Study On The Data Analysis And Processing Method For AC Contactor Life Test

AC contactor is one of low voltage control electric that used to connect and break the AC circuits frequently.The life test of the AC contactor has the characteristics of long cycle and high testing expenses,if the data processing method can be used to predict the residual life of the product under the condition of less test data,the test time can be effectively shortened.And wavelet neural network is a new kind of data processing tools,has a fast convergence rate,high fault-tolerance and ability to learn etc.Therefore,the research of applying wavelet neural network to predict AC contactor electric life has great significance.In this paper,the contact resistance and arc energy are explained how to affect the electrical life of ac contactor based on the data analysis and processing.Using piecewise average method to deal with contact resistance and accumulative method to deal with the arc energy,can find that with the increase of operations,contact resistance tends to increase,and the maximum phase of arc energy accumulation increases linearly.Then to contact resistance and arc energy as input,wavelet neural network is adopted to establish the single and double parameters life prediction model,to estimate the electrical life of ac contactor.Prediction error are within the acceptable range,and double parameter prediction model prediction accuracy is higher than single parameter life prediction model,double parameters life prediction model prediction error of less than 2%.