Pareto-adaptive epsilon-dominance

Efficiency has become one of the main concerns in evolutionary multiobjective optimization during recent years. One of the possible alternatives to achieve a faster convergence is to use a relaxed form of Pareto dominance that allows us to regulate the granularity of the approximation of the Pareto front that we wish to achieve. One such relaxed forms of Pareto dominance that has become popular in the last few years is epsilon-dominance, which has been mainly used as an archiving strategy in some multiobjective evolutionary algorithms. Despite its advantages, epsilon-dominance has some limitations. In this paper, we propose a mechanism that can be seen as a variant of epsilon-dominance, which we call Pareto-adaptive epsilon-dominance (paepsilon-dominance). Our proposed approach tries to overcome the main limitation of epsilon-dominance: the loss of several nondominated solutions from the hypergrid adopted in the archive because of the way in which solutions are selected within each box.     

Multi-objective optimization using teaching-learning-based optimization algorithm

Two major goals in multi-objective optimization are to obtain a set of nondominated solutions as closely as possible to the true Pareto front (PF) and maintain a well-distributed solution set along the Pareto front. In this paper, we propose a teaching-learning-based optimization (TLBO) algorithm for multi-objective optimization problems (MOPs). In our algorithm, we adopt the nondominated sorting concept and the mechanism of crowding distance computation. The teacher of the learners is selected from among current nondominated solutions with the highest crowding distance values and the centroid of the nondominated solutions from current archive is selected as the Mean of the learners. The performance of proposed algorithm is investigated on a set of some benchmark problems and real life application problems and the results show that the proposed algorithm is a challenging method for multi-objective algorithms.     

Multi-objective topology and sizing optimization of truss structures based on adaptive multi-island search strategy

This paper uses genetic algorithm to handle the topology and sizing optimization of truss structures, in which a sparse node matrix encoding approach is used and individual identification technique is employed to avoid duplicate structural analysis to save computation time. It is observed that NSGA-II could not improve the convergence of non-dominated front at latter generations when solving multi-objective topology and sizing optimization of truss structures. Therefore, an adaptive multi-island search strategy for multi-objective optimization problem (AMISS-MOP) is developed to enhance the convergence. Meanwhile, an elitist strategy based on archive set is introduced to reduce the size of non-dominated sorting to improve computation efficiency. Two numeric examples are presented to demonstrate the performance of AMISS-MOP. Results show that the global Pareto front could be found by AMISS-MOP, the convergence is improved as generation increases, and the time spent on non-dominated sorting is reduced.     

Improving the efficiency of ?-dominance based grids

In this paper, we deal with the problem of handling solutions in an external archive with the use of a relaxed form of Pareto dominance called ?-dominance and a variation of it called pa?-dominance. These two relaxed forms of Pareto dominance have been used as archiving strategies in some multi-objective evolutionary algorithms (MOEAs). The main objective of this work is to improve the ?-dominance based schemes to handle nondominated solutions, or to retain nondominated solutions in an external archive. Thus, our main contribution is to add an extra objective function only at the time of accepting a nondominated solution into the external archive, in order to preserve some solutions which are normally lost when using any of the aforementioned relaxed forms of Pareto dominance. Such a proposal is inexpensive (computationally speaking) and quite effective, since it is able to produce Pareto fronts of much better quality than the aforementioned archiving techniques.     

Multiobjective optimization using an adaptive weighting scheme

A new Pareto front approximation method is proposed for multiobjective optimization problems (MOPs) with bound constraints. The method employs a hybrid optimization approach using two derivative-free direct search techniques, and intends to solve black box simulation-based MOPs where the analytical form of the objectives is not known and/or the evaluation of the objective function(s) is very expensive. A new adaptive weighting scheme is proposed to convert a multiobjective optimization problem to a single objective optimization problem. Another contribution of this paper is the generalization of the star discrepancy-based performance measure for problems with more than two objectives. The method is evaluated using five test problems from the literature, and a realistic engineering problem. Results show that the method achieves an arbitrarily close approximation to the Pareto front with a good collection of well-distributed nondominated points for all six test problems.     

Towards faster convergence of evolutionary multi-criterion optimization algorithms using Karush Kuhn Tucker optimality based local search

Evolutionary multi-criterion optimization (EMO) algorithms emphasize non-dominated and less crowded solutions in a population iteratively until the population converges close to the Pareto optimal set. During the search process, non-dominated solutions are differentiated only by their local crowding or contribution to hypervolume or using a similar other metric. Thus, during evolution and even at the final iteration, the true convergence behavior of each non-dominated solutions from the Pareto optimal set is unknown. Recent studies have used Karush Kuhn Tucker (KKT) optimality conditions to develop a KKT Proximity Measure (KKTPM) for estimating proximity of a solution from Pareto optimal set for a multi-objective optimization problem. In this paper, we integrate KKTPM with a recently proposed EMO algorithm to enhance its convergence properties towards the true Pareto optimal front. Specifically, we use KKTPM to identify poorly converged non-dominated solutions in every generation and apply an achievement scalarizing function based local search procedure to improve their convergence. Assisted by the KKTPM, the modified algorithm is designed in a way that maintains the total number of function evaluations as low as possible while making use of local search where it is most needed. Simulations on both constrained and unconstrained multi- and many objectives optimization problems demonstrate that the hybrid algorithm significantly improves the overall convergence properties. This study brings evolutionary optimization closer to mainstream optimization field and should motivate researchers to utilize KKTPM measure further within EMO and other numerical optimization algorithms.     

