鲁棒线性优化问题研究综述

鲁棒优化（ＲＯ）是从计算复杂性的角度研究不确定优化模型鲁棒最优解的数学方法．从单阶段鲁棒优化和多阶段鲁棒优化两个方面对鲁棒线性优化（ＲＬＯ）理论的研究进展进行综述,前者的研究主要基于不同形式的不确定集合,后者的研究则基于前者的方法．研究多阶段不确定决策中决策变量受不确定参数实现值影响的情况,其核心是影响函数连续时的仿射可调鲁棒对应模型和函数离散时的有限适应性模型．最后对ＲＬＯ 的研究前景作了展望．

     

一类不确定线性系统的鲁棒线性控制器设计

针对一类不确定性不满足匹配条件的线性系统,利用Lyapunov方程和不确定项的范数界,分别设计了具有可调参数和鲁棒线性状态和输出控制器。这些可调参数可以依据不确定项的范数界的大小来选取,具有一定的灵活性。     

一类线性不确定组合系统的鲁棒控制器和观测器设计*

本文研究了一类带有结构不确定性的线性组合系统的鲁控制器和观测器设计问题，文中给出一种设计分散控制器和分散观测器的方法，其中状态反馈增益 观测器增益阵由两个黎卡提议方程的解给出。     

线性不确定系统的鲁棒D——稳定性分析

基于一种Ｇｕａｒｄｉａｎ映射，将具有结构不确定性的线性系统在指定扇区的鲁棒Ｄ－稳定性问题转化成相应参数矩阵的非奇异性问题，得到了系统极点保持在左半平面某一扇区中的Ｄ－稳定界，同时得到了系统的鲁棒稳定界。该界不再是系统的二次稳定或Ｄ－稳定界，因而非常接近系统的实际稳定与Ｄ－稳定界。最后给出了一种鲁棒Ｄ－稳定性的分析方法，并通过实例加以说明。     

线性时滞不确定系统的强稳定鲁棒控制器设计

研究了线性时滞不稳定系统的强稳定鲁棒控制器设计的设计问题，其中不稳定性是时变的，不要求满足匹配条件，主要通过黎卡方程方法给出强稳定鲁棒控制器存在的充分条件，并具例说明方法的有效性。     

带有非线性不确定参数的线性系统的鲁棒稳定性和鲁棒镇定问题

研究带有非线性不确定参数的线性系统的鲁棒稳定性和鲁棒镇定问题。讨论一种有很强实际应用背景并允许带有二次不确定参数的模型,研究该系统的鲁棒稳定性和鲁棒镇定问题。以ＬＭＩ的形式给出了判据,并举例证明了该方法的优越性。     

不确定线性时滞互联大系统的鲁棒镇定

本文考虑了一类不确定线性时滞互联大系统，基于LIM方法给出了系统可鲁棒镇定的充分条件，利用MATLAB的工具箱可以方便的给出该问题的解，具有一定的理论及实际价值。     

一类线性时变不确定周期奇异系统的鲁棒镇定

讨论了一类线性时变不确定周期广义系统的鲁棒镇定问题．基于线性时变不确定周期广义系统的鲁棒稳定的概念, 提出鲁棒稳定的充分必要条件, 并基于对偶系统的等价性得到鲁棒镇定的充分必要条件. 通过引入自由矩阵, 所得结果表示为线性矩阵不等式, 验证过程更简单、可靠.     

