid according to their voltage sag characteristics and identify the substations that urgently need to be prevented from voltage sag.

Cluster analysis is a classification method of multivariate statistics, which can divide data with similar statistical characteristics into one category according to index data, and is suitable for fast and effective classification processing of a large number of objects. Among the existing clustering analysis methods, the fuzzy C-means clustering method is a relatively mature method, which is widely used.

The establishment of clustering indicators is a key step in clustering analysis. To conduct clustering analysis on substations for the purpose of preventing voltage sags, the clustering indicators should first consider the monitoring information of voltage sags, such as the voltage sags in the latest period The number of times, the average value of voltage sag amplitude, the average value of voltage sag duration, etc. In addition, more serious voltage sags are mostly caused by short-circuit faults. In places with high lightning density, the probability of lightning strikes is high, and places with high thunderstorm levels are prone to lightning short-circuit accidents, which in turn cause voltage sags. In heavily polluted lines, the probability of pollution flashover is high, and pollution flashover accidents are prone to occur in places with high levels of pollution, which in turn leads to voltage sags. Therefore, the level of thunderstorm and the level of pollution area are important correlation factors for the voltage sag in the external environment dimension of the substation. Similarly, the total capacity of the main transformer and the cable conversion rate are important factors related to voltage sag in the substation electrical parameter dimension.

When using a clustering algorithm for calculation, if too many indicators are used, it is easy to cause the disaster of dimensionality, and the clustering effect is not good, so dimension reduction processing is required. At present, the algorithms used for dimensionality reduction include principal component analysis, factor analysis, etc., but these algorithms determine which indicators are key indicators based on the distribution of the data itself, and the selected indicators may not be the indicators that people are most concerned about.

Contents of the invention

The purpose of the present invention is to overcome the shortcomings and deficiencies of the prior art, and provide a regional substation clustering method for voltage sag monitoring information and related factors. Based on information and related factors, three-dimensional clustering indicators are established, and through multi-dimensional clustering, voltage sag information is deeply excavated to obtain scientific classification results. This method has the advantages of comprehensive index setting, good classification effect, and rich information extraction. It can provide effective support for power supply enterprises to analyze the multi-dimensional characteristics of voltage sag in regional power grid substations and strengthen the prevention and control of voltage sag.

The purpose of the present invention is achieved through the following technical solutions:

The substation clustering method taking into account the voltage sag monitoring information and its associated factors is characterized in that it includes the following steps:

S1. Establish three-dimensional clustering indicators, including voltage sag event monitoring information indicators and associated factor indicators, where the associated factor indicators include substation external environment indicators and electrical parameter indicators;

S2. Normalize the three-dimensional index data of the power grid substation in the area to be analyzed;

S3. Using the fuzzy C-means clustering method to cluster the power grid substations in the area to be analyzed, and obtain three-dimensional clustering results;

S4. Extract a relatively important indicator from each of the three dimensions as a key indicator;

S5. Based on the key indicators, the fuzzy C-means clustering method is used to cluster the power grid substations in the area to be analyzed to obtain a comprehensive clustering result;

S6. According to the clustering results of S3 and S5, comprehensively analyze the voltage sag characteristics of each type of substation, and give preliminary suggestions for voltage sag control.

Specifically, the formula for the normalization process described in step S2 is as follows:

${\mathrm{<span\; class="notranslate">\; b</span>}}_{\mathrm{<span\; class="notranslate">\; ij</span>}}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; ij</span>}}}{\underset{\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; \&le;</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; \&le;</span>\mathrm{<span\; class="notranslate">\; no</span>}}{\mathrm{<span\; class="notranslate">\; max</span>}}<span\; class="notranslate">\; \{</span>{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; ij</span>}}<span\; class="notranslate">\; \}</span>}$

In the formula, i=1,2,3...,n, j=1,2,3...,m, a _ij is the original value of the jth indicator of the i-th object, b _ij is the normalized value of a _ij value.

Specifically, the fuzzy C-means clustering method includes the following steps:

(1) Initialize the membership matrix U with a random number between 0 and 1, so that it satisfies the formula

$\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; c</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; u</span>}}_{\mathrm{<span\; class="notranslate">\; ij</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; \&ForAll;</span>\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; no</span>}$

Constraints in;

(2) Calculate c cluster centers c _i , where i=1,...,c;

(3) Calculate the objective function. If the value of the objective function is greater than the threshold, calculate a new membership matrix and return to step (2); if the value of the objective function is less than the threshold, or its change relative to the value of the previous value function is less than a certain threshold, the algorithm stops and the output cluster result.

