ANFIS APPLIED TO THE PREDICTION OF SURFACE ROUGHNESS IN GRINDING OF ADVANCED CERAMICS

摘要

This paper introduces a methodology for predicting thernsurface roughness of advanced ceramics using AdaptivernNeuro-Fuzzy Inference System (ANFIS). To this end, arngrinding machine was used, equipped with an acousticrnemission sensor and a power transducer connected to thernelectric motor rotating the diamond grinding wheel. Thernalumina workpieces used in this work were pressed andrnsintered into rectangular bars. Acoustic emission andrncutting power signals were collected during the tests andrndigitally processed to calculate the mean, standardrndeviation, and two other statistical data. These statistics,rnas well the root mean square of the acoustic emission andrncutting power signals were used as input data for ANFIS.rnThe output values of surface roughness (measured duringrnthe tests) were implemented for training and validation ofrnthe model. The results indicated that an ANFIS network isrnan excellent tool when applied to predict the surfacernroughness of ceramic workpieces in the grinding process.