Article Preview
Top1. Introduction
Engineering industries today are witnessing the growth and development of hard and difficult -to-cut materials. These materials possess superior mechanical properties such as high specific shear strength, very high corrosion, wear and fatigue strength combined to weight ratio (Arunachalam & Mannan, 2000). Inconel 718 is one of the alloys falls in this category having its ability to withstand several operations in gas turbines, airplane and defense parts. Ball end milling is used to generate complex and contoured surfaces on the above parts. Dimensional accuracy and the surface quality of the ball end milled parts developed on the machining parameters employed during the process. The appropriate and wise selection of the process parameters is a challenging task to achieve the desired accuracy with higher productivity. This can be possible through the use of proper optimization techniques for selecting suitable process parameters and their ranges while machining (Ezugwu, 2003; Bhopale, 2015; Pawade, 2008; Bhopale & Pawade, 2012).
Many optimization techniques are employed to determine the optimal set of process parameters in machining. They include genetic algorithm (GA), grey relational method (GR), artificial neural network (ANN), differential evolutionary algorithm (DE), particle swarm optimization (PSO) and simulated annealing (SA). There are few approaches that can be used to establish mathematical models like statistical, numerical or analytical. Many difficulties are associated with the optimization of manufacturing processes such as multimodality, dimensionality and differentiability. Techniques such as linear programming, dynamic programming, steepest decent, etc. generally fail to solve such large-scale problems especially with non-linear objective functions. Hence, there is a need for efficient and effective optimization techniques. Continuous research is being conducted in this field. Nature-inspired meta-heuristic optimization techniques are proving to be better than the traditional techniques and thus are widely used. The convergence of evolutionary algorithms is better than the traditional optimization algorithms.
Recently, Yildiz (2012), presented a comparison of evolutionary optimization techniques to solve multi-pass turning optimization problems. Population-based algorithms such as cuckoo search algorithm, differential evolution algorithm, particle swarm optimization algorithm, genetic algorithm and simulated annealing have been preferred in many applications instead of conventional techniques (Chen & Tsai, 1996; Arun & Raymohan, 2012). The population-based algorithms may have premature convergence towards a local minimum. To find a remedy to the mentioned weakness, they have been integrated with other techniques.