Specifically, the voltage sag event monitoring information indicators described in step S1 include the number of voltage sags, the average value of the voltage sag amplitude, and the average value of the voltage sag duration; the external environmental indicators of the substation include thunderstorm levels, polluted areas Grade; the electrical parameter index of the substation, including the total capacity of the main transformer and the cable conversion rate.

Specifically, the number of voltage sags refers to the number of voltage sags monitored by the power quality monitoring device of the substation within one year; the average value of the voltage sag amplitude refers to the number of times the power quality monitoring device of the substation monitors within one year The average value of all voltage sag amplitudes; the average voltage sag duration refers to the average value of all voltage sag durations monitored by the power quality monitoring device of the substation within one year.

Specifically, the thunderstorm level is an index reflecting the intensity of lightning strikes in the area where the substation is located, and is divided into five levels: 1, 2, 3, 4, and 5, with 5 being the highest level, and the higher the level, the higher the intensity of lightning strikes.

Specifically, the polluted area level is an index reflecting the pollution degree of the substation line, which is divided into five levels: 1, 2, 3, 4, and 5, with 5 being the highest level, and the higher the level, the higher the pollution degree.

Specifically, the total capacity of the main transformer refers to the sum of the capacities of all transformers in the substation.

Specifically, the cable conversion rate refers to the proportion of cable length in all incoming and outgoing lines of the substation, and its calculation formula is as follows:

$\mathrm{<span\; class="notranslate">\; \&eta;</span>}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; l</span>}}{\mathrm{<span\; class="notranslate">\; l</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; d</span>}}$

Among them, l is the sum of the lengths of the cables in the incoming and outgoing lines of the substation, and d is the sum of the lengths of the overhead lines in the incoming and outgoing lines of the substation.

Specifically, the comprehensive analysis in the step S6 refers to analyzing the severity of voltage sag events in each category of substations and the degree of influence by related factors based on the comprehensive clustering results of substations, combined with the clustering results in three dimensions .

Compared with the prior art, the present invention has the following advantages and beneficial effects:

1. The regional power grid substation clustering method of the present invention taking into account the voltage sag monitoring information and its related factors takes the regional power grid substation as the research object, establishes clustering indicators around the voltage sag event and its related factors, and adopts fuzzy C-means clustering method The substations are clustered using the class method, which provides a new method for power supply enterprises to control voltage sags and clarify key control objects. This method reduces the reliance on expert experience in the decision-making process, reduces the workload of substation data analysis, and has good scalability.

2. The regional power grid substation clustering method of the present invention, which takes into account voltage sag monitoring information and its associated factors, establishes clustering indicators from three dimensions such as voltage sag monitoring information, external environment, and electrical parameters. The index establishment is scientific and reasonable, and can A more comprehensive response to voltage dips in substations.

3. The regional power grid substation clustering method of the present invention, which takes into account the voltage sag monitoring information and its associated factors, can obtain a clustering result that comprehensively reflects the voltage sag situation of the substation by screening key indicators and clustering based on the key indicators. It solves the problem of the curse of dimensionality that may be caused by too many indicators.

4. The regional power grid substation clustering method of the present invention, which takes into account the voltage sag monitoring information and its associated factors, performs clustering according to the three-dimensional indicators such as the voltage sag monitoring information, external environment, and electrical parameters, and performs clustering on the index data. A more in-depth excavation of the contained information can provide more references for decision makers.

Description of drawings

Fig. 1 is the flow chart of the substation clustering method taking into account the voltage sag monitoring information and related factors thereof according to the present invention;

Fig. 2 is the fuzzy C-means clustering method flowchart that method described in Fig. 1 adopts;

Fig. 3 is a three-dimensional scatter diagram of the comprehensive clustering results of regional power grid substations in the embodiment of the present invention.

Detailed ways

The present invention will be further described in detail below in conjunction with the embodiments and the accompanying drawings, but the embodiments of the present invention are not limited thereto.

