Research Of Transmission Line Electric Power Fittings Wear In Xinjiang Wind Region

With the rapid development of power industry,xinjiang region assumes the task of "xinjiang electricity delivery",however,affected by bad weather such as strong wind,sandstorm occur,transmission line hardware winds in xinjiang region have serious wear,affect the entire power grid security.Wear is a complex process,affected by a variety of nonlinear,strong coupling factors,the process of friction pair,working parameters,environment and wear time all have relations.In this paper,for the more real and effectively research transmission line hardware wear problem,first,in the form of test area according to the wind the actual situation of transmission line to design test,have the different load and wear number of u-shaped ring wear test,Using metallo graphic and scanning electron microscope and micro hardness analysis study the reason and mechanism of the u-shaped ring wear.The wear rate of the u-rings is roughly constant,no accelerated wear;The upper U-shaped ring is more likely to wear than the lower U ring.0.8 t,150000 times wear of u-shaped ring residual strength had less than half of the original state,wear in the middle region with extrusion shearing action have obvious deformation,hardness decreased by 38%;The ring wear of U-shaped ring is mainly adhesion wear.under SEM there are obvious layers feature.Using finite element simulation software of ABAQUS build the model of u-shaped ring friction and wear,and simulates the u-shaped ring movement form and the load,combin accumulation failure of materials and Johnson-cook constitutive model to realize the u-shaped ring wear simulation.Through the model found that the maximum contact stress changing in the process of the relative friction and the change range is larger,the biggest effect point always appeared at the area of contact surface.The model simulates can be more effective in reflecting the wear of U-shaped ring,the error between simulate and test is 7.5%.Finally,using SPSS statistical analysis software and neural network model to predict the wear of U ring.Both models have good fitting results,averaging 1.86% and 1.45% respectively.Because of the complexity of the wear itself and randomness,two models predict results have slightly error,the neural network model average predict deviation is 6.0 KN,but individual point prediction deviation is more;The average deviation of the regression model is 6.3 KN,the error of each point is more average,regression model predicts better stability.