Research On The Prediction Of Damage To The Power Line Along The Route After Typhoon Landing

With the deterioration of the global ecological environment,typhoon occurs frequently around the world.Located on the west coast of the western North Pacific,China is affected by the tropical ocean surface,with an average of 9.09 tropical cyclones including 3.17typhoon landing in.As the energy supply of production and life,electricity plays a supporting role in economic development and social stability.However,typhoon is very easy damaging the power transmission system,which poses a serious threat to the stability of power supply.With the characteristics of long transmission distance,wide range,continuity and complicated structure,power lines are greatly affected by strong typhoon and heavy rain,leading to power failure accidents such as tilting,collapse and line breakage of the tower.Typhoon frequently causes serious damage to the power lines in the southeastern coastal areas in China,which not only cut off electricity supply for social production and residents'daily life,but also hinder the effective development of rescue.Therefore,it is of great practical significance to study the influencing factors of typhoon damage to power lines and predict the damage amount.Not only can it improve the wind resistance of the power lines,but also facilitate the repair and restore power supply in time after disaster.Based on the characteristics of typhoon and power lines,this paper selects the damage of towers as the research object,and obtains 17 influencing factors through grey correlation analysis,including typhoon landing grade,longitudinal mileage,moving speed,storm level,rainfall duration,design grade of wind resistance,typhoon radius,number of lines in tower,path tortuosity,line running time,voltage level,wind direction angle,conductor cross section,distance from typhoon center,depth of tower base,number of conductor splits,line maintenance frequency.The correlation between factors and dependent variables is higher than 0.74.In order to eliminate the correlation between various influencing factors and effectively reduce the dimension,this paper divides the 17 influencing factors into 5 types,and extracts 6 common factors through factor analysis,including precipitation factor₁F,wind intensity factorF_{2 1},wind direction factorF₂₂,wind resistance factor₃F,wire stress surface factor₄F,and the duration factor₅F.Extracting common factor can reduce the correlation between the independent variables while preserving the original information as much as possible.Introducing the improved particle swarm optimization algorithm optimized by genetic mutation operator,this paper establishes a prediction model of extreme learning machine based on improved particle swarm optimization algorithm?IPSO-ELM?,and uses the common factor as an input variable to predict the amount of damge to the power lines along the typhoon route.As the results shown,the prediction model of typhoon damage on power lines based on IPSO-ELM has high accuracy.Compared with ELM and PSO-ELM,IPSO-ELM prediction error reaches the minimum value,which is more suitable for predicting the typhoon damage to the power line along the route.Meanwhile,the effectiveness of factor analysis has also been verified,which can effectively improve the accuracy of model prediction.Finally,based on the above research,this paper proposes relevant countermeasures and suggestions for strengthening the power line's resistance to typhoon and improving the efficiency of post-disaster repair.