摘要稳健参数设计(Robust Parameter Design, RPD)作为持续性质量改进活动的重要技术，在产品或过程的优化设计中得到了广泛地应用，并且有效地提高了产品的质量。随着产品复杂程度以及顾客需求层次的变化，在实际的产品或者工艺过程中，往往需要考虑多个质量特性，因此多响应的稳健参数优化设计在质量改进活动中显示出越来越重要的地位和作用。
本文首先利用熵权理论对优化目标的均值满意度函数和方差满意度函数所占权重进行了客观性的赋值，将满意度函数与双响应曲面方法相结合构建了改进的满意度函数，然后运用混合遗传算法对所构建的非线性函数进行参数优化，最后结合具体的实例验证了所提方法的有效性。64476
毕业论文关键词 多响应稳健设计；满意度函数；熵权理论；双响应曲面；混合遗传算法
毕 业 论 文 外 文 摘 要
Title The Multi-response Robust Optimization Design Based on Desirability Function
Abstract Robust Parameter Design (RPD)，as an important technology of continuous quality improvement activity, has been applied to the optimization design of products or procedure, and it effectively improves product quality. As product complexity levels and customer demands change, in actual products or process engineering, multiple quality characteristics should often be considered. Therefore, multi-response optimization design of RPD plays a very important part in quality improvement activity.
In this article it is presented that Entropy Weight Theory is used to conduct objective valuation on the weight of optimization objective’s Typical-value Desirability Function and Variance Desirability Function; an improved Desirability Function is formulated by combining Desirability Function approach with dual-response surface approach; and Hybrid Genetic Algorithm is used to conduct optimization on various kinds of complicated nonlinear functions, which, with specific examples, will eventually test and verify the validity of the proposed approach.
Keyword Multi-response robust design; Desirability function; Entropy weight theory; dual-responses surface method; Hybrid genetic algorithm
1引言 1
1.1 多响应优化设计概述 1
1.2.1 满意度函数 3
1.2.2 双响应曲面方法 4
1.2.3 其他方法 4
1.3 研究意义 6
2 多响应稳健参数设计的基本理论与方法 6
2.1满意度函数法 7
2.2 双响应曲面法 9
2.3混合遗传算法 10
3 基于改进满意度函数的稳健参数设计 11
3.1 设定方案，收集数据进行计算 11
3.3 构建基于双响应曲面模型的满意度函数 12
3.4 利用熵权理论方法确定优化目标的权重 13
3.5 利用混合遗传算法进行优化 15
4 案例分析 16
4.1 关于胶态气体泡沫的特性问题 基于改进理想函数多响应稳健参数设计:http://www.751com.cn/guanli/lunwen_71665.html