Fuzzy logic to predict thermal damages of ground parts

Abstract

One of the critical problems in implementing an intelligent grinding process is the automatic detection of workpiece surface burn. This work uses fuzzy logic as a tool to classify and predict burn levels in the grinding process. Based on acoustic emission signals, cutting power, and the mean-value deviance (MVD), linguistic rules were established for the various burn situations (slight, intermediate, severe) by applying fuzzy logic using the Matlab Toolbox. Three practical fuzzy system models were developed. The first model with two inputs resulted only in a simple analysis process. The second and third models have an additional MVD statistic input, associating information and precision. These two models differ from each other in terms of the rule base developed. The three developed models presented valid responses, proving effective, accurate, reliable and easy to use for the determination of ground workpiece burn. In this analysis, fuzzy logic translates the operator's human experience associated with powerful computational methods.