Multiobjective Evolutionary Algorithm With Controllable Focus on the Knees of the Pareto Front

The optimal solutions of a multiobjective optimization problem correspond to a nondominated front that is characterized by a tradeoff between objectives. A knee region in this Pareto-optimal front, which is visually a convex bulge in the front, is important to decision makers in practical contexts, as it often constitutes the optimum in tradeoff, i.e., substitution of a given Pareto-optimal solution with another solution on the knee region yields the largest improvement per unit degradation. This paper presents a selection scheme that enables a multiobjective evolutionary algorithm (MOEA) to obtain a nondominated set with controllable concentration around existing knee regions of the Pareto front. The preference-based focus is achieved by optimizing a set of linear weighted sums of the original objectives, and control of the extent of the focus is attained by careful selection of the weight set based on a user-specified parameter. The fitness scheme could be easily adopted in any Pareto-based MOEA with little additional computational cost. Simulations on various two- and three-objective test problems demonstrate the ability of the proposed method to guide the population toward existing knee regions on the Pareto front. Comparison with general-purpose Pareto based MOEA demonstrates that convergence on the Pareto front is not compromised by imposing the preference-based bias. The performance of the method in terms of an additional performance metric introduced to measure the accuracy of resulting convergence on the desired regions validates the efficacy of the method.      

基于分解的多目标花朵授粉算法

在过去几十年里，许多多目标进化算法被广泛应用于解决多目标优化问题，其中一种比较流行的多目标进化算法是基于分解的多目标进化算法(MOEA/D)。花朵授粉算法是一种启发式优化算法，但迄今为止，花朵授粉算法在基于分解的多目标进化算法领域的研究还非常少。本文在基于分解的多目标进化算法的框架下，将花朵授粉算法拓展至多目标优化领域，提出一种基于分解的多目标花朵授粉算法(MOFPA/D)。此外，为了保证非支配解的多样性，本文提出一种基于网格的目标空间分割法，该方法从找到的Pareto最优解集中筛选出一定数量且分布均匀的Pareto最优解。实验结果表明，基于分解的多目标花朵授粉算法在收敛性与多样性方面均优于基于分解的多目标进化算法。     

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.     

基于粒子记忆体的多目标微粒群算法*

针对多目标微粒群算法(MOPSO)解的多样性分布问题,提出一种基于粒子记忆体的多目标微粒群算法(dp-MOPSO)。dp-MOPSO算法为每个微粒分配一个记忆体,保存寻优过程中搜索到的非支配pbest集,以避免搜索信息的丢失。采用外部存档保存种群搜索到的所有Pareto解,并引入动态邻域的策略从外部存档中选择全局最优解。利用几个典型的多目标测试函数对dp-MOPSO算法的性能进行测试,并与两种著名的多目标进化算法m-DNPSO、SPEA2进行比较。实验结果表明,dp-MOPSO算法可以更好地逼近真实Pareto沿,同时所得Pareto解分布更均匀。     

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.     

Multiobjective particle swarm optimization with nondominated local and global sets

In multiobjective particle swarm optimization (MOPSO) methods, selecting the local best and the global best for each particle of the population has a great impact on the convergence and diversity of solutions, especially when optimizing problems with high number of objectives. This paper presents an approach using two sets of nondominated solutions. The ability of the proposed approach to detect the true Pareto optimal solutions and capture the shape of the Pareto front is evaluated through experiments on well-known non-trivial multiobjective test problems as well as the real-life electric power dispatch problem. The diversity of the nondominated solutions obtained is demonstrated through different measures. The proposed approach has been assessed through a comparative study with the reported results in the literature.     

混合多目标粒子群优化算法在热精轧负荷分配优化中的应用

通过对热精轧负荷分配过程的分析,选取负荷均衡、板形良好和轧制功率最低为目标,建立了热精轧负荷分配多目标优化模型.为了提高多目标优化算法解集的分布性和收敛性,提出了一种混合多目标粒子群优化算法(HMOPSO),该算法根据Pareto支配关系得到Pareto前沿进而保证种群收敛;采用分解策略维护外部存档,该策略首先根据Pareto前沿求出上界点对目标空间进行归一化处理,然后对种群进行分区处理进而保证种群的分布性能.仿真结果表明,HMOPSO的收敛性和分布性都好于MOPSO和d MOPSO;采用模糊多属性决策的方法从Pareto最优解集中选择一个Pareto最优解,通过与经验负荷分配方法相比,表明该Pareto最优解可以使轧制方案更加合理.     

求解偏好多目标优化的克隆选择算法

目标维数较高的多目标优化问题的难题在于非支配解急剧增加,经典算法由于缺乏足够的选择压力导致性能急剧下降.提出了基于偏好等级的免疫记忆克隆选择优化算法,用于解决目标维数较高的多目标优化问题.利用决策者提供的偏好信息来为抗体分配偏好等级,根据该值比例克隆抗体,增大抗体的选择压力,加快收敛速率.根据偏好信息来缩减Pareto前沿,并用有限的偏好解估计该前沿.同时,建立了免疫记忆种群来保留较好的非支配抗体,采用ε支配机制来保持记忆抗体种群的多样性.实验结果表明,对于2目标的偏好多目标问题以及高达8目标的DTLZ2和DTLZ3问题,该算法取得了一定的实验效果.     

一种基于群智能的快速多目标优化算法

粒子群优化算法是一种典型的仿真群智能的算法。探讨了利用粒子群算法求解多目标优化问题,为了提高算法速度,采用了几何Pareto选择算法作为文档算法,用多方向搜索的办法寻找极端点。实验表明：该算法得到的解的数量多,速度快并且近似前沿的程度比较高。     

A New Evolutionary Algorithm for Solving Many-Objective