针对传统的多响应优化方法很少同时考虑到优化结果的最优性与稳健性以及容易忽视各个响应质量损失的问题,通过引入双响应曲面法的思想,提出了一种基于改进质量损失函数的多响应稳健优化设计新方法。通过双响应曲面法与多变量质量损失函数的相互结合,同时考虑了各个响应的均值和方差,首先通过拟合各个响应的均值回归方程和方差回归方程并将其结合到多变量质量损失函数中,然后通过主客观权重法科学设置各响应均值和方差质量损失的权重,最后将多响应的稳健优化问题转化为求解综合质量损失最小值的问题。以细胞注射3自由度柔顺并联微动平台中的双曲柔性铰链为研究对象进行研究,结果表明,该方法能够有效地解决具有多个响应问题的稳健优化,为多响应稳健优化设计提供新的思路。
Because the conventional multi-response optimization methods rarely consider the optimization and the robustness simultaneously and tend to overlook the quality loss of multiple responses, we propose a new method of multi-response robust optimization based on the improved quality loss function to achieve the robustness and optimization of multiple responses by introducing the dual response surface method. With the dual response surface method and the multivariate loss function combined, the new method considers both the mean and variance of multiple responses. Firstly, the regression equation of the mean and standard deviation of multiple responses are fitted and are incorporated into the multivariate quality loss function. Secondly, the subjective and objective weight theory is used to set the weights of quality losses of mean and standard deviations of each response. Finally, the multi-response robust and optimal problem is transformed into the minimization of overall quality loss. The hyperbolic flexure hinge in the cell injection 3-DOF compliant parallel micro motion platform is used to prove the effectiveness of the new method. The analysis results show that the new method can more effectively solve multi- response robust optimization problems.