Fuzzy logic and neural networks for design of process parameters: A grinding process application

Abstract

The design of a grinding process is a difficult task since there are so many characteristics to consider. In this study, a generic scheme to establish the norm for automation of design by employing fuzzy logic and neural networks for a surface grinding process is proposed. Design of a grinding process is accomplished by initial determination of a set of optimal design variables in order to achieve a set of desired process variables. First, the important features of a surface grinding process are identified. Next, advisory systems for surface grinding design are reviewed. After that, a 'fuzzy grinding optimizer' (FGO) and a 'neural grinding optimizer' (NGO) are proposed. In addition, a generic scheme called 'bi-directional construction of fuzzy and neural systems' (BCFNS) is proposed for performance evaluation and comparison between fuzzy logic and neural networks. Finally, future research directions are pointed out concerning performance evaluation for various types of grinding optimizers.