全局替换的自适应权重调整MOEA/D

当多目标问题的帕累托前沿形状较为复杂时,基于分解的多目标进化算法MOEA/D的解的均匀性将受到很大的影响. MOEA/D利用相邻子问题的信息来优化,但早期因为种群中的个体与子问题的关联是随机分配的,仅在邻居间更新会浪费优秀解的信息,影响收敛速度.针对这些问题,本文提出一种MOEA/D的改进算法(MOEA/DGUAW).该算法使用种群全局更新的策略,来提高收敛速度;使用自适应调整权重向量的策略来获得更均匀分布的解集.将MOEA/D-GUAW算法与现有的MOEA/D, MOEA/D-AWA, RVEA和NSGA-III算法在10个广泛应用的测试问题上进行了实验比较.实验结果表明,提出的算法在大部分问题上,反转世代距离评价指标IGD优于其他算法,收敛速度也快于其他算法.     

一种改进的基于分解的多目标进化算法

利用基于分解的多目标进化算法框架(MOEA/D),将混合策略的进化算法用于求解分解后的若干单目标优化子问题,提出了一种带局部搜索的基于分解的多目标混合策略进化算法(LMS-MOEA/D)。算法利用均匀设计产生子问题的聚合权重向量,混合交叉策略能够充分利用不同交叉算子的优势;同时算法针对演化过程收敛的特点,结合局部搜索策略,获得逼近Pareto前沿的最优解集。最后通过实验验证算法在多样性和收敛性方面的有效性。     

一种基于MOEA/D的组合权重方法

为了准确地求解组合权重的组合系数,将基于分解的多目标进化算法(multi-objective evolutionary algorithm based on decomposition,MOEA/D)思想引入评估领域,提出一种基于MOEA/D的组合权重方法.通常,利用加权和法将组合权重模型转化为单目标模型时,模型加权系数难以准确确定.对此,引入MOEA/D算法的分解思想,将组合权重模型转化为多个单目标子模型.MOEA/D算法仅适用于无约束优化问题,而较为常用的惩罚函数法难以表达进化初期无可行解的情况,因而提出改进自适应惩罚函数(improved adaptive penalty function,IAPF),将组合权重模型转化为无约束优化模型.应用所提出方法与其他方法进行仿真实验,实验结果表明,所提出算法具有有效性.     

基于幂变换的多目标进化算法MOEA?D权重设计方法

在多目标最优化问题中,如何求解一组均匀散布在前沿界面上的有效解具有重要意义.MOEA?D是最近出现的一种杰出的多目标进化算法,当前沿界面的形状是某种已知的类型时,MOEA?D使用高级分解的方法容易求出均匀散布在前沿界面上的有效解.然而,多目标优化问题的前沿界面的形状通常是未知的.为了使MOEA?D能求出一般多目标优化问题的均匀散布的有效解,利用幂函数对目标进行数学变换,使变换后的多目标优化问题的前沿界面在算法的进化过程中逐渐接近希望得到的形状,提出了一种求解一般的多目标优化问题的MOEA?D算法的权重设计方法,并且讨论了经过数学变换后前沿界面的保距性问题.采用建议的权重设计方法,MOEA?D更容易求出一般的多目标优化问题均匀散布的有效解.数值结果验证了算法的有效性.     

基于参考向量的自适应约束多目标进化算法

针对目前用多目标进化算法(MOEA)处理约束多目标优化问题(CMOP)的研究通常以解决单一类型约束为主,而在面对不同种类的复杂约束时算法难以收敛或者种群分布性差的问题,以基于分解的多目标进化算法(MOEA/D)框架为基础,提出一种基于参考向量的自适应约束多目标进化算法(ARVCMOEA).首先将参考向量分成主参考向量及...     

