In this paper, four parameters (cutting depth, feed rate, speed, tool nose runoff) with three levels (low, medium, high) are considered to optimize the tool wear for CNC (computer numerical control) finish turning based on L 9(34) orthogonal array. Additionally, twenty- seven fuzzy control rules using trapezoid membership function with respective to seventeen linguistic grades for tool wear are constructed. Considering thirty input and eighty output intervals, the defuzzification using center of gravity is moreover completed. Through the Taguchi experiment, the optimum fuzzy deduction parameters can then be received. The confirmation experiment for optimum deduction parameters is furthermore performed on an ECOCA-3807 CNC lathe. It is shown that the tool wear ratio from the fuzzy deduction optimization parameters is significantly advanced comparing to that from the benchmark. This paper not only proposes a parametric deduction optimization scheme using orthogonal array, but also contributes the satisfactory fuzzy approach to the tool wear for CNC turning with profound insight.
Journal of Statistics & Management Systems , 14(5), pp.885-898