Research On Estimation Model Of Pollution Flashover Voltage Of DC Transmission Insulators Based On Artificial Pollution Test

With the development of social economy,there are more and more pollutants in the atmosphere,and the pollution degree on the surface of the insulator also increases.There is a strong electrostatic adsorption effect during DC transmission,which makes the pollution degree on the surface of the insulator more serious than that of AC.All of this has a great impact on the safe and stable operation of the DC system.Therefore,the estimation of the pollution flashover voltage of the DC insulator is an effective method to prevent pollution flashover accidents.The method of equivalent artificial pollution test is most commonly used in estimation of pollution flashover voltage,but with the increase of the voltage level of DC transmission engineering,the artificial pollution test of extra high voltage level in the laboratory is limited.In addition,the pollution flashover voltage has a certain randomness and in order to obtain its statistical law,a large number of tests are required,which greatly increases the difficulty and cost of the test.Therefore,it is of great engineering significance to estimate the pollution flashover voltage of long strings of insulators by using the test of pollution flashover voltage of short strings of insulators through the estimation model.In this paper,the pollution test of the FXBW-10/70,XP-160,LXZP-210 insulators were carried out in the artificial fog chamber.The variation of pollution flashover voltage and pollution flashover voltage gradient of glass,ceramics and composite insulators with salt density and the relationship between the string length and contamination flashover voltage under different salt density were analyzed.On this basis,through the pollution flashover data,the quantitative relationship between the pollution flashover voltage of insulators under different salt density and the flashover voltage of insulators under saturated salt density was derived,and the model of estimating the flashover voltage of insulator under saturated salt density is given by the flashover voltage at a given salt density.In addition,in order to reduce the error caused by the direct linear estimation between insulators with different string length,an improved estimation based on the effective creepage coefficient was established to estimate the pollution flashover voltage of long string insulators.Through the optimization of BP neural network by genetic algorithm,the optimal umbrella model in the improved algorithm was extracted,and the estimation model was verified by the pollution flashover voltage data of composite insulator with different string length obtained by artificial pollution test,which proves the accuracy of the method.The research work in this paper gives an estimation model of the low-salt pollution flashover voltage of the insulator to the pollution flashover voltage under the saturated salt density;the estimation model of the short string insulator pollution flashover voltage to the long string insulator pollution flashover voltage.It provides a theoretical reference for the estimation of insulator flashover voltage.