Particle Swarm Optimization for Milling Titanium Alloy
- PDF / 265,623 Bytes
- 6 Pages / 432 x 648 pts Page_size
- 106 Downloads / 254 Views
Particle Swarm Optimization for Milling Titanium Alloy I. Escamilla1, L. Torres1, B. Gonzalez1, P. Zambrano1 1 Facultad de Ingeniería Mecánica y Eléctrica. Ave. Universidad s/n. San Nicolás de los Garza, N. L. [email protected] ABSTRACT Optimum machining parameters are of great concern in manufacturing environments, where economy of machining operation plays a key role in competitiveness in the market. Many researchers have dealt with the optimization of machining parameters for milling operations. In this paper, optimization procedures based on particle swarm optimization algorithm are developed for find machining parameters in milling operation. It describes development and utilization of the methodology that determines optimum Pareto’s front analyzing feed, speed and depth for milling operation. The relationships between machining parameters and the performance measures of interest are obtained by using experimental data and a swarm intelligent neural network system. Results show that particle swarm optimization is an effective method for solving multi-objective optimization problems, and also, that an integrated system of neural networks and swarm intelligence can be used to solve complex machining optimization problems. Keywords: Multi-objective optimization, particle swarm optimization, machining parameters, Titanium Alloy INTRODUCTION Determining and optimizing the parameters involved in a machining process is a critical and very important task. Machining of Titanium into finished components is receiving increasing attention in especially aerospace and automotive industries. Titanium is also one of the most expensive metals to produce, which is another drawback to manufacturing parts using this metal. Surface finish is an important parameter in the machining process. Surface quality is one of the most specified customer requirements. A major indication of surface quality on machined parts is surface roughness. Thus machinists and companies specializing in the machining of titanium materials generally develop techniques to maximize surface integrity of titanium alloys. Thus optimum properties usually are achieved during the production machining of titanium. [1, 2]. Neural networks (NNs) are computational intelligent methods, which can establish a mapping through the numerical inputs and outputs. The NNs extract a sensitive and realistic relation from some experimental input–output data, called training set. In this way, they can interpolate synthetic data that estimate the results of the experiments that have not been established, and predict optimum reasonable processing conditions [3]. Particle Swam Optimization (PSO) is an evolutionary computation (EC) method inspired by flocking birds. This population based stochastic optimization technique has been developed by Eberhart and Kennedy [4] and applied to many different systems including machining. Coello and Lechuga [5] find that PSO is an efficient alternative over other stochastic and population-based search algorithms, especially when dealing
Data Loading...
