Development of Fuzzy Logic Tool to predict Wear Rate in Aluminum Composite

Abstract

Industrial automation is the need of the hour. Success and failure of automation depends on the selection and efficient utilization of soft computing tools such as artificial neural network, expert systems, and fuzzy logic. This article explores the idea of constructing a fuzzy logic model to predict wear rate of fabricated composites under predetermined conditions of input variables such as applied load, sliding velocity and distance travelled. Taguchi design of experiments is employed for experimentation. 27 sets of experiments are performed by varying the parameters of applied load, sliding velocity and distance travelled. Experiment is repeated for average values of wear rate are measured to reduce experimental error. Matlab toolbox functions are used to build the model. Confirmatory experiments are done and the results are measured in terms of accuracy and time. It is found that the developed model predicts wear rate with acceptable limit. Hence the constructed model can be forwarded to predict wear rate of the developed composite under different conditions.