Predictive Modeling and Optimization of Surface Roughness and Cutting Zone Temperature in Turning of Hardened Steel Using RSM, ANN, Genetic Algorithm, and Particle Swarm Optimization

Abstract

The purpose of this research is to establish a combined surface roughness and cutting zone temperature study for modeling and optimizing cutting parameters for turning operations of D3/1.2080 steel with coated carbide under cryogenic cooling and MQL conditions. Cutting speed, feed, and depth of cut are investigated as input factors, while surface roughness and cutting zone temperature are chosen as responses. The experiment has been conducted using the Box-Behnken design (BBD) with three levels and three parameters. Response surface methodology (RSM) and Artificial neural network (ANN) techniques are used for modeling to perform prediction. Where ANN outclassed RSM. Then RSM is coupled with Genetic algorithm (GA) and Particle swarm optimization (PSO). RSM-GA and RSM-PSO approaches are used for optimization. It has been discovered that combining RSM with PSO yields superior outcomes. Finally, a comparison between cryogenic cooling and MQL is done. MQL showed supremacy over cryogenic for both responses.

Cite as

Sajjad Hossain, & Inzamam Ul Haq. (2023). Predictive Modeling and Optimization of Surface Roughness and Cutting Zone Temperature in Turning of Hardened Steel Using RSM, ANN, Genetic Algorithm, and Particle Swarm Optimization. Research and Applications of Thermal Engineering, 6(3), 13–26. https://doi.org/10.5281/zenodo.10251145