DEVELOPMENT OF AN OPTIMIZATION MODEL FOR TURNING PROCESS USING AN INVERSE MODEL NEURAL NETWORK

Abstract

Neural networks are now a major interest as when it is connected to computer it mimics the brain. These have shown promise for solving many combinatorial optimization problem. Neural networks have become increasingly popular because their speedy computation , they also provide close optimal solution which are sometimes not possible by other computation methods. Here a feedforward neural network scheme is used to synthesize optimal inputs ( cutting speed, feed rate) is used for turning process. The inverse model neural network is implemented in a multilayer feedforward neural network. Optimization of turning process parameter based on a cost/quality performance index is a key component in machining process planning . Neural network produces the desired result (i.e... cutting speed, feed rate for minimum cost) for a single pass turning in a fraction of second irrespective of the complexity of the problem. In this dissertation work, an inverse model neural network has been developed to optimize machining parameters for single pass finish turning