基于主成分分析的多响应稳健性优化方法研究

Study on Multi-Response Robust Design Based on Principal Component Analysis

将主成分分析法与双响应曲面法相结合,提出了一种同时优化响应均值和标准差的多响应优化方法。通过主成分分析,将多个响应的位置和散布特性值转化为主成分综合得分,并以二者的加权和作为优化指标,克服了传统的多响应优化方法只考虑响应均值的不足,将响应的标准差也纳入优化范畴,实现优化的稳健性。实例表明,该方法充分考虑了响应标准差对优化结果的影响,优化结果得到一定改善。

An approach to simultaneously optimize both the mean and standard deviation of multiple responses is proposed by combining principal component analysis and dual response surface method. The proposed approach aims to identify the settings of input variables to optimize the multi-response performance index which is the weighted sum of principal values of both the location and dispersion of all the responses. The proposed method successfully achieves the robustness,overcomes the limitation of some multi-response approaches, which typically ignore the standard deviation of the responses. The proposed method is illustrated through an example from the literature.