一类带约束动态多目标优化问题的进化算法

动态多目标约束优化问题是一类NP-Hard问题,定义了动态环境下进化种群中个体的序值和个体的约束度,结合这两个定义给出了一种选择算子.在一种环境变化判断算子下给出了求解环境变量取值于正整数集Z+的一类带约束动态多目标优化问题的进化算法.通过几个典型的Benchmark函数对算法的性能进行了测试,其结果表明新算法能够较好地求出带约束动态多目标优化问题在不同环境下质量较好、分布较均匀的Pareto最优解集.     

A single front genetic algorithm for parallel multi-objective optimization in dynamic environments

This paper proposes a new parallel evolutionary procedure to solve multi-objective dynamic optimization problems along with some measures to evaluate multi-objective optimization in dynamic environments. These dynamic optimization problems appear in quite different real-world applications with actual socio-economic relevance. In these applications, the objective functions, the constraints, and hence, also the solutions, can change over time and usually demand to be solved online whilst the size of the changes is unknown. Although parallel processing could be very useful in these problems to meet the solution quality requirements and constraints, to date, not many parallel approaches have been reported in the literature. Taking this into account, we introduce a multi-objective optimization procedure for dynamic problems that are based on PSFGA, a parallel evolutionary algorithm previously proposed by us for multi-objective optimization. It uses an island model where a process divides the population among the remaining processes and allows the communication and coordination among the subpopulations in the different islands. The proposed algorithm makes an exclusive use of non-dominating individuals for the selection and variation operator and applies a crowding mechanism to maintain the diversity and the distribution of the solutions in the Pareto front. We also propose a model to understand the benefits of parallel processing in multi-objective problems and the speedup figures obtained in our experiments.     

多目标演化算法的收敛性研究

基于群体搜索的演化算法求解多目标优化问题有独特的优势,多目标演化算法已有的研究大多为算法的设计和数值试验效果的比较,理论研究往往被忽视．该文讨论了多目标演化算法的收敛性问题,针对一种网格化的简单易于实现的多目标演化算法模型定义了多目标演化算法强收敛和弱收敛等概念,给出了判断算法收敛性的一般性条件;在变异算子为高斯变异、目标函数连续的条件下,证明了提出的算法强收敛．数值实验验证了算法的可行性和有效性．     

Evolutionary Algorithms for Multi-Objective Optimization: Performance Assessments and Comparisons

Evolutionary techniques for multi-objective(MO) optimization are currently gainingsignificant attention from researchers invarious fields due to their effectiveness androbustness in searching for a set of trade-offsolutions. Unlike conventional methods thataggregate multiple attributes to form acomposite scalar objective function,evolutionary algorithms with modifiedreproduction schemes for MO optimization arecapable of treating each objective componentseparately and lead the search in discoveringthe global Pareto-optimal front. The rapidadvances of multi-objective evolutionaryalgorithms, however, poses the difficulty ofkeeping track of the developments in this fieldas well as selecting an existing approach thatbest suits the optimization problem in-hand.This paper thus provides a survey on variousevolutionary methods for MO optimization. Manywell-known multi-objective evolutionaryalgorithms have been experimented with andcompared extensively on four benchmark problemswith different MO optimization difficulties.Besides considering the usual performancemeasures in MO optimization, e.g., the spreadacross the Pareto-optimal front and the abilityto attain the global trade-offs, the paper alsopresents a few metrics to examinethe strength and weakness of each evolutionaryapproach both quantitatively and qualitatively.Simulation results for the comparisons areanalyzed, summarized and commented.     

用多目标演化优化算法解决约束选址问题

约束选址问题是一个多目标约束优化问题,传统算法（加权法）一次只能得到一个候选解,用多目标演化优化算法对其进行求解,可以一次得到多个候选解,给决策者提供更多的选择余地,以期获得更大的利益,数字试验表明,该方法优于传统多目标优化方法。     

Plateaus can be harder in multi-objective optimization

In recent years a lot of progress has been made in understanding the behavior of evolutionary computation methods for single- and multi-objective problems. Our aim is to analyze the diversity mechanisms that are implicitly used in evolutionary algorithms for multi-objective problems by rigorous runtime analyses. We show that, even if the population size is small, the runtime can be exponential where corresponding single-objective problems are optimized within polynomial time. To illustrate this behavior we analyze a simple plateau function in a first step and extend our result to a class of instances of the well-known SetCover problem.     

