为了解决产品质量特性优化设计过程中设计参数发生变差时,最优解变为违反约束条件的不可行解的问题,对产品质量特性的稳健性和稳健性度量进行了分析,建立了基于最小灵敏域估计的产品质量特性稳健优化数学模型,提出了多目标产品质量特性稳健优化设计流程。根据该模型,采用广义差分多目标进化算法求得多目标质量特性稳健优化问题的Pareto稳健优化解集,并使用基于模糊集合理论的方法选择出最优解。在满足产品质量特性设计要求的前提下,优化产品质量特性,提高产品质量特性性能对设计参数变差的稳健性。最后,应用该方法对高速工业平缝机挑线机构进行质量特性稳健优化设计,证明了该方法在工程应用中的正确性与高效性。
In product quality characteristics design optimization,it didn't consider the effects of parameter variations which might lead to change from feasible solutions to infeasible solutions.A mathematical model of product quality characteristics robust optimization design was constructed based on Minimum Sensitivity Region Estimation(MSRE).An optimization process using Generalized Differential Evolution algorithm(GDE)was put forward based on the analysis and measurement of product quality characteristics sensitivity region.Based on the mathematical model of product quality characteristics robust optimization design,the Pareto set for the multiobjective optimization problem was obtained using GDE.An approach based on fuzzy set theory was developed to extract one of the multi-objective robust Pareto-optimal solutions as the best compromise one.The multiobjective robust Pareto-optimal solutions were satisfied with the design requirements and insensitive to parameter variations.The efficiency and effectiveness of the proposed method were illustrated by the product quality characteristics robust optimization design of thread-taking-up mechanism of sewing machine.