Research On Lightning Warning Method For Traction Power Supply System Of High Speed Railway

The high-speed railway traction power supply system is vulnerable to the impact of lightning.Lightning strike on the catenary can lead to power supply equipment damage,power failure,train outage,etc.,causing property losses of railway operation departments and passengers,and even threatening the personal safety of passengers.The safe and reliable operation of high-speed railway traction power supply system is directly related to the stability of railway operation.In order to ensure the safe operation of high-speed railway,it is very important to predict the lightning disaster effectively.In this paper,the probability of lightning disaster along the high-speed railway is analyzed and calculated based on Bayesian network.And on this basis,the corresponding lightning protection measures are proposed.Firstly,according to the structure of traction power supply system and characteristics of lightning activities,analyzing the impact of lightning disaster on high-speed railway traction power supply system.This paper establishes the model of high-speed railway traction power supply system using the PSCAD,to simulate the changes of Current and Voltage of high-speed railway traction network.Secondly,based on Bayesian network,establishing the lightning warning model.This paper selects 40 groups of data to learn the structure and parameters of Bayesian network.At the same time,based on the regression analysis method,establishing the lightning warning model based on Logistic regression,and carrying out its significance.The results show that all warning parameters fail to pass the significance test,which indicates that this model is not suitable for lightning warning of high-speed railway traction power supply system.Thirdly,the lightning warning model based on Bayesian network is analyzed by an example.The causal reasoning and diagnostic reasoning are carried out by using the software of Netica Bayesian network analysis.Then calculating the posterior probability and compared with the prior probability.The results show that Bayesian network model can effectively analyze the complex relationship between lightning activity and influencing factors,and can get the key factors in the lightning early warning.Finally,based on the practical application,exploring some lightning protection measures of high-speed railway traction power supply system.And simulating the changes of lightning Voltage and Current of traction network before and after installing arrester.The results show that the Voltage and Current produced obviously after installing arrester,which indicates that lightning protection effect is remarkable.