Authors: Nafis Ahmad; Tomohisha Tanaka; Yoshio Saito
Addresses: Department of Industrial and Production Engineering, Bangladesh University of Engineering and Technology, Dhaka-1000, Bangladesh ' Department of Mechanical and Control Engineering, Tokyo Institute of Technology, Tokyo 152-8550, Japan ' Department of Mechanical and Control Engineering, Tokyo Institute of Technology, Tokyo 152-8550, Japan
Abstract: Machining operations are used for creating surfaces by cutting away unwanted materials from workpieces. These operations are highly constrained and non-linear in nature. As a result traditional techniques are not suitable for machining parameter optimisation. Turning and milling are the two most commonly used machining operations where machining time or cost is minimised by optimising cutting parameters. The important constraints are maximum cutting force, machine power, available rotational speed, tool deflection, required surface finish cusp height etc. Here, a genetic algorithm (GA) and artificial neural network (ANN)-based hybrid approach is presented. The proposed approach gives more emphasis on searching optimum cutting parameters near boundaries of feasible and infeasible solution spaces. The optimum solution obtained by this method also does not violate constraints for a specific machining operation. An example of ball end milling operation is presented to explain this technique.
Keywords: machining parameters; cutting parameters; ball end milling; genetic algorithms; GAs; artificial neural networks; ANNs; parameter optimisation.
International Journal of Data Analysis Techniques and Strategies, 2014 Vol.6 No.3, pp.261 - 274
Available online: 25 Jun 2014Full-text access for editors Access for subscribers Purchase this article Comment on this article