As shown in Figure 1, the substation clustering method taking into account the voltage sag monitoring information and its associated factors, the specific steps are as follows:

First, according to the voltage sag event information monitored by the regional substation power quality monitoring system and the related factors of voltage sag, the characteristic indicators of voltage sag in regional power grid substations are established, including voltage sag event monitoring information indicators, substation external environment indicators and substation electrical parameters. Parameter index.

Voltage sag monitoring information indicators refer to the indicators established based on the details of voltage sag events recorded by the substation power quality monitoring system in the last year, specifically including the number of voltage sags and voltage sag amplitudes that occurred in the substation in the last year Average value, average value of voltage sag duration.

The external environment index of the substation is established from the perspective of factors related to voltage sag. Lightning strikes and pollution flashover accidents are important causes of voltage sag. Therefore, two indicators of thunderstorm level and pollution area level are established. Thunderstorm level refers to the level of thunderstorms, which is an indicator of the intensity of lightning in the area where the substation is located. It is divided into five levels: 1, 2, 3, 4, and 5. 5 is the highest level. The higher the level, the higher the intensity of lightning. Pollution level is an index reflecting the pollution degree of substation lines. It is divided into five levels: 1, 2, 3, 4, and 5. 5 is the highest level. The higher the level, the higher the pollution level.

The substation electrical parameter index is also established from the perspective of voltage sag related factors, and two indexes are established: the total capacity of the main transformer and the cable conversion rate. The total capacity of the main transformer refers to the sum of the capacities of all transformers in the substation. The cable ratio refers to the proportion of cables in all incoming and outgoing lines of the substation, and its calculation formula is as follows:

$\mathrm{\&eta;}=\frac{l}{l+d}$ ①

Among them, l is the sum of the lengths of the cables in the incoming and outgoing lines of the substation, and d is the sum of the lengths of the overhead lines in the incoming and outgoing lines of the substation.

Assuming that the number of substations in the regional power grid to be analyzed is n, input the voltage sag characteristic index data of the power grid substations in the region to be analyzed to form the original data matrix A=(a _ij ) _n×7 , where a _ij is the i-th substation’s The value of j indicators.

Normalize the original data matrix A to obtain a normalized matrix B=(b _ij ) _n×7 , and the normalized formula is as follows:

$b_{\mathrm{ij}}=\frac{a_{\mathrm{ij}}}{\underset{1\&le;i\&le;\mathrm{no}}{\max }\left\{a_{\mathrm{ij}}\right\}}$ ②

In the formula, i=1,2,3...,n, j=1,2,3...,m, a _ij is the original value of the jth indicator of the i-th object, b _ij is the normalized value of a _ij value.

The principle of the fuzzy C-means clustering method is as follows:

The basic idea of the fuzzy C-means algorithm is to divide n vectors X _j (j=1,2,…,n) into c fuzzy groups, and find the cluster center of each group, so that the objective function of the non-similarity index reaches minimum. The fuzzy C-means algorithm uses fuzzy division, and the degree to which each given data point belongs to each group is expressed by the degree of membership between 0 and 1. The sum of membership degrees of a data point is always equal to 1:

$\underset{i=1}{\overset{c}{\mathrm{\&Sigma;}}}{u}_{\mathrm{ij}}=1,\&ForAll;j=1,...,\mathrm{no}$ ③

The objective function is:

$J(u,c_1,...,c_c)=\underset{i=1}{\overset{c}{\mathrm{\&Sigma;}}}{J}_i=\underset{i=1}{\overset{c}{\mathrm{\&Sigma;}}}\underset{j}{\overset{\mathrm{no}}{\mathrm{\&Sigma;}}}{u}_{\mathrm{ij}}^m{d}_{\mathrm{ij}}^2,$ ④

Among them, c _i is the cluster center of fuzzy group I, u _ij is the membership degree of vector X _j to fuzzy group I, U is the membership degree matrix, and d _ij = ||ci -x _j || is the _Ith cluster The Euclidean distance between the center and the jth data point; m∈[1,∞) is the weighting index.

