Artificial Neural Network and Genetic Algorithm-Based Models for Predicting Cutting Force in Turning of Hardened H13 Steel

Abstract

The manufacturing sector in the modern era is striving hard to reduce the cost of production by employing innovative techniques. One such technique is hard turning where the workpiece is heat treated to the requisite hardness, and the final size and shape of the component are obtained directly through hard turning process. Hard turning is generally carried out with a huge amount of cutting fluid to enhance the output performance. Since petroleum-based emulsions are easily available in the market at reasonable price, they are widely used in industries. Petroleum-based cutting fluids create a number of environmental and health issues. In this perspective, pure dry turning is a logical substitute as it does not possess the harmful effects connected with the cutting fluids. The feasible tool life and surface quality are often disturbed while carrying out the machining operation under pure dry condition. Under such circumstances, the concept called minimal cutting fluid application (MCFA) performs itself as a possible solution. This paper investigates the effect of applying cutting fluid using MCFA technique at the critical contact zones while hard turning of H13 steel. An artificial neural network (ANN) model was developed for the prediction of the main cutting force, and its ability to predict cutting force (F z) was analyzed. An effort is made to optimize the cutting parameters to accomplish minimum cutting force using genetic algorithm.