A levy flight particle swarm optimizer for machining performances optimization

Abstract

Machining processes has been used widely in manufacturing industry and manufacturers have realized the important of these processes to improve the machining performance that would lead to an increase in production. However, one of the problems identified is how to minimize the values of machining performance in terms of surface roughness (Ra), tool wear (VB) and power consumption (Pm). To provide better machining performance, it is essential to optimize cutting parameters which are cutting speed (V), feed rate (f) and cutting time (T). This research has developed a hybridization technique using particle swarm optimization (PSO) and Levy flight labeled as Levy flight particle swarm optimizer (LPSO) aimed at optimizing the cutting parameters to obtain minimum values of machining performance for a specific machining performance such as turning process. The simulation results obtained were compared with particle swarm optimization (PSO), regression analysis (RA), response surface method (RSM), artificial neural network (ANN) and support vector regression (SVR) and validated using regression model, analysis of variance (ANOVA) and determination of optimum level for each machining performance. The results showed that the LPSO could minimize the values of Ra, VB and Pm nearly 95% in comparison to the other research techniques listed in this research. The LPSO technique could minimize the values of machining performance substantially for the manufacturing industry.