Constructing the Lagrangian equation can obtain the expressions of membership u _ij and cluster center c _i :

$c_i=\frac{\underset{j=1}{\overset{\mathrm{no}}{\mathrm{\&Sigma;}}}{u}_{\mathrm{ij}}^m{x}_j}{\underset{j=1}{\overset{\mathrm{no}}{\mathrm{\&Sigma;}}}{u}_{\mathrm{ij}}^m}$ ⑤

and

$u_{\mathrm{ij}}=\frac{1}{\underset{k=1}{\overset{c}{\mathrm{\&Sigma;}}}{\left(\frac{d_{\mathrm{ij}}}{d_{\mathrm{kj}}}\right)}^{2/(m-1)}}$ ⑥

The fuzzy C-means clustering method, its sub-steps are introduced as follows:

(1): Initialize the membership matrix U with a random number between 0 and 1, so that it meets the constraints in formula ③;

(2): Use formula ⑤ to calculate c cluster centers c _i (i=1,...,c);

(3): Calculate the objective function according to formula ④. If the value of the objective function is greater than the threshold, use formula ⑥ to calculate the new membership matrix, and return to step (2); Stop and output the clustering results.

In order to express the calculation steps more intuitively, Figure 2 shows the flow chart of the fuzzy C-means clustering method. The fuzzy C-means clustering method is used to cluster regional power grid substations based on the three indicators of the number of voltage sags, the average value of voltage sag amplitude, and the average value of voltage sag duration, and obtain the clustering of the monitoring information dimension of voltage sag events result.

The fuzzy C-means clustering method is used to cluster regional power grid substations based on the two indicators of thunderstorm level and pollution area level, and the clustering results of the external environment dimension are obtained.

The fuzzy C-means clustering method is used to cluster the substations of the regional power grid based on the two indicators of the total capacity of the main transformer and the cable conversion rate, and the clustering results of the electrical parameter dimension are obtained.

Clustering has been carried out from three dimensions above, but the clustering results of each dimension cannot reflect the comprehensive characteristics of the substation. If clustering is performed based on eight indicators, it will cause a disaster of dimensionality, and the clustering results are poorly differentiated and difficult to explain. In order to avoid the curse of dimensionality, dimensionality reduction is required. According to expert experience, one indicator is extracted from the three dimensions of voltage sag event monitoring information, external environment and electrical parameters as the key indicator.

According to the experience of experts, important indicators are selected from each dimension as key indicators. Although it has a certain degree of subjectivity, the clustering results have better comprehensiveness and interpretability.

Using the fuzzy C-means clustering method, according to the key indicators, the power grid substations in the area to be analyzed are clustered, and the comprehensive clustering results are obtained;

Based on the comprehensive clustering results of substations, combined with the three-dimensional clustering results, the severity of voltage sag events in various types of substations and the degree of influence by related factors are analyzed, and preliminary suggestions for voltage sag control are given.

The following is a further description combined with an application example. Taking 25 substations in a certain city (both voltage levels are 110kV and 220kV) as the object of analysis, the original data of eight clustering indicators are obtained, as shown in Table 1:

Table 1 Raw data of substation clustering index

After normalizing the original data of the eight clustering indicators, as shown in Table 2:

Table 2 Normalized data of substation clustering index

The fuzzy C-means clustering method is used to cluster regional power grid substations based on the three indicators of the number of voltage sags, the average value of the voltage sag amplitude, and the average value of the duration of voltage sags. Classes III, IV, and V, and the clustering results of voltage sag event monitoring information dimensions are given, as shown in Table 3:

Table 3 Clustering results of voltage sag event monitoring information dimensions

The fuzzy C-means clustering method is used to cluster regional power grid substations based on the two indicators of thunderstorm level and pollution level, and they are divided into 5 categories, which are recorded as I, II, III, IV, and V. The clustering results are shown in Table 4:

Table 4 Clustering results of external environment dimensions

category serial number Pollution level thunderstorm level Class I (4) 2, 6, 24, 25 2 4,5 Class II (3) 11, 13, 21 2 1, 2 Class III (4) 3, 5, 16, 20 3 1 Category IV (6) 7, 9, 12, 14, 17, 23 2, 3, 4 3, 4 Class V (8) 1, 4, 8, 10, 15, 18, 19, 22 3, 4 2

The fuzzy C-means clustering method is used to cluster the substations of the regional power grid based on the two indicators of the total capacity of the main transformer and the cable conversion rate, and they are divided into 5 categories, which are recorded as I, II, III, IV, and V. The clustering results of dimensions are shown in Table 5:

Table 5 Clustering results of electrical parameter dimensions

According to the experience of experts, the key indicators of voltage sag amplitude average value, thunderstorm level and cable conversion rate are respectively selected from the three dimensions of voltage sag event monitoring information, external environment and electrical parameters. Using the fuzzy C-means clustering method, according to the key indicators, the power grid substations in the area to be analyzed are clustered, and when they are divided into five categories, the comprehensive clustering results are obtained, as shown in Table 6:

Table 6 Comprehensive clustering results

From the comprehensive clustering results and their index values, a three-dimensional scatter diagram can be drawn, as shown in Figure 3. The distribution of different types of substations can be seen intuitively from the figure.