A Study on Distribution Preservation Mechanism in Evolutionary Multi-Objective Optimization

This paper reviews a number of popular distribution preservation mechanisms and examines their characteristics and effectiveness in evolutionary multi-objective (MO) optimization. A conceptual framework consisting of solution assessment and elitism is presented to better understand the search guidance in evolutionary MO optimization. Simulation studies among different distribution preservation techniques are performed over fifteen representative distribution samples and the performances are compared based upon two distribution metrics proposed in this paper. The results and findings reported in this paper are valuable for better understanding of the working principle and characteristics of distribution preservation mechanisms, which are very useful for incorporating different distribution preservation features into MO evolutionary algorithms in a modular fashion or improving the effectiveness of existing preservation approaches.     

Minimum spanning trees made easier via multi-objective optimization

Many real-world problems are multi-objective optimization problems and evolutionary algorithms are quite successful on such problems. Since the task is to compute or approximate the Pareto front, multi-objective optimization problems are considered as more difficult than single-objective problems. One should not forget that the fitness vector with respect to more than one objective contains more information that in principle can direct the search of evolutionary algorithms. Therefore, it is possible that a single-objective problem can be solved more efficiently via a generalized multi-objective model of the problem. That this is indeed the case is proved by investigating the computation of minimum spanning trees.This work was supported by the Deutsche Forschungsgemeinschaft (DFG) as part of the Collaborative Research Center Computational Intelligence (SFB 531) and by the German-Israeli Foundation (GIF) in the project Robustness Aspects of Algorithms.     

求解多目标优化问题的改进蚁群算法

蚁群算法是一种模拟蚂蚁行为进行优化的启发式优化算法,该算法在许多领域已经得到应用.针对多目标优化问题优化与求解较困难的问题,提出一种嵌入变尺度算法的改进蚁群算法用于求解,为蚁群算法在连续空间中的应用提供了怂一个可行的方案.给出了该算法的详细定义及实现步骤,实例仿真表明,该算法能加快收敛速率,对连续空间的蚁群算法研究具有重要的意义.     

优化设计中的多目标进化算法

近十多年来多目标进化算法是人工智能领域的一个相当活跃的研究热点。该文从非Pareto方法、基于Pareto方法及贝叶斯多目标优化算法等角度对当今多目标进化算法进行了分析,归纳了新出现的各种方法和技术,探讨了这个领域发展中存在的问题,并进一步给出了发展方向。此外文中分别对后两类提出了解决一般问题的计算效果较好的改进算法和新的算法。     

基于密集度的搅动多目标演化算法

多目标演化算法的研究目标是使算法种群快速收敛并均匀分布于问题的非劣最优域.定义和使用密集度来保持群体中个体的均匀分布,将个体的Pareto强度值和密集度合并到个体的适应值定义中.提出搅动策略,以提高算法对解空间的遍历性,从而较大程度上避免算法的早熟,对每次搅动得到的部分非劣解个体进行邻域搜索以加快非劣解前沿的进化.最后,测试函数的实验结果表明了算法的可行性和有效性.     

一种基于决策图贝叶斯网络的强度Pareto进化算法

提出了一种基于决策图贝叶斯网络的强度Pareto进化算法,该算法把贝叶斯概率模型结合到多目标进化算法中,通过构造和学习网络来替代传统进化算法中的交叉重组和变异等遗传操作,避免对大量参数的人工设置和重要构造块的破坏．求解多目标背包问题的仿真结果表明,所提算法可以快速收敛到较好的Pareto前沿,有很强的鲁棒性．     

基于进化计算的轨道转移时间-能量优化方法

针对空间飞行器轨道转移的时间.能量优化问题,提出了一种基于进化计算的多目标优化方法.该方法在非支配解排序和密度估计的基础上,设计了一种新的选择算子从父代中选择进入繁殖池的个体,并使用外部集合保存进化过程所得的非支配解.实验结果表明,该方法可以有效求解优化目标存在约束的轨道转移时间一能量优化问题,并显著提高Pareto前沿的散布性能.     

