Application of differential evolution to tolerance design

Most of the optimization algorithms in machining operations deal with the methods, which derive the exact optimal machining variables for an optimal manufacturing process design. Actually, the exact design is not easy to be set because of multi-constraints, like variables bounds and such as machining quality and manufacturing cost. In addition, most of these variables obtained from the optimal methods would not be robust subject to environment disturbances. In real situation, a robust solution has to perform and maintain its functions in routine circumstances, as well as hostile or unexpected circumstances. It would be reasonably that it should be a feasible solution even when some specific tolerances are imposed on. In this paper, a new robust optimization approach, which combines the differential evolution algorithm with the Taguchi outer array, is proposed to solve the problem of optimization for the surface grinding process when considering the tolerance design of machining variables. The purpose of this paper is to use the proposed new approach, named the IDETA, optimize the grinding variables when considering the tolerance design, simultaneously subject to a comprehensive set of multi-constraints. Simultaneously, a distinctive characteristic of reliability has been defined and handled as a measure of robustness for the variables.