城市道路建设时序决策的鲁棒优化

为提高城市道路建设时序决策的鲁棒性,提出了城市道路建设时序决策优化的双 层规划模型。模型假定出行需求在一定范围内扰动,上层规划是在有限资金的约束下寻求各建设阶段的系统总出行时间与系统总出行时间对出行需求的灵敏度之间的综合最小值,下层规划为各建设阶段的随机用户均衡配流。文中推导出了系统总出行时间对出行需求灵敏度的计算式,并给出了模型的求解算法。最后以一个测试路网为例,对基于系统总出行时间、基于灵敏度、基于系统总出行时间与灵敏度综合出行时间的决策优化模型进行了计算分析,结果显示3种决策优化模型均可寻求到各自目标最优的城市道路建设时序,但在需求不确定的情景下基于灵敏度、基于系统总出行时间与灵敏度综合出行时间的决策优化结果更具鲁棒性。     

一类不确定线性时滞系统的输出反馈鲁棒镇定

研究一类不确定线性时滞系统的输出反馈鲁棒镇定问题,其中不确定性不必满足匹配条件。以二次Lyapunov泛函保证系统的渐近稳定性,利用线性矩阵不等式给出了系统可以利用动态输出反馈鲁棒镇定的充分条件。当此条件成立时,基于线性矩阵不等式的解构造了全阶动态输出反馈镇定控制器。     

Energy supply planning and supply chain optimization under uncertainty

In energy supply planning and supply chain design, the coupling between long-term planning decisions like capital investment and short-term operation decisions like dispatching present a challenge, waiting to be tackled by systems and control engineers. The coupling is further complicated by uncertainties, which may arise from several sources including the market, politics, and technology. This paper addresses the coupling in the context of energy supply planning and supply chain design. We first discuss a simple two-stage stochastic program formulation that addresses optimization of an energy supply chain in the presence of uncertainties. The two-stage formulation can handle problems in which all design decisions are made up front and operating parameters act as ‘recourse’ decisions that can be varied from one time period to next based on realized values of uncertain parameters. The design of a biodiesel production network in the Southeastern region of the United States is used as an illustrative example. The discussion then moves on to a more complex multi-stage, multi-scale stochastic decision problem in which periodic investment/policy decisions are made on a time scale orders of magnitude slower than that of operating decisions. The problem of energy capacity planning is introduced as an example. In the particular problem we examine, annual acquisition of energy generation capacities of various types are coupled with hourly energy production and dispatch decisions. The increasing role of renewable sources like wind and solar necessitates the use of a fine-grained time scale for accurate assessment of their values. Use of storage intended to overcome the limitations of intermittent sources puts further demand on the modeling and optimization. Numerical challenges that arise from the multi-scale nature and uncertainties are reviewed and some possible modeling and numerical solution approaches are discussed.     

Multi-period portfolio optimization with linear control policies

This paper is concerned with multi-period sequential decision problems for financial asset allocation. A model is proposed in which periodic optimal portfolio adjustments are determined with the objective of minimizing a cumulative risk measure over the investment horizon, while satisfying portfolio diversity constraints at each period and achieving or exceeding a desired terminal expected wealth target. The proposed solution approach is based on a specific affine parameterization of the recourse policy, which allows us to obtain a sub-optimal but exact and explicit problem formulation in terms of a convex quadratic program.In contrast to the mainstream stochastic programming approach to multi-period optimization, which has the drawback of being computationally intractable, the proposed setup leads to optimization problems that can be solved efficiently with currently available convex quadratic programming solvers, enabling the user to effectively attack multi-stage decision problems with many securities and periods.     

The use of metamodeling techniques for optimization under uncertainty

Metamodeling techniques have been widely used in engineering design to improve efficiency in the simulation and optimization of design systems that involve computationally expensive simulation programs. Many existing applications are restricted to deterministic optimization. Very few studies have been conducted on studying the accuracy of using metamodels for optimization under uncertainty. In this paper, using a two-bar structure system design as an example, various metamodeling techniques are tested for different formulations of optimization under uncertainty. Observations are made on the applicability and accuracy of these techniques, the impact of sample size, and the optimization performance when different formulations are used to incorporate uncertainty. Some important issues for applying metamodels to optimization under uncertainty are discussed.     

Delay-independent robust stability of uncertain linear systems

In this paper we present some new sufficient conditions for robust stability of linear systems with uncertain time delay as well as structured parameter uncertainty. The stability robustness bounds obtained from the new sufficient conditions are independent of the size of time delay. Some illustrative examples are given.     

