基于两类因子波动下的稳健优化设计

发布时间：2018-01-30 11:25

本文关键词： 噪声因子可控因子波动 Kriging模型稳健优化设计　出处：《工业工程与管理》2017年06期 　论文类型：期刊论文

【摘要】：针对稳健优化设计中噪声因子和可控因子波动对质量特性影响的问题,结合Kriging模型和稳健优化思想,提出了一种同时考虑噪声因子和可控因子波动的稳健优化方法。该方法首先假定噪声因子和可控因子服从正态分布,提出了基于两类因子波动下的稳健设计方法;其次,在融合两类波动信息的空间填充设计基础上,构建Kriging均值模型和标准差模型;然后,基于稳健优化思想,建立优化目标函数;最后,结合经济订购批量(Economic Order Quantity,EOQ)模型和蒙特卡洛仿真,验证所提方法的有效性和稳健性。结果表明,所提方法能有效解决噪声因子和可控因子波动情况下的稳健优化问题。
[Abstract]:Aiming at the influence of noise factor and controllable factor fluctuation on quality characteristics in robust optimization design, Kriging model and robust optimization idea are combined. A robust optimization method considering both noise factor and controllable factor fluctuation is proposed, which assumes that noise factor and controllable factor are obeyed from normal distribution. A robust design method based on two kinds of factor fluctuations is proposed. Secondly, the Kriging mean model and the standard deviation model are constructed on the basis of the space filling design which combines the two kinds of fluctuation information. Then, based on the idea of robust optimization, the optimization objective function is established. Finally, the validity and robustness of the proposed method are verified by combining the economic Order quantity EOQ model with Monte Carlo simulation. The proposed method can effectively solve the robust optimization problem in the case of noise factor and controllable factor fluctuation.
【作者单位】：南京理工大学经济管理学院;南京航空航天大学经济与管理学院;
【基金】：国家自然科学基金资助项目(71471088,71371099) 中央高校基本科研业务专项资金资助项目(3091511102) 江苏省自然科学基金资助项目(BK20170810)
【分类号】：F273
【正文快照】： 1引言稳健参数设计是由日本质量管理专家田口玄一博士于20世纪60年代提出,基本原理是通过选择可控因子的水平组合来减少系统对噪声变化的敏感性,从而达到减少整个系统性能波动的目的[1-3]。Taguchi把试验输入参数分为两大类:一类为可控因子,是指管理者可以控制其水平变化的因

【相似文献】