基于满意度函数的多响应稳健优化模型及实证研究

针对传统的综合满意度模型在研究多响应时较少考虑噪声因素的问题,本文通过引入噪声因素对传统的满意度模型实施改进,进行多响应稳健优化分析。首先,构建包含可控因素和噪声因素的多质量特性的响应曲面模型;其次,借助传统综合满意度模型的构建方法,将多质量特性的响应曲面模型整合为改进的满意度模型;最后,将信噪比作为衡量改进的满意度模型稳健优化的指标,得到稳健优化的参数组合。实证研究表明:构建同时包含可控因素和噪声因素的改进的满意度模型是可行的,在此模型的基础上利用信噪比能够有效地找到多响应稳健优化的参数组合。     

多响应优化方法的比较和应用研究

本文对目前几种多响应优化的主要方法进行了分析和总结,考察了马氏距离法、多元损失函数法、满意度函数法、概率法以及双响应曲面等几种主要方法在多响应优化中的应用,并对实例的多响应的预测值和方差进行了优化,比较了这几种方法的优缺点。研究结果表明,满意度函数法和概率法在处理多响应优化过程中能够取得较好的优化效果。     

响应建模法在改进产品设计中的应用

本文指出双重响应曲面方法的一些局限性,介绍了响应建模法在参数设计中的应用及其优点,也指出了响应建模法与双重响应曲面方法的联系。本文还通过一个实例说明响应建模法能从试验数据中获取更多的信息。我们认为响应建模法能在改进产品的参数设计起到重要作用,具有一些双重响应曲面方法不能替代的优点。     

基于模糊逻辑推理的改进多响应参数优化方法研究

研究一种基于模糊逻辑推理的改进多响应参数优化方法.通过满意度函数求解各个响应的满意度值,然后经过模糊逻辑推理将多个响应的满意度值转化为模糊推理等级(FRG).模糊推理规则可以充分考虑实际工艺操作经验和专家知识,利用响应间相互关系的机理和统计信息,得到的模糊推理等级能够客观反映多个质量特性的整体水平.将基于模糊逻辑推理的改进多响应参数优化方法运用到铣切削生产过程的参数优化设计中,使两个响应的粗糙程度均有所下降,验证了方法的有效性.     

循环搜索法在三参数双指数曲面拟合中的应用

在最小二乘意义下用 Gauss-Newton法拟合三参数双指数曲面z =γ[exp( -αx) +exp( -βy) ]时 ,充分利用观测值 ,应用循环搜索法确定了参数初始值 .并成功地以由 Meyer &Roth给出的观测值为例对此进行了验证 .     

变步长非单调模式搜索法

A varied steplength nonmonotone pattern search method is proposed in this paper. The varied steplength search strategy is designed in this method such that the pattern direction is more approximated to efficient descent direction. The interpolation and nonmonotone technique are used for improving local search and global convergence. The theoretical and numerical results show that this method is an efficient direct search method.     

基于模糊推理的改进多响应参数稳健优化方法

提出一种基于模糊逻辑推理的改进多响应稳健参数优化方法,建立模糊逻辑推理系统,由多个响应的满意度值推算出一个模糊推理等级.拟合含有噪声因子的多响应过程的神经网络预测模型,可以预测参数优化结果,减少实验量.通过案例分析验证了所提稳健参数优化方法的有效性.     

参数化方法在解多目标优化中的应用

求解多目标优化的 参数化方法本质上是将多目标评价函数中的权系数视为可变参数。本文从一般的含参数的优化问题出发，论述了最优解连续依赖于参数的变化。本文的数值例子将表是，采用这种处理方法，可达到人们的预期目的。     

非单调多步曲线搜索方法的收敛性

提出一种求解无约束优化问题的非单调多步曲线搜索方法.此方法具有如下特点:(1)算法在产生下一个迭代点时不仅利用了当前迭代点的信息,而且还可能利用前m个迭代点的信息.这就是多步法;(2)下降方向和步长同时确定,而不是先找到方向,再由线性搜索寻找步长.这就是曲线搜索技术;(3)采用非单调搜索技巧.在较弱的条件下,我们证明了此方法的收敛性.     

多值函数在复变函数中的应用

通过讨论初等多值函数的单值解析分枝问题,重点研究了初等多值函数在复变函数中的具体应用.     

