| In the past ten years,China’s high-speed railway has developed rapidly,and a number of trunk high-speed railways,such as Harbin-Dalian high-speed railway and Lanzhou-Xinjiang high-speed railway,have been connected and planned in frigid region.With the increasing coverage of high-speed railway in frigid region,it is crucial to ensure the safety of high-speed railway operation,which means higher requirements should be put forward for the power supply reliability of high-speed railway traction power supply system.The traction power supply system is the only power source of high-speed railway,whose central equipment is the traction transformer.The attached high voltage bushing of the transformer plays an important role in leading high and low voltage leads from the inside of the traction transformer and the external electrical network.Therefore,the reliable power supply of the traction power supply system depends on high voltage bushings’ safe operation.In order to ensure the safe and reliable operation of traction power supply system in frigid environment and to provide suggestions for the maintenance of power equipment in specific areas,this paper studied the development law of partial discharge and evaluated the severity of partial discharge in oilpaper insulated bushings at low temperature.In order to simulate the operation condition of oil-paper insulated bushings in high-speed railway in frigid region,a partial discharge test platform at low temperature was established.The development process of partial discharge in oil paper insulation was studied from three aspects: average discharge magnitude,maximum discharge magnitude and pulse repetition rate.Meanwhile,the development process of partial discharge was divided into three stages based on the variation trend of those three aspects.Then,the influence mechanism of temperature and moisture content on partial discharge development of oil-paper insulation at low temperature and the difference of partial discharge development process of bushings with different defect types were analyzed by control variable method.As the PRPD spectrum has evolved by the development of partial discharge characteristics,based on feature parameters extracted by gray gradient co-occurrence matrix and color moment,this paper proposed a whale optimization-extreme learning machine partial discharge stage identification method.The final recognition effect was compared with that of extreme learning machine,support vector machine and back propagation neural network,which meant using whale optimization algorithm to optimize extreme learning machine can effectively identify the partial discharge stage of oil paper insulation with different insulation defect types at different environmental temperature.According to model identification results,the severity of partial discharge of oil-paper insulated bushings in high-speed railway in frigid region was evaluated,which provided suggestions for later bushings maintenance plan. |
