Intelligent parameter identification of machining Ti64 alloy

In this paper, a methodology approach based on analysis of multidimensional Pareto front is proposed. A new optimization approach helps the user to set the optimal parameters of a machining process. Four neural networks are used to model desire output responses, and they are used as objective functions. Particle swarm optimization (PSO) is used to find the best parameters that improve process. As application of approached proposed, an analysis of a multidimensional Pareto front is made considering a minimization of time, temperature, vibration, and surface roughness in a milling process of Ti64 alloy. Physical parameters for experimental approach are tool diameter, number of cutting edge of the tool, cutting speed, feed, and depth of cut. Analysing the 2D and 3D multidimensional Pareto front is generated a user table of machining parameters.