不确定条件下控制分配问题的鲁棒优化方法

针对具有冗余执行机构的过驱动系统, 在考虑控制效率不确定性的条件下, 提出了一种基于鲁棒优化理论的控制分配算法. 研究了原始不确定鲁棒优化模型的建立和基于椭球不确定集的鲁棒对等式的转化问题, 并推广到可由锥二次不等式表示的不确定集的情况. 讨论了鲁棒优化控制分配算法的求解方法及其计算复杂度. 最后, 针对多操纵面飞机的最优控制分配问题与传统算法进行了仿真比较, 结果表明鲁棒优化算法能有效降低控制效率不确定性的影响, 使分配结果更为合理, 从而具有更好的鲁棒性, 同时能有效提高操纵面故障情况下闭环系统的控制重构能力, 很好地改善了飞控系统的性能.     

Data envelopment analysis and robust optimization: A review

This paper reviews the milestone approaches for handling uncertainty in data envelopment analysis (DEA). This paper presents the detailed classifications of robust data envelopment analysis (RDEA). RDEA is appropriate for measuring the efficiencies of decision-making units in the presence of the data and distributional uncertainties. This paper reviews scenario-based and uncertainty set of DEA models. It covers 73 studies from 2008 to 2019. The paper concludes with suggestions about the guidelines for future researches in the field of RDEA.     

电梯群的可调整鲁棒优化调度

为解决电梯群控调度(GES)中乘客交通流不确定问题, 提出基于可调整鲁棒优化的电梯群控调度方法. 基于对电梯交通流的不确定特性分析, 建立了电梯群控调度的不确定优化模型. 利用可调整鲁棒优化方法将电梯群控调度的不确定模型转化为其可调整鲁棒对等式. 在此基础上, 证明了在不确定集为椭球集直积时, 电梯群控调度模型的可调整鲁棒对等式(ARC)是可计算的. 仿真验证表明, 与其他的调度方法相比, 该方法具有较好的调度性能, 提高了调度对不同乘客交通流模式的适应性.     

基于在线优化的线性系统状态反馈鲁棒镇定

建立关于Sylvester方程的鲁棒极点配置梯度流优化算法模型,在线求解相应的二次优化问题,并在线计算状态反馈增益矩阵.使得闭环系统矩阵的伞部特征值位于给定的区域中.对于一切满足条件的扰动,具有最小灵敏度,闭环系统人范围一致渐近稳定.仿真结果验证了该方法适用十非线性系统的鲁棒镇定问题和在线鲁棒极点配置问题.     

蚁群优化算法在求解随机组合优化问题中的应用综述*

不确定条件下的优化问题更贴近真实世界环境,因而日益受到广泛关注。综述了蚁群优化在求解一组不确定条件下的组合优化问题,即随机组合优化问题方面的应用。首先介绍了不确定条件下组合优化问题的概念分类模型,给出了随机组合优化问题的一般定义;然后指出了其与求解传统确定性组合优化问题的不同之处,即目标函数的计算存在不确定性,并详细论述了目前解决方法的进展;最后分析了该领域值得重点关注的几个研究方向,并对其未来发展进行了展望。     

Tail mean and related robust solution concepts

Robust optimisation might be viewed as a multicriteria optimisation problem where objectives correspond to the scenarios although their probabilities are unknown or imprecise. The simplest robust solution concept represents a conservative approach focused on the worst-case scenario results optimisation. A softer concept allows one to optimise the tail mean thus combining performances under multiple worst scenarios. We show that while considering robust models allowing the probabilities to vary only within given intervals, the tail mean represents the robust solution for only upper bounded probabilities. For any arbitrary intervals of probabilities the corresponding robust solution may be expressed by the optimisation of appropriately combined mean and tail mean criteria thus remaining easily implementable with auxiliary linear inequalities. Moreover, we use the tail mean concept to develope linear programming implementable robust solution concepts related to risk averse optimisation criteria.