Artificial Neural Network on Tool Condition Monitoring in Hard Turning of AISI4140 Steel Using Carbide Tool

Hard turning has replaced conventional grinding in production processes recent years as an emerging technique. Nowadays, coated carbide tools are replacing expensive CBN inserts turning. Wear is a significant concern when with carbide; it immediately affects the acceptability of machined surface, which causes machine downtime and loss due to wastage parts. Online tool condition monitoring (TCM) required prevent such critical conditions. differs from energy balance during metal cutting, resulting greater thrust force; hence, TCM model presented for may not be suitable hard Hence, wear prediction projected based on force using artificial neural network (ANN). All tests were done design experiments called full factorial (FFD). The specimens made AISI 4140 steel that had been hardened 47 HRC, carbide. most impactful input features wear, selected experimental outputs, given trained. Tool estimated output training set validated satisfactory results random 5–10–1 structure Levenberg–Marquardt (LM) learning algorithm, R2 values 0.996602 0.969437 testing data, mean square error 0.000133152 0.004443 respectively, gave best results. MEP 0.575407 2.977617 very low (5%). LM algorithm-based ANN good at predicting how well predicts both set.