ON-LINE CUTTING TOOL CONDITION MONITORING IN TURNING PROCESSES USING ARTIFICIAL INTELLIGENCE AND VIBRATION SIGNALS

摘要

This study deals with developing an artificial neural network (ANN) cutting force and surface roughness prediction model as a function of cutting parameters and vibration signals in the turning process of AISI 4140 steel. An experimental turning dataset is used to train and evaluate the model. Input dataset includes cutting speed, feed rate, depth of cut, vibration levels along the three axes on the tool holder (ax, ay, az). The Output dataset includes cutting force (Fc) and surface roughness (Ra). A comparison between the predicted force and, surface roughness with their experimental counterparts shows an excellent agreement. The accuracy between the experimental and the predicted values is as high as 99.95%. The results show that the model can reliability and accurately be used to predict cutting force and surface roughness as a function of cutting parameters and tool vibrations.