基于最小距离和聚合策略的分解多目标进化算法

针对基于帕累托（Pareto）支配的多目标进化算法在解决高维问题时选择压力降低,以及基于分解的多目标进化算法在提高收敛性和分布性的同时降低了种群多样性的问题,提出了一种基于最小距离和聚合策略的分解多目标进化算法。首先,使用基于角度分解的技术将目标空间分解为指定个数的子空间来提高种群的多样性;然后,在生成新解的过程中加入基于聚合的交叉邻域方法,使生成的新解更接近于父代解;最后,分两阶段在每个子空间内基于最小距离和聚合策略来选择解以提高收敛性和分布性。为了验证所提算法的可行性,采用标准测试函数ZDT和DTLZ进行仿真实验,结果表明所提算法的总体性能均优于经典的基于分解的多目标进化算法（MOEA/D）、MOEA/D-DE、NSGA-Ⅲ和GrEA。可见,所提算法在提高多样性的同时可以有效平衡收敛性和多样性。     

基于分解的演化多目标优化算法综述

基于分解的演化多目标优化算法(MOEA/D)的基本思想是将一个多目标优化问题转化成一系列子问题(单目标或者多目标)来进行优化求解.自2007年提出以来, MOEA/D受到了国内外学者的广泛关注,已经成为最具代表性的演化多目标优化算法之一.总结过去13年中关于MOEA/D的一些研究进展,具体内容包括:(1)关于MOEA/D的算法改进;(2) MOEA/D在超多目标优化问题及约束优化问题上的研究;(3) MOEA/D在一些实际问题上的应用.然后,实验对比几个具有代表性的MOEA/D改进算法.最后,指出一些MOEA/D未来的研究方向.     

混合分解和强度帕累托多目标进化算法

在多目标进化优化中,使用分解策略的基于分解的多目标进化算法(MOEA/D)时间复杂度低,使用〖BP(〗强度帕累托策略的〖BP)〗强度帕累托进化算法-2(SPEA2)能得到分布均匀的解集。结合这两种策略,提出一种新的多目标进化算法用于求解具有复杂、不连续的帕累托前沿的多目标优化问题(MOP)。首先,利用分解策略快速逼近帕累托前沿;然后,利用强度帕累托策略使解集均匀分布在帕累托前沿,利用解集重置分解策略中的权重向量集,使其适配于特定的帕累托前沿;最后,利用分解策略进一步逼近帕累托前沿。使用的反向世代距离(IGD)作为度量标准,将新算法与MOEA/D、SPEA2和paλ-MOEA/D在12个基准问题上进行性能对比。实验结果表明该算法性能在7个基准问题上最优,在5个基准问题上接近于最优,且无论MOP的帕累托前沿是简单或复杂、连续或不连续的,该算法均能生成分布均匀的解集。     

一种基于偶图匹配的多目标分解进化算法

针对基于分解的多目标进化(MOEA/D)算法在选择下一代解时未考虑解和子问题之间的相对距离,可能导致算法得到的最终解多样性较差的问题,提出一种基于偶图匹配的多目标分解进化(MOEA/D-BM)算法.所提算法利用偶图匹配模型对解和子问题的相互关系进行建模,在选择下一代解的同时,考虑收敛性和多样性,以提高算法性能.通过与其他3种经典的多目标分解进化算法在多个测试函数上进行实验,验证了所提出算法的有效性.     

基于在线感知Pareto前沿划分目标空间的多目标进化优化

多目标进化优化是求解多目标优化问题的可行方法.但是, 由于没有准确感知并充分利用问题的Pareto前沿, 已有方法难以高效求解复杂的多目标优化问题.本文提出一种基于在线感知Pareto前沿划分目标空间的多目标进化优化方法, 以利用感知的结果, 采用有针对性的进化优化方法求解多目标优化问题.首先, 根据个体之间的拥挤距离与给定阈值的关系感知优化问题的Pareto前沿上的间断点, 并基于此将目标空间划分为若干子空间; 然后, 在每一子空间中采用MOEA/D (Multi-objective evolutionary algorithm based on decomposition)得到一个外部保存集; 最后, 基于所有外部保存集生成问题的Pareto解集.将提出的方法应用于15个基准数值函数优化问题, 并与NSGA-Ⅱ、RPEA、MOEA/D、MOEA/DPBI、MOEA/D-STM和MOEA/D-ACD等比较.结果表明, 提出的方法能够产生收敛和分布性更优的Pareto解集, 是一种非常有竞争力的方法.     

