Research On Lightning Risk Classification Of 10kV Distribution Line Tower

With the rapid development of my country’s basic economic construction,the scale of the power grid continues to grow,and the power sector is paying more and more attention to the safe operation and reliable power supply of distribution lines in mountainous areas.Lightning strikes are one of the main reasons for the failure of distribution lines.The mountainous terrain is complex and changeable,the soil resistivity is high,and the weather conditions are bad.The 10kV distribution line has a wide range of installation and the overall insulation level is low.It is often in areas with frequent lightning strikes.There was a lightning strike.The current lightning protection measures for distribution lines are often simply divided into mountainous areas and plains,without further detailed division of mountainous terrain and landforms,resulting in a large discrepancy between the final assessment results and actual conditions.Therefore,study the relationship between topography and landforms and lightning damage to lines,accurately identify the risk of lightning damage to power distribution line towers under different topography and landforms,and promote the intelligent development of power grid lightning protection,and improve the lightning protection of 10kV distribution lines in mountainous areas.And safe operation has important practical engineering significance.Taking a certain mountainous area in the south as the research object,the line and topography data are obtained through field investigation,the relationship between the lightning activity and the topography and topography of the line tower is studied,and the characteristic parameters affecting the lightning risk of the tower are determined,and the characteristic parameters are determined.Assign value to prepare for the subsequent lightning strike risk classification of power distribution line towers.Based on the historical lightning data of a 10kV distribution line corridor in Shaoyang,training samples and test samples are established,and a lightning risk classification and identification model for distribution line towers based on BP neural network and support vector machine is proposed.The improved fruit fly algorithm is used to The kernel function parameter g and penalty variable c in the support vector machine were optimized,and the experimental results were compared.It was found that the proposed support vector machine optimization model of the improved fruit fly algorithm can better identify the lightning strikes of the towers under different terrains.The probability of the risk.In order to further improve the classification accuracy of the model,a multi-source data fusion technology for pole tower lightning risk classification research algorithm is proposed.The thesis adopts the evidence synthesis method of the similarity coefficient between the evidences,and takes the credibility as the weight of the evidence to carry out the weighted average,and then uses the D-S evidence theory to synthesize,which effectively improves the synthesis of conflicting evidence.The test results show that,based on the conflict evidence synthesis method,the data fusion method using BP neural network(or multi-class SVM)combined with DS evidence theory has better fault tolerance and higher accuracy,which is effective and feasible.The lightning risk classification method for poles and towers solves practical problems to a certain extent,and can provide a certain reference and reference for the future work of lightning protection in mountainous areas.