全球技术与优化杂志

Statistical analyses and regression modeling for influence of process parameters on material removal rate in ultrasonic machining

Abstract

Gaoyan Zhonga*, Jiangyan Xua, Yuetong Wub, Shoufeng Yangc

The traditional regression model in machining process neglects nonlinear influence of machining parameters on process response, which causes the analyses to have a low accuracy. The primary objective of this study is to propose an optimal regression model to analyze the material removal rate in ultrasonic machining through the experimental tests, statistical analyses and regression modeling. Three main factors affecting the machining process response, namely abrasive granularity, feed pressure and feed speed, were selected for this purpose, and the experiments were performed in accordance with an L16 orthogonal array using Taguchi method. Analysis of variance (ANOVA) was used to investigate the statistical significance of the parameters at 95% confidence level and to determine the percentage contribution of the parameters to the process response. On this basis, the optimal regression model was proposed. Compared with traditional regression model, the analytical precision of the optimal regression model is quite higher than that of traditional regression model. The results obtained from the new experimental conditions show that the optimal regression model can correctly reflect the influence of machining parameters on process response, which can provide a theoretical basis for selection of machining parameters to improve its machining efficiency.