Optimisation of machining parameters using Hopfield-type neural networks

The variables affecting the economics of machining operations are numerous and include machine tool capacity, work piece geometry, cutting conditions such as velocity, feed rate, depth of cut, etc. Optimum selection of cutting conditions importantly contributes to the increase of productivity and the reduction of costs. Therefore, in recent years, more attention has been paid to the problem of optimum selection of cutting conditions for multi-pass operations. In this paper, a solution approach based on minimum unit cost criterion is proposed for this problem. The objective of the problem is to minimise unit production cost without violating any technological, economical and organisational constraints. A Hopfield-type dynamical network which employs a penalty function approach is used for solving the problem formulated by mixed integer linear programming. The results of the proposed approach tested on an illustrative example show that the approach provides better or least the same unit costs compared to existing approaches. Since the proposed approach is both effective and efficient, it can be integrated into an intelligent process planning system for solving complex machining parameters optimisation problems. The optimal solution of the proposed approach makes it attractive and suitable for the determination of optimum cutting conditions where there is no enough time for deep analysis.