Robust optimization for multiple responses using response surface methodology

Typically in the analysis of industrial data for product/process optimization, there are many response variables that are under investigation at the same time. Robustness is also an important concept in industrial optimization. Here, robustness means that the responses are not sensitive to the small changes of the input variables. However, most of the recent work in industrial optimization has not dealt with robustness, and most practitioners follow up optimization calculations without consideration for robustness. This paper presents a strategy for dealing with robustness and optimization simultaneously for multiple responses. In this paper, we propose a robustness desirability function distinguished from the optimization desirability function and also propose an overall desirability function approach, which makes balance between robustness and optimization for multiple response problems. Simplex search method is used to search for the most robust optimal point in the feasible operating region. Finally, the proposed strategy is illustrated with an example from the literature. Copyright © 2009 John Wiley & Sons, Ltd.     

基于模矢搜索和遗传算法的混合约束优化算法

近年,免梯度方法又开始引起大家的注意，由于不需要计算函数的梯度．特别适合用来求解那些无法得到梯度信息或需要花很大计算量才能得到梯度信息的问题．本文构造了一个基于模矢搜索和遗传算法的混合优化算法．在模矢搜索方法的搜索步，用一个类似于遗传算法的方法产生一个有限点集．算法是全局收敛的.     

A Filter-Based Pattern Search Method for Unconstrained Optimization

We discuss a filter-based pattern search method for unconstrained optimization in this paper. For the purpose to broaden the search range we use both filter technique and frames, which are fragments of grids, to provide a new criterion of iterate acceptance. The convergence can be ensured under some conditions. The numerical result shows that this method is practical and efficient.     

基于引力搜索和粒子群混合优化算法的证券投资组合问题研究

本文研究考虑交易成本的投资组合模型,分别以风险价值(VAR)和夏普比率(SR)作为投资组合的风险评价指标和效益评价指标。为有效求解此模型,本文在引力搜索和粒子群算法的基础上提出了一种混合优化算法(IN-GSA-PSO),将粒子群算法的群体最佳位置和个体最佳位置与引力搜索算法的加速度算子有机结合,使混合优化算法充分发挥单一算法的开采能力和探索能力。通过对算法相关参数的合理设置,算法能够达到全局搜索和局部搜索的平衡,快速收敛到模型的最优解。本文选取上证50股2014年下半年126个交易日的数据,运用Matlab软件进行仿真实验,实验结果显示,考虑交易成本的投资组合模型可使投资者得到更高的收益率。研究同时表明,基于PSO和GSA的混合算法在求解投资组合模型时比单一算法具有更好的性能,能够得到满意的优化结果。     

Experimental Testing of Advanced Scatter Search Designs for Global Optimization of Multimodal Functions

Scatter search is an evolutionary method that, unlike genetic algorithms, operates on a small set of solutions and makes only limited use of randomization as a proxy for diversification when searching for a globally optimal solution. The scatter search framework is flexible, allowing the development of alternative implementations with varying degrees of sophistication. In this paper, we test the merit of several scatter search designs in the context of global optimization of multimodal functions. We compare these designs among themselves and choose one to compare against a well-known genetic algorithm that has been specifically developed for this class of problems. The testing is performed on a set of benchmark multimodal functions with known global minima.     

BFGS算法对非凸函数优化问题的收敛性

ＢＦＧＳ算法是无约束最优化中最著名的数值算法之一,对非凸函数ＢＦＧＳ算法是否具有整体收敛性,这是一个ｏｐｅｎ问题,本文考虑Ｗｏｌｆｏ线搜索下目标函数非凸的ＢＦＧＳ算法,我们给出一个使该算法收敛的充分条件。     

光线寻优算法的寻优机理分析

Based on Fermat’s principle and the automatic optimization mechanism in the propagation process of light,an optimal searching algorithm named light ray optimization is presented,where the laws of refraction and reflection of light rays are integrated into searching process of optimization.In this algorithm,coordinate space is assumed to be the space that is full of media with different refractivities,then the space is divided by grids,and finally the searching path is assumed to be the propagation path of light rays.With the law of refraction,the search direction is deflected to the direction that makes the value of objective function decrease.With the law of reflection,the search direction is changed,which makes the search continue when it cannot keep going with refraction.Only the function values of objective problems are used and there is no artificial rule in light ray optimization,so it is simple and easy to realize.Theoretical analysis and the results of numerical experiments show that the algorithm is feasible and effective.     

Optimization of generalized desirability functions under model uncertainty

Desirability functions are increasingly used in multi-criteria decision-making which we support by modern optimization. It is necessary to formulate desirability functions to obtain a generalized version with a piecewise max type-structure for optimizing them in different areas of mathematics, operational research, management science and engineering by nonsmooth optimization approaches. This optimization problem needs to be robustified as regression models employed by the desirability functions are typically built under lack of knowledge about the underlying model. In this paper, we contribute to the theory of desirability functions by our robustification approach. We present how generalized semi-infinite programming and disjunctive optimization can be used for this purpose. We show our findings on a numerical example. The robustification of the optimization problem eventually aims at variance reduction in the optimal solutions.