一种考虑概率分布的鲁棒优化模型  

A robust optimization model considering probability distribution

作　　者：丁然[1] 李歧强[1] 张元鹏[2] 

机构地区：[1]山东大学控制科学与工程学院,济南250061 [2]山东大学电气工程学院,济南250061

出　　处：《中国工程科学》2008年第9期70-73,83,共5页Strategic Study of CAE

基　　金：山东省优秀中青年科学家奖励基金资助(2007BS06074);国家863项目资助(2007AA04Z157)

摘　　要：文章以随机规划中的机会约束思想为指导,根据随机参数的概率分布情况,提出了两种鲁棒性条件约束,并在此基础上建立了一种新的鲁棒优化模型,使模型的可行解控制在一定的鲁棒性指标的范围内。该模型不但可处理约束两端同时含有随机参数的情况,还可以方便地推广到非线性模型中。仿真实例说明了模型的有效性。Robust optimization is a method to process optimization problem under uncertainty. The current robust optimization methods have some deficiencies in application conditions and probability utilization. Based on the chance constraints programming, two kinds of robust constraints according to two different kinds of probability distribution of the stochastic parameters are proposed, and a novel robust optimization model is proposed. The feasible solutions of this model can be controlled to satisfy the robust index. This model can be used in the situations that both sides of the constraints contain stochastic parameters, and can be easily extended to non-liner models. The simulation results illustrate the validity of the model.

关 键 词：不确定性 鲁棒优化 随机规划 机会约束 

分 类 号：TB114.1[理学—运筹学与控制论]