目标空间映射策略的高维多目标粒子群优化算法

为了平衡优化算法在高维多目标优化问题中收敛性和多样性之间的关系,增加算法的选择压力,本文提出了一种基于目标空间映射策略的高维多目标粒子群优化算法(many-objective particle swarm optimization algorithm based on objective space mapping strategy,MOPSO-OSM)。在求解高维多目标优化问题时,Pareto准则难以从众多的非支配解中确定最优“折中”解,因此将高维多目标空间映射为以收敛性和多样性评价指标的2维空间,再将上述2维空间根据性能指标的优劣划分为4个不同区域。同时,使用反向学习策略提高算法跳出局部最优的能力。实验表明,MOPSO-OSM算法可以有效平衡收敛性和多样性之间的关系,达到求解复杂多目标优化问题的目的。     

基于参考点的高维多目标粒子群算法

高维多目标优化问题一般指目标个数为4个 或以上时的多目标优化问题.由于种群中非支配解数量随着目标数量的增加而急剧增多,导致进化算法的进化压力严重降低,求解效率低.针对该问题,提出一种基于粒子群的高维多目标问题求解方法,在目标空间中引入一系列的参考点,根据参考点筛选出能兼顾多样性和收敛性的非支配解作为粒子的全局最优,以增大选择压力.同时,提出了基于参考点的外部档案维护策略,以保持最后所得解集的多样性.在标准测试函数DTLZ2上的仿真结果表明,所提方法在求解高维多目标问题时能够得到收敛性和分布性都较好的解集.     

基于分解和差分进化的多目标粒子群优化算法

为了提高多目标优化算法解集的分布性和收敛性,提出一种基于分解和差分进化的多目标粒子群优化算法(dMOPSO-DE).该算法通过提出方向角产生一组均匀的方向向量,确保粒子分布的均匀性;引入隐式精英保持策略和差分进化修正机制选择全局最优粒子,避免种群陷入局部最优Pareto前沿;采用粒子重置策略保证群体的多样性.与非支配排序(NSGA-II)算法、多目标粒子群优化(MOPSO)算法、分解多目标粒子群优化(dMOPSO)算法和分解多目标进化-差分进化(MOEA/D-DE)算法进行比较,实验结果表明,所提出算法在求解多目标优化问题时具有良好的收敛性和多样性.     

改进的r支配高维多目标粒子群优化算法

高维多目标优化问题是广泛存在于实际应用中的复杂优化问题,目前的研究方法大都限于进化算法.本文利用粒子群优化算法求解高维多目标优化问题,提出了一种基于r支配的多目标粒子群优化算法.采用r支配关系进行粒子的比较与选择,并结合粒子群优化算法收敛速度快的优势,使得算法在目标个数增加时仍保持较强的搜索能力;为了弥补由此造成的群体多样性的丢失,优化非r支配阈值的取值策略;此外,引入决策空间的拥挤距离测度,并给出新的外部存储器更新方法,从而进一步防止算法陷入局部最优.对多个基准测试函数的仿真结果表明所得解集在收敛性、多样性以及围绕参考点的分布性上均优于其他两种算法.     

进化高维多目标优化算法研究综述

首先针对常规多目标优化算法求解高维多目标优化时面临的选择压力衰减问题进行论述;然后针对该问题,按照选择机制的不同详细介绍基于Pareto支配、基于分解策略和基于性能评价指标的典型高维多目标优化算法,并分析各自的优缺点;接着立足于一种全新的性能评价指标-----R2指标,给出R2指标的具体定义,介绍基于R2指标的高维多目标优化算法,分析此类算法的本质,并按照R2指标的4个关键组成部分进行综述;最后,发掘其存在的潜在问题以及未来发展空间.     

基于R2指标和参考向量的高维多目标进化算法

在高维多目标优化中, 不同的优化问题存在不同形状的Pareto前沿(PF), 而研究表明大多数多目标进化算法(Multi-objective evolutionary algorithms, MOEAs) 在处理不同的优化问题时普适性较差. 为了解决这个问题, 本文提出了一个基于R2指标和参考向量的高维多目标进化算法(An R2 indicator and reference vector based many-objective optimization evolutionary algorithm, R2-RVEA). R2-RVEA基于Pareto支配选取非支配解来指导种群进化, 仅当非支配解的数量超过种群规模时, 算法进一步采用种群分解策略和R2指标选择策略进行多样性管理. 通过大量的实验证明, 本文提出的算法在处理不同形状的PF时具有良好的性能.     

基于多样性检测的双子群多目标粒子群算法

为了平衡多目标粒子群算法的多样性和收敛性,提出一种基于多样性检测的多子群多目标粒子群算法.首先,将多样性检测方法引入到多目标粒子群算法中,并结合多目标粒子群算法的特点进行改进.然后,将种群分为两个不同分工的子群,一个子群保持较好的多样性,在搜索空间进行全局搜索;另一个子群保持较好的收敛性,在Pareto前沿附近进行局部搜索.最后,根据多样性度量指标调整两个子群的搜索行为,以达到兼顾多样性和收敛性的目的.在标准测试问题上的仿真结果表明了所提算法的有效性.     