The following is based on the comprehensive clustering results of substations, combined with the clustering results of three dimensions, analyzes the severity of voltage sag events in various types of substations, and the degree of influence by related factors, and gives preliminary suggestions for voltage sag control.

It can be seen from Table 6 and Figure 3 that the 25 substations are divided into 5 categories. The distance between the substations of the same category is relatively close, which means that the similarity is high. The substations of different categories are distributed in different areas, and the boundaries between categories are clear. , which verifies the rationality of the clustering method.

Analyze the clustering results in Table 6 and Figure 3 by category, 1), 4 substations in category I have low thunderstorm level, low cable conversion rate, and high average voltage sag amplitude. The situation is not serious, but due to the high proportion of overhead lines, it is necessary to do a good job in line operation and maintenance; 2), 9 substations in category II have low thunderstorm levels, low cable conversion rates, and low average voltage sag amplitudes. The voltage sag in Class I substations is the most serious among the five categories, and it is the primary control object. The low level of thunderstorms indicates that the probability of voltage sag caused by lightning accidents is relatively small, and the cause should be found from other aspects. The proportion of overhead lines 3) The five substations in category III have medium thunderstorm levels, medium cable conversion rates, and average voltage sag amplitudes. The voltage sag of this type of substation is relatively serious and can be used as a key point Control object; 4), the cable conversion rate of the two substations in category IV is 1, and the incoming and outgoing lines are all cables, and there is no overhead line, so there is no need to consider the factor of lightning strikes. The severity of voltage sag in this type of substation is average; 5 ), the 5 substations in category V have high thunderstorm levels, low cable conversion rate, and the average voltage sag amplitude fluctuates between 41% and 66%. This type of substation should first strengthen the lightning protection measures for lines and equipment.

The comprehensive clustering results in Table 6 can be analyzed in combination with the clustering results of the three dimensions in Table 3, Table 4, and Table 5. Taking the 4 substations in Class I in Table 6 as an example, these 4 substations are listed in Table 3. The third category in the clustering results of the sag monitoring information dimension, as can be seen from Table 3, the number of voltage sags recorded in these 4 substations is only 2 to 16 times, the average value of the voltage sag amplitude is above 50%, and the voltage sag The average value of the duration is less than 0.15s, and the severity of the voltage sag is very low; in the clustering results of the external environment dimension in Table 4, these 4 substations mainly belong to their fifth category, with low thunderstorm level and high pollution area level, which need to be strengthened The lines are clean to prevent the occurrence of pollution flashover; in the clustering results of electrical parameters in Table 5, these 4 substations belong to the third category, the cable conversion rate is low, and the total capacity of the main transformer is below 200MVA, which belongs to the small-scale substation .

It can be seen that through the clustering method, 25 substations are divided into 5 categories, and each category has similar characteristics. In this way, the number of analysis objects is changed from 25 to 5, which greatly reduces the workload of analysis. The established clustering index covers voltage sag monitoring information and voltage sag related factors. By screening key indicators, the obtained comprehensive clustering results can more comprehensively reflect the voltage sag situation of substations, and clarify the governance objects and governance measures provide effective support. The substations are clustered from three dimensions, and the information contained in the clustering indicators is mined more deeply to provide more detailed references for decision makers.

The above-mentioned embodiment is a preferred embodiment of the present invention, but the embodiment of the present invention is not limited by the above-mentioned embodiment, and any other changes, modifications, substitutions, combinations, Simplifications should be equivalent replacement methods, and all are included in the protection scope of the present invention.

