FBF-NN-based modelling of cylindrical plunge grinding process using a GA

In order to make a machining process cost effective as well as to assure the desired objective(s), it is important to find the optimal machining parameter(s) in considering all related output variable(s). However developing many models (each one corresponding to a single output variable) leads to more time consuming as well as difficult to interact them. Moreover it has been observed that some of the output variables are inter-related with each other. In this paper, an intelligent approach based on fuzzy basis function neural network (FBF-NN) is proposed to model the cylindrical plunge grinding process. Two approaches are adopted here for automatic design of rule base (each rule is characterized by the antecedent and consequent parts) of FBF-NN using a genetic algorithm. In the first approach the rule-consequent parts are determined individually and in the second approach, those are obtained based on the inter-relationship exist among the output variables. The results of the FBF-NNs and that obtained using the empirical expressions are compared with the real experimental values, which shows that the FBF-NN-based models give better predictions than mathematical models.