演化多目标优化中的几何热力学选择

热力学遗传算法(Thermodynamical Genetic Algorithms,TDGAs)借鉴热力学中的自由能极小过程来统一处理多目标优化在逼近性和多样性两方面的任务.为提高TDGA的运行效率和解集分布均匀性,提出了一种几何热力学选择.在该选择中首先定义角度熵通过扇形采样来度量种群逼近方向的多样性.然后利用距离精英定义距离能量来度量种群的逼近程度,避免了耗时的非劣分层操作.此外,引入分量热力学替换规则以较低计算代价驱动种群的几何自由能快速下降.在多目标0/1背包问题上的实验结果表明,几何热力学选择极大地提高了TDGA的运行效率和解集分布均匀性;采用该选择的TDGA算法可生成与NSGA-II在逼近性和分布多样性上性能相当的解,但在运行效率上明显优于NSGA-II.     

基于粗糙集理论与差分进化的混合多目标优化算法

针对传统算法求解约束多目标优化所得近似解精度不高、分布性能不好的问题,提出一种基于粗糙集理论与差分进化的混合算法.首先利用多目标差分进化生成一个初始的近似 Pareto 前沿;然后利用粗糙集理论提高Pareto 前沿的分布质量.选取一组标准的多目标约束测试问题,采用混合算法与 NSGA-II 算法进行仿真求解,对比结果表明,所提出的算法在求解约束多目标优化问题时具有更好的近似解分布和更优越的近似解性能.     

基于演化算法实现多目标优化的岛屿迁徙模型

多目标演化算法(MOEA)利用种群策略，尽可能地找出多目标问题的Pareto最优集供决策者选择，为决策者提供了更大的选择余地，与其它传统的方法相比有了很大的改进．但提供大量选择的同时，存在着不能为决策者提供一定的指导性信息，不能反映决策者的偏好，可扩展性差等问题．本文提出了一个新的多目标演化算法(MOEA)计算模型…岛屿迁徙模型，该模型体现了一种全新的多目标演化优化的求解思想，对多目标优化问题的最优解集作了新的定义．数值试验结果表明，岛屿迁徙模型在求解MOP时有效地解决了以上问题，并且存在进一步改进的潜力．     

基于正交设计的多目标演化算法

提出一种基于正交设计的多目标演化算法以求解多目标优化问题(MOPs)．它的特点在于：(1)用基于正交数组的均匀搜索代替经典EA的随机性搜索,既保证了解分布的均匀性,又保证了收敛的快速性;(2)用统计优化方法繁殖后代,不仅提高了解的精度,而且加快了收敛速度;(3)实验结果表明,对于双目标的MOPs,新算法在解集分布的均匀性、多样性与解精确性及算法收敛速度等方面均优于SPEA;(4)用于求解一个带约束多目标优化工程设计问题,它得到了最好的结果——Pareto最优解,在此之前,此问题的Pareto最优解是未知的．     

并行多目标遗传算法在控制系统优化中的应用

为了提高复杂控制系统设计问题中的效率,提出了一种基于主从模型的并行多目标遗传算法的优化器模型。主进程进行各类遗传操作和最优排序操作,所有进程都进行目标函数值和约束函数值的运算操作,并采用动态负载平衡策略。将该优化器应用在飞行器控制系统设计中,显示出了该算法的优良效果。     

环境友好过程的建模和多目标优化

以环境和经济为目标的过程设计是过程系统工程的重要研究内容。为有效地实施环境友好过程设计,有必要从系统的角度来考虑和研究过程建模和优化问题。本文从物质变化、能量利用、环境影响、经济效益等角度对过程结构进行重新分析,得出多目标问题宏结构,并对其7种模块建立了线性模型。提出了以环境影响最小和经济效益最大为目标的多目标优化模型,多目标模型的求解采用搜索寻优和多目标决策相结合的求解策略。最后以反应系统为例,对环境友好过程建模和优化进行实例研究,