Claims (10)

1. take into account transformer station's clustering method of voltage dip monitoring information and relation factor thereof, it is characterized in that comprising the following steps:

S1. set up the clustering target of three dimensions, comprise voltage dip event monitoring information index and relation factor index thereof, wherein relation factor index comprises transformer station's external environment condition index and transformer station's electric parameter index;

S2. treat three the dimension achievement datas analyzing regional electricity grid substation to be normalized;

S3. adopt fuzzy C-means clustering method to treat the regional electricity grid substation of analysis and carry out cluster, obtain the cluster result of three dimensions;

S4. a relatively important index is respectively extracted, as key index from three dimensions;

S5. adopt fuzzy C-means clustering method, treat the regional electricity grid substation of analysis based on key index and carry out cluster, obtain synthesize cluster result;

S6. comprehensively analyze the voltage dip feature of each classification transformer station according to the cluster result of S3, S5, provide preliminary voltage dip disposing suggestion.

2. the transformer station's clustering method taking into account voltage dip monitoring information and relation factor thereof according to claim 1, is characterized in that: the formula of normalized described in step S2 is as follows:

$b_{\mathrm{ij}}=\frac{a_{\mathrm{ij}}}{\underset{1\&le;i\&le;n}{\max }\left\{a_{\mathrm{ij}}\right\}}$

In formula, i=1,2,3 ..., n, j=1,2,3 ..., m, a _ijbe the original value of a jth index of i-th object, b _ijfor a _ijnormalized value.

3. the regional power grid transformer station clustering method of consideration voltage dip monitoring information according to claim 1 and relation factor thereof, is characterized in that: described fuzzy C-means clustering method comprises the steps:

(1) with being worth 0, the random number initialization Subject Matrix U between 1, makes it meet formula

$\underset{i=1}{\overset{c}{\mathrm{\&Sigma;}}}{u}_{\mathrm{ij}}=1,\&ForAll;j=1,...,n$

In constraint condition;

(2) c cluster centre c is calculated _i, wherein i=1 ..., c;

(3) calculating target function: if target function value is greater than threshold value, then calculate new Subject Matrix, returns step (2); If target function value is less than threshold value, or its relative last time cost function value knots modification be less than certain threshold value, then algorithm stops, and exports cluster result.

4. the transformer station's clustering method taking into account voltage dip monitoring information and relation factor thereof according to claim 1, it is characterized in that: the event monitoring of voltage dip described in step S1 information index, comprise voltage dip number of times, voltage dip amplitude mean value, voltage dip duration mean value; Described transformer station external environment condition index, comprises thunderstorm grade, dirty district grade; Described transformer station electric parameter index, comprises main transformer total volume, cable rate.

5. the transformer station's clustering method taking into account voltage dip monitoring information and relation factor thereof according to claim 4, is characterized in that: described voltage dip number of times refers to the voltage dip number of times that the equipment for monitoring power quality of transformer station monitored in a year; Described voltage dip amplitude mean value refers to the mean value of all voltage dip amplitudes that the equipment for monitoring power quality of transformer station monitored in a year; Described voltage dip duration mean value refers to the mean value of all voltage dip duration that the equipment for monitoring power quality of transformer station monitored in 1 year.

6. the transformer station's clustering method taking into account voltage dip monitoring information and relation factor thereof according to claim 4, it is characterized in that: described thunderstorm grade is the index of reaction transformer station region thunderbolt dense degree, be divided into 1,2,3,4,5 five grade, 5 is highest level, higher grade, represents that thunderbolt dense degree is higher.

7. the transformer station's clustering method taking into account voltage dip monitoring information and relation factor thereof according to claim 4, it is characterized in that: described dirty district grade is the index of reaction substation line pollution level, be divided into 1,2,3,4,5 five grade, 5 is highest ranking, higher grade, represents that pollution level is higher.

8. the transformer station's clustering method taking into account voltage dip monitoring information and relation factor thereof according to claim 4, is characterized in that: described main transformer total volume refers to the capacity sum of all transformers of transformer station.

9. the transformer station's clustering method taking into account voltage dip monitoring information and relation factor thereof according to claim 4, is characterized in that: described cable rate refers to the ratio in all lines of transformer station shared by cable length, and its computing formula is as follows:

$\mathrm{\&eta;}=\frac{l}{l+d}$

Wherein, l is the length sum of cable in transformer station's line, and d is the length sum of pole line in transformer station's line.

10. the transformer station's clustering method taking into account voltage dip monitoring information and relation factor thereof according to claim 1, it is characterized in that: the comprehensive analysis in described step S6 refers to based on transformer station's synthesize cluster result, in conjunction with the cluster result of three dimensions, analyze the voltage dip event order of severity of each classification transformer station, and by relation factor effect.