Experimental Investigation and Optimization of Machining Parameters in Milling of Al6351 Using Hybrid – Artificial Bee Colony Algorithm

Volume 9 Issue 3 May - July 2019

Research Paper

Experimental Investigation and Optimization of Machining Parameters in Milling of Al6351 Using Hybrid – Artificial Bee Colony Algorithm

P. Hema*, G. Padmanabhan**, T. Eswar***

*_*** Department of Mechanical Engineering, S. V. University college of Engineering, Tirupati, Andhra Pradesh, India.

Hema, P., Padmanabhan, G., and Eswar, T. (2019). Experimental Investigation and Optimization of Machining Parameters in Milling of Al6351 Using Hybrid – Artificial Bee Colony Algorithm. i-manager’s Journal on Mechanical Engineering, 9(3),9-18. https://doi.org/10.26634/jme.9.3.16059

Abstract

Simplifying any process of machining is a profoundly difficult, since it basically includes forecasts of ideal cutting parameters and working requirements that are unpredictable and extremely non-linear in nature which influence the overall production costs and workpiece quality. One of the Nature Inspired Algorithms (NIA) is Artificial Bee Colony (ABC) algorithm for process optimization that imitates honey bees intelligent foraging behavior. This paper describes an experimental study of cutting parameters optimization like Cutting Speed, Feed rate, Aluminum Alloy 6351 response depth cut by using Swarm-based optimization. Based on Taguchi design of experiments L18 orthogonal array is selected with three levels of input parameters at various machining conditions. The experiments are performed and predicted the responses like Surface Roughness, Material Removal Rate, Resultant Forces and Temperature. A recent evolutionary heuristic swarm intelligence algorithm called the Hybrid Artificial Bee Colony (HABC) is used to optimize conventional milling processes. This algorithm is used to minimize responses by estimating the optimum parameters of the process. Comparison of the results with the Harmony Search Algorithm (HSA), Particle Swarm Optimization (PSO) and Genetic Algorithm (GA) are done to examine the performances of various methods. The results suggest that the HABC algorithm outperforms the solution's HSA, PSO and GA quality. Additionally, Multi-Objective Optimization is performed and a combined normalized objective function (Z) is formulated by considering equal weightages to all the objectives. The optimized values of milling parameters are obtained through the HABC algorithm. Confirmatory experiments reveal that the experimental values are moderately close with optimized values.