On the effect of reference point in MOEA/D for multi-objective optimization

Multi-objective evolutionary algorithm based on decomposition (MOEA/D) has continuously proven effective for multi-objective optimization. So far, the effect of weight vectors and scalarizing methods in MOEA/D has been intensively studied. However, the reference point which serves as the starting point of reference lines (determined by weight vectors) is yet to be well studied. This study aims to fill in this research gap. Ideally, the ideal point of a multi-objective problem could serve as the reference point, however, since the ideal point is often unknown beforehand, the reference point has to be estimated (or specified). In this study, the effect of the reference point specified in three representative manners, i.e., pessimistic, optimistic and dynamic (from optimistic to pessimistic), is examined on three sets of benchmark problems. Each set of the problems has different degrees of difficulty in convergence and spread. Experimental results show that (i) the reference point implicitly impacts the convergence and spread performance of MOEA/D; (ii) the pessimistic specification emphasizes more of exploiting existing regions and the optimistic specification emphasizes more of exploring new regions; (iii) the dynamic specification can strike a good balance between exploitation and exploration, exhibiting good performance for most of the test problems, and thus, is commended to use for new problems.     

Utopian point based decomposition for multi-objective optimization problems with complicated Pareto fronts

Multi-objective evolutionary algorithm based on decomposition (MOEA/D) has been considered as a promising method for solving multi-objective optimization problems (MOPs). It devotes most of its effort on convergence by optimizing a set of scalar optimization subproblems in a collaborative manner, while maintaining the diversity by using a set of uniformly distributed weight vectors. However, more recent studies illustrated that MOEA/D faces difficulties on MOPs with complicated Pareto fronts, mainly because the uniformity of weight vectors no longer lead to an evenly scattered approximation of the Pareto fronts in these cases. To remedy this, we suggest replacing the ideal point in the reciprocal Tchebycheff decomposition method with a more optimistic utopian point, with the aim of alleviating the sensitivity of MOEA/D to the Pareto front shape of MOPs. Experimental studies on benchmark and real-world problems have shown that such simple modification can significantly improve the performances of MOEA/D with reciprocal Tchebycheff decomposition on MOPs with complicated Pareto fronts.     

基于多点非均匀变异的多目标极值优化算法研究

多目标进化算法因其在解决含有多个矛盾目标函数的多目标优化问题中的强大处理能力,正受到越来越多的关注与研究。极值优化作为一种新型的进化算法,已在各种离散优化、连续优化测试函数以及工程优化问题中得到了较为成功的应用,但有关多目标EO算法的研究却十分有限。本文将采用Pareto优化的基本原理引入到极值优化算法中,提出一种求解连续多目标优化问题的基于多点非均匀变异的多目标极值优化算法。通过对六个国际公认的连续多目标优化测试函数的仿真实验结果表明：本文提出算法相比NSGA-II、 PAES、SPEA和SPEA2等经典多目标优化算法在收敛性和分布性方面均具有优势。     

MOEA/D线性插入方向向量策略研究

MOEA/D具有良好的收敛性、均匀的分布性、求解效率高等优点,普遍应用于求解多目标优化问题.然而对于Pareto前端复杂的多目标优化问题,预先设定均匀的权重向量并不能够维持Pareto最优解集的良好分布性.本文,首先分析均匀分布的权重向量、均匀分布的搜索方向二者与均匀分布的解集之间的关系,提出一种新的权重向量设置方式;其次基于进化过程中解集的分布,提出线性插入搜索方向策略,并将其转换为对应的权重向量,同时在MOEA/D中周期性应用该策略调整搜索方向,获取分布均匀的解集;最后将该算法在WFG系列测试问题上进行性能测试,并采用世代距离指标(GD)、Spacing指标(S)、超体积指标(HV)对算法收敛性和多样性进行对比分析,实验结果表明,与原始的MOEA/D、使用均匀分布的搜索方向MOEA/D、使用预处理的M OEA/D、M OEA/D-DU相比,改进的算法求出解集的多样性极大提高,收敛性明显增强,解集的整体质量显著提高.     

