Establishment and Research of Prediction Model of Discharge Gap Voltages in Discharge Arc Milling

It was difficult to directly measure the discharge gaps during arc milling. The changes of the discharge gaps were judged by the changes of the discharge gap voltages. The system identification theory was used to determine the structure and model parameters of the discharge gap voltage prediction model. The method for establishing the prediction model was elaborated, and the degree of fit of the prediction model was verified by experiments. The experimental results show that the fit accuracy of the prediction model decreases with the increase of the fitting time. Therefore, the recursive least squares method was used to perform the online discharge gap voltage measurements. It is predicted that the average error of online prediction is as 6.82%. The results show that the discharge gap voltage prediction model may predict the discharge gap voltage stably and effectively one step ahead, with fewer model identification parameters and high prediction accuracy. © 2022 China Mechanical Engineering Magazine Office. All rights reserved.