Multi-objective optimization of hard turning of AISI 6150 using PCA-based desirability index for correlated objectives

摘要

The turning process, one of the most popular material removal processes in industry, has several performance measures which are usually found to be correlated, such as tool wear, cutting force and surface finish. In order to apply optimization methods, such as the desirability index, the conditional independence assumption is usually made. However, this assumption rarely holds true in real world applications and the optimal solution obtained might be biased towards the performance measures which have strong positive correlations with the others. Despite the fact that the desirability index has been developed and frequently applied in industry for a long time, only a few studies have been carried out to solve optimization problems with correlated objectives. The modified desirability index which provides a solution for integrating the expert's preferences and the correlation information of the performance measures into the overall performance index, the principal component analysis (PCA) based desirability index (DI), has been only recently developed. In this paper, an optimization using the PCA-based DI is demonstrated based on empirical models of hard turning of AISI 6150 steel in which uncertainties are propagated by model errors. The results show that the degree of importance of each performance measure has been adjusted by the integration of the covariance information into the overall performance index.