XU Han, WANG Weizhe, HAN Wen

(School of Mechanical and Electrical Engineering, Jingdezhen Ceramic Institute, Jingdezhen 333403, Jiangxi, China)

Abstract: Aiming at the problem of control and prediction of edge damage in ceramic body milling process, prediction model of ceramic blank edge damage based on BP neural network is established through the mapping relationship between edge damage size and milling parameters. Orthogonal experiments of milling speed, feed speed, cutter cone angle, milling depth and milling width are designed. It is demonstrated that the prediction model has high accuracy, with a prediction error of less than 10%. Also, the milling width and depth have been found to show the greatest impact on edge damage in milling.
Key words: ceramic body; edge damage; prediction model; BP neural network; orthogonal experiment