MOEA/D with uniform decomposition measurement for many-objective problems

Many-objective problems (MAPs) have put forward a number of challenges to classical Pareto-dominance based multi-objective evolutionary algorithms (MOEAs) for the past few years. Recently, researchers have suggested that MOEA/D (multi-objective evolutionary algorithm based on decomposition) can work for MAPs. However, there exist two difficulties in applying MOEA/D to solve MAPs directly. One is that the number of constructed weight vectors is not arbitrary and the weight vectors are mainly distributed on the boundary of weight space for MAPs. The other is that the relationship between the optimal solution of subproblem and its weight vector is nonlinear for the Tchebycheff decomposition approach used by MOEA/D. To deal with these two difficulties, we propose an improved MOEA/D with uniform decomposition measurement and the modified Tchebycheff decomposition approach (MOEA/D-UDM) in this paper. Firstly, a novel weight vectors initialization method based on the uniform decomposition measurement is introduced to obtain uniform weight vectors in any amount, which is one of great merits to use our proposed algorithm. The modified Tchebycheff decomposition approach, instead of the Tchebycheff decomposition approach, is used in MOEA/D-UDM to alleviate the inconsistency between the weight vector of subproblem and the direction of its optimal solution in the Tchebycheff decomposition approach. The proposed MOEA/D-UDM is compared with two state-of-the-art MOEAs, namely MOEA/D and UMOEA/D on a number of MAPs. Experimental results suggest that the proposed MOEA/D-UDM outperforms or performs similarly to the other compared algorithms in terms of hypervolume and inverted generational distance metrics on different types of problems. The effects of uniform weight vector initializing method and the modified Tchebycheff decomposition are also studied separately.     

A decomposition based memetic algorithm for multi-objective vehicle routing problem with time windows

Multi-objective evolutionary algorithm based on decomposition (MOEA/D) provides an excellent algorithmic framework for solving multi-objective optimization problems. It decomposes a target problem into a set of scalar sub-problems and optimizes them simultaneously. Due to its simplicity and outstanding performance, MOEA/D has been widely studied and applied. However, for solving the multi-objective vehicle routing problem with time windows (MO-VRPTW), MOEA/D faces a difficulty that many sub-problems have duplicated best solutions. It is well-known that MO-VRPTW is a challenging problem and has very few Pareto optimal solutions. To address this problem, a novel selection operator is designed in this work to enhance the original MOEA/D for dealing with MO-VRPTW. Moreover, three local search methods are introduced into the enhanced algorithm. Experimental results indicate that the proposed algorithm can obtain highly competitive results on Solomon׳s benchmark problems. Especially for instances with long time windows, the proposed algorithm can obtain more diverse set of non-dominated solutions than the other algorithms. The effectiveness of the proposed selection operator is also demonstrated by further analysis.     

多目标进化算法及其在控制领域中的应用综述

多目标进化算法在求解多目标优化问题方面具有独特的优势.对此,介绍了多目标进化算法的基本原理,讨论了多目标进化算法的一系列改进方法;论述了近年来多目标进化算法在自动控制领域中的最新研究成果,并对其未来的发展方向进行了展望.     

An adaptive evolutionary multi-objective approach based on simulated annealing

A multi-objective optimization problem can be solved by decomposing it into one or more single objective subproblems in some multi-objective metaheuristic algorithms. Each subproblem corresponds to one weighted aggregation function. For example, MOEA/D is an evolutionary multi-objective optimization (EMO) algorithm that attempts to optimize multiple subproblems simultaneously by evolving a population of solutions. However, the performance of MOEA/D highly depends on the initial setting and diversity of the weight vectors. In this paper, we present an improved version of MOEA/D, called EMOSA, which incorporates an advanced local search technique (simulated annealing) and adapts the search directions (weight vectors) corresponding to various subproblems. In EMOSA, the weight vector of each subproblem is adaptively modified at the lowest temperature in order to diversify the search toward the unexplored parts of the Pareto-optimal front. Our computational results show that EMOSA outperforms six other well established multi-objective metaheuristic algorithms on both the (constrained) multi-objective knapsack problem and the (unconstrained) multi-objective traveling salesman problem. Moreover, the effects of the main algorithmic components and parameter sensitivities on the search performance of EMOSA are experimentally investigated.