采用放松支配关系的高维多目标微分进化算法

为了提高进化算法在求解高维多目标优化问题时的收敛性和多样性,提出了采用放松支配关系的高维多目标微分进化算法。该算法采用放松的Pareto支配关系,以增加个体的选择压力;采用群体和外部存储器协同进化的方案,并通过混合微分变异算子,生成子代群体;采用基于指标的方法计算个体的适应度并对群体进行更新;采用基于Lp范数（0相似文献   

基于目标空间分解和连续变异的多目标粒子群算法

针对当前多目标粒子群优化算法收敛性和多样性不佳等问题,提出了一种基于目标空间分解和连续变异的多目标粒子群优化算法。利用目标空间分解方法将粒子群分配到预先设定好的子区域中,在该过程中,通过一种新适应值公式来对每个子区域中的粒子进行择优筛选,该适应值公式融入了支配强度因素;在全局搜索过程中,使用差分变异、高斯变异和柯西变异对全局引导粒子的位置进行连续变异操作。将该算法与当前主流的一些多目标优化算法进行对比实验,结果表明,本文提出的算法在提高粒子收敛性的同时,多样性也得到了提升。     

基于超球形模糊支配的高维多目标粒子群优化算法

高维多目标优化问题（MAOP）会随着待优化问题维度的增加形成巨大的目标空间,导致在目标空间中非支配解的比例急剧增加,削弱了进化算法的选择压力,降低了进化算法对MAOP的求解效率。针对这一问题,提出一种以超球型支配关系降低种群中非支配解数量的粒子群优化（PSO）算法。算法以模糊支配策略来维持种群对MAOP的选择压力,并通过全局极值的选择和外部档案的维护来保持种群个体在目标空间中的分布。在标准测试集DTLZ和WFG上的仿真结果表明,所提算法在求解MAOP时具备较优的收敛性和分布性。     

Handling many-objective optimisation problems with R2 indicator and decomposition-based particle swarm optimiser

An R2 indicator-based multi-objective particle swarm optimiser (R2-MOPSO) can obtain well-convergence and well-distributed solutions while solving two and three objectives optimisation problems. However, R2-MOPSO faces difficulty to tackle many-objective optimisation problems because balancing convergence and diversity is a key issue in high-dimensional objective space. In order to address this issue, this paper proposes a novel algorithm, named R2-MaPSO, which combines the R2 indicator and decomposition-based archiving pruning strategy into particle swarm optimiser for many-objective optimisation problems. The innovations of the proposed algorithm mainly contains three crucial factors: (1) A bi-level archiving maintenance approach based on the R2 indicator and objective space decomposition strategy is designed to balance convergence and diversity. (2) The global-best leader selection is based on the R2 indicator and the personal-best leader selection is based on the Pareto dominance. Meanwhile, the objective space decomposition leader selection adopts the feedback information from the bi-level archive. (3) A new velocity updated method is modified to enhance the exploration and exploitation ability. In addition, an elitist learning strategy and a smart Gaussian learning strategy are embedded into R2-MaPSO to help the algorithm jump out of the local optimal front. The performance of the proposed algorithm is validated and compared with some algorithms on a number of unconstraint benchmark problems, i.e. DTLZ1-DTLZ4, WFG test suites from 3 to 15 objectives. Experimental results have demonstrated a better performance of the proposed algorithm compared with several multi-objective particle swarm optimisers and multi-objective evolutionary algorithms for many-objective optimisation problems.     

DAV-MOEA:一种采用动态角度向量支配关系的高维多目标进化算法

现实中不断涌现的高维多目标优化问题对传统的基于Pareto支配的多目标进化算法构成巨大挑战.一些研究者提出了若干改进的支配关系,但仍难以有效地平衡高维多目标进化算法的收敛性和多样性.提出一种动态角度向量支配关系动态地刻画进化种群在高维目标空间的分布状况,以较好地在收敛性与多样性之间取得平衡;另外,提出一种改进的基于Lp...     

Indicator-based set evolution particle swarm optimization for many-objective problems

Multi-objective particle swarm optimization (MOPSO) has been well studied in recent years. However, existing MOPSO methods are not powerful enough when tackling optimization problems with more than three objectives, termed as many-objective optimization problems (MaOPs). In this study, an improved set evolution multi-objective particle swarm optimization (S-MOPSO, for short) is proposed for solving many-objective problems. According to the proposed framework of set evolution MOPSO (S-MOPSO), including quality indicators-based objective transformation, the Pareto dominance on sets, and the particle swarm operators for set evolution, an enhanced S-MOPSO method is developed by updating particles hierarchically, i.e., a set of solutions is first regarded as a particle to be updated and then the solutions in a selected set are further evolved by a modified PSO. In the set evolutionary stage, the strategy for efficiently updating the set particle is proposed. When further evolving a single solution in the initial decision space of the optimized MaOP, the global and local best particles are dynamically determined based on those ideal reference points. The performance of the proposed algorithm is empirically demonstrated by applying it to several scalable benchmark many-objective problems.     

