Power transmission line typical ice wind disaster analysis method based on fusion of FCE and SVM
Technical Field
The invention belongs to the technical field of ice wind disaster analysis of a power transmission line, and particularly relates to a typical ice wind disaster analysis method of the power transmission line based on fusion of FCE and SVM.
Background
The ice wind disaster of the power transmission line is a big problem facing a power supply system. The ice wind disasters of different degrees occur all over the world, the power grid system in China also has the ice wind disasters for many times, and the occurrence frequency of the disasters is in an ascending trend. The method warns people of huge economic loss and social influence caused by ice wind disasters of the power transmission line, and developing disaster prevention research of the power system and ensuring safe and stable operation of the power system in the ice wind disasters are challenges and difficult tasks which are not slow at present. When the ice wind disaster of the power transmission line is coped with, besides the ice covering prevention technology, the ice melting technology and the deicing technology, the method has very important significance for effectively identifying and predicting the ice wind disaster.
The mature current risk assessment of the power transmission line and the wide application are equipment risk assessment, and the risk assessment refers to the comprehensive consideration of risks in the aspects of equipment safety, economy, social influence and the like and the determination of equipment risk degree. However, in the field of power transmission line disaster prevention and reduction, the influence of a specific disaster type on a line is faced, evaluation factors relate to aspects such as historical records, operation and maintenance live conditions, body states, meteorological monitoring and forecasting, the related range is wide, and the correlation among the factors is unclear, so that the work in the field of power transmission line disaster prevention and reduction mainly aims at performing prevention and post-disaster maintenance work on a certain kind of events at present. Therefore, a method for analyzing the ice wind disaster caused disaster by coupling different influence factors by analyzing the key influence indexes of the ice wind disaster is very urgent.
Disclosure of Invention
The invention aims to provide a typical ice wind disaster analysis method of a power transmission line based on fusion of a Fuzzy Comprehensive Evaluation (FCE) and a Support Vector Machine (SVM), which can construct a line ice wind disaster analysis model based on the FCE and the SVM and provide theoretical and practical bases for analysis and judgment of typical ice wind disasters by analyzing key influence indexes of ice wind disasters.
In order to realize the purpose, the technical scheme adopted by the invention is as follows:
a power transmission line typical ice wind disaster analysis method based on FCE and SVM fusion is characterized in that: the method comprises the following steps:
the first step is as follows: determining an evaluation index set according to analysis of the factors causing the ice wind disaster of the power transmission line;
the second step is as follows: determining the weight of the evaluation index type and the weight of each subtype index;
the third step: for the evaluation of the ice wind disaster of the line, determining an evaluation grade according to the correlation analysis of the influence indexes and the fault occurrence;
the fourth step: establishing a fuzzy comprehensive evaluation mathematical model;
the fifth step: analyzing and correcting an FCE evaluation result;
a sixth step: establishing a nonlinear SVM regression model;
a seventh step of: and taking the typical parameter indexes determined by the fuzzy comprehensive evaluation method as input vectors of the SVM regression model, taking the probability of possible occurrence of the fault as output vectors, and grading the fault probability for judging the ice wind disaster.
In the second step, the evaluation index types include meteorological conditions, terrain and geographical conditions, and line structures and parameters;
in the second step, the subtype indexes comprise temperature, relative humidity, rainfall, wind speed, wind direction, altitude, landform, slope direction, condensation height, line trend, ground wire suspension height, wire diameter, load current and electric field, wire rigidity and wire split number.
In the seventh step, the grade is classified into a high possibility of failure, a low possibility of failure, and no failure.
The invention also provides a typical ice wind disaster analysis device of the power transmission line based on the fusion of the FCE and the SVM, which comprises the following components:
an index collection unit: determining an evaluation index set according to analysis of the factors causing the ice wind disaster of the power transmission line;
a weighting unit: determining the weight of the evaluation index type and the weight of each subtype index;
an evaluation unit: determining the evaluation level of the ice wind disaster of the line according to the correlation analysis of the influence indexes and the fault occurrence;
a mathematical model unit: establishing a fuzzy comprehensive evaluation mathematical model;
an analysis and correction unit: analyzing and correcting an FCE evaluation result;
a regression model unit: establishing a nonlinear SVM regression model;
a judging unit: and taking the typical parameter indexes determined by the fuzzy comprehensive evaluation method as input vectors of the SVM regression model, taking the probability of possible occurrence of the fault as output vectors, and grading the fault probability for judging the ice wind disaster.
Compared with the prior art, the invention has the advantages that:
according to the typical ice wind disaster analysis method for the power transmission line based on the fusion of the FCE and the SVM, the correlation degree of ice wind disaster influence indexes of the power transmission line is analyzed by using a fuzzy comprehensive evaluation method, key disaster influence indexes are extracted, meanwhile, a non-linear SVM small sample disaster analysis model adopting a radial basis RBF kernel function is provided, the key influence indexes are used as input quantity, the dimensionality of a classification algorithm is reduced, the problems of complex structure, low training speed, low efficiency and the like are effectively solved, the ice wind disaster of the power transmission line can be reliably analyzed under the condition of small sample data, the early warning effect of the ice wind disaster of the power transmission line is achieved, the disaster prevention and reduction measures are implemented in advance, and the loss of a power grid is avoided.
Drawings
FIG. 1 is a model structure diagram of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the following specific examples:
a power transmission line typical ice wind disaster analysis method based on FCE and SVM fusion is characterized in that:
step 1: and determining an evaluation index set.
Setting the evaluation index set as U-M11,M12,M13…MijIn the formula: m is an evaluation index; i is the total number of evaluation index types; j is the total number of each subtype index.
Evaluation indexes shown in table 1 are constructed by researching and analyzing the influence indexes of the ice wind disaster and combining with the expert experience of operation and maintenance.
TABLE 1
Step 2: and confirming the index weight.
Determining the weight A of the evaluation index type by combining the expert experience and the historical ice wind disaster fault information1And the weight A of each subtype index2Comprises the following steps:
A1=(M1,M2,M3)=(0.5,0.3,0.2)
A2=(M11,M12,M13…M36)
=(0.3,0.15,0.1,0.3,0.15,0.25,0.35,0.15,0.05,0.2,0.25,0.25,0.1,0.15,0.1,0.15)
and step 3: and determining the evaluation grade.
Let the evaluation grade be V ═ V (V)1,V2…Vj…Vn) In the formula: n is the total number of evaluation results; vjJ is 1,2, … n as the jth evaluation result. For the evaluation of the ice wind disaster of the line, the correlation analysis between the influence indexes and the fault occurrence is carried out, namely V1,V2,V3Respectively, high, middle and low 3 grades.
And 4, step 4: and establishing a fuzzy comprehensive evaluation mathematical model.
Performing single-factor fuzzy evaluation, and determining an evaluation matrix as follows:
in the formula: r isijFuzzy subset V of grade from factor U for a certain evaluated objectjDegree of membership.
The fuzzy comprehensive evaluation result vector B of each evaluated object is obtained by synthesizing the weight A of the evaluation index and the evaluation matrix R, and can be expressed as follows:
in the formula: bjFuzzy subset V of rating for rated object as a wholejDegree of membership.
Under meteorological conditions M1For example, the evaluation matrix is established as follows:
then B is1(0.475,0.37,0.155) in terms of maximum membership, M1The correlation with the occurrence of a failure is high.
And 5: and analyzing and correcting the FCE evaluation result.
And analyzing the membership degree and the weight coefficient of each graded index and the comprehensive evaluation index result to obtain an index with the highest correlation with the occurrence of the ice wind fault, and then providing a correction factor of the index needing to be corrected by combining with the analysis of the occurrence mechanism of the ice wind fault.
Step 6: and establishing a nonlinear SVM regression model.
Assume that the training sample set is: s { (x)i,yi) 1,2 … N, wherein: x is the number ofiIs an input vector; y isiIs an output vector; and N is the number of samples.
The nonlinear SVM uses functional regression estimation: (x) ω · Φ (x) + b, wherein: omega is an m-dimensional vector; phi (x) is a non-linear transformation function; b is a threshold value.
The structure of the established typical ice wind disaster model of the power transmission line based on the fusion of the fuzzy comprehensive evaluation method and the support vector machine method is shown in fig. 1.
And 7: taking the typical parameter index determined by the fuzzy comprehensive evaluation method as an input vector of the SVM regression model, wherein the probability of possible fault occurrence is an output vector yiThe failure probability is classified for use in determining the ice wind disaster, as shown in table 2.
TABLE 2
Influence of faults
|
yiRange of (1)
|
Is likely to malfunction
|
0.85≤yi<1
|
Is likely to malfunction
|
0.5≤yi<0.85
|
The possibility of failure is less
|
0.2≤yi<0.5
|
Will not malfunction
|
0≤yi<0.2 |
Therefore, the possibility of ice and wind disaster faults of the power transmission line in different states can be analyzed.
Those not described in detail in this specification are within the knowledge of those skilled in the art.