An angle based evolutionary algorithm with infeasibility information for constrained many-objective optimization

Recently, angle-based approaches have shown promising for unconstrained many-objective optimization problems (MaOPs), but few of them are extended to solve constrained MaOPs (CMaOPs). Moreover, due to the difficulty in searching for feasible solutions in high-dimensional objective space, the use of infeasible solutions comes to be more important in solving CMaOPs. In this paper, an angle based evolutionary algorithm with infeasibility information is proposed for constrained many-objective optimization, where different kinds of infeasible solutions are utilized in environmental selection and mating selection. To be specific, an angle-based constrained dominance relation is proposed for non-dominated sorting, which gives infeasible solutions with good diversity the same priority to feasible solutions for escaping from the locally feasible regions. As for diversity maintenance, an angle-based density estimation is developed to give the infeasible solutions with good convergence a chance to survive for next generation, which is helpful to get across the large infeasible barrier. In addition, in order to utilize the potential of infeasible solutions in creating high-quality offspring, a modified mating selection is designed by considering the convergence, diversity and feasibility of solutions simultaneously. Experimental results on two constrained many-objective optimization test suites demonstrate the competitiveness of the proposed algorithm in comparison with five existing constrained many-objective evolutionary algorithms for CMaOPs. Moreover, the effectiveness of the proposed algorithm on a real-world problem is showcased.     

A novel two-archive strategy for evolutionary many-objective optimization algorithm based on reference points

Current evolutionary many-objective optimization algorithms face two challenges: one is to ensure population diversity for searching the entire solution space. The other is to ensure quick convergence to the optimal solution set. In this paper, we propose a novel two-archive strategy for evolutionary many-objective optimization algorithm. The uniform archive strategy, based on reference points, is used to keep population diversity in the evolutionary process, and to ensure that an evolutionary algorithm is able to search the entire solution space. The single elite archive strategy is used to ensure that individuals with the best single objective value are able to evolve into the next generation and have more opportunities to generate offspring. This strategy aims to improve the convergence rate. Then this novel two-archive strategy is applied to improving the Non-dominated Sorting Genetic Algorithm (NSGA-III). Simulation experiments are conducted on benchmark test sets and experimental results show that our proposed algorithm with the two-archive strategy has a better performance than other state-of-art algorithms.     

基于种群分区的多策略自适应多目标粒子群算法

在多目标粒子群优化算法中, 平衡算法收敛性和多样性是获得良好分布和高精度Pareto前沿的关键, 多数已提出的方法仅依靠一种策略引导粒子搜索, 在解决复杂问题时算法收敛性和多样性不足. 为解决这一问题, 提出一种基于种群分区的多策略自适应多目标粒子群优化算法. 采用粒子收敛性贡献对算法环境进行检测, 自适应调整粒子的探索和开发过程; 为准确制定不同性能的粒子的搜索策略, 提出一种多策略的全局最优粒子选取方法和多策略的变异方法, 根据粒子的收敛性评价指标, 将种群划分为3个区域, 将粒子性能与算法寻优过程结合, 提升种群中各个粒子的搜索效率; 为解决因选取的个体最优粒子不能有效指导粒子飞行方向, 使算法停滞, 陷入局部最优的问题, 提出一种带有记忆区间的个体最优粒子选取方法, 提升个体最优粒子选取的可靠性并加快粒子收敛过程; 采用包含双性能测度的融合指标维护外部存档, 避免仅根据粒子密度对外部存档维护时, 删除收敛性较好的粒子, 导致种群产生退化, 影响粒子开发能力. 仿真实验结果表明, 与其他几种多目标优化算法相比, 该算法具有良好的收敛性和多样性.     

解决约束优化问题的改进粒子群算法

针对约束优化问题的求解,提出一种改进的粒子群算法（CMPSO）。在CMPSO算法中,为了增加种群多样性,提升种群跳出局部最优解的能力,引入种群多样性阈值,当种群多样性低于给定阈值时,对全局最优粒子位置和粒子自身最优位置进行多项式变异;并根据粒子违背约束条件的程度,提出一种新的粒子间比较准则来比较粒子间的优劣,该准则可以保留一部分性能较优的不可行解;为提升种群向全局最优解飞行的概率,采取一种广义学习策略。对经典测试函数的仿真结果表明,所提出的算法是一种可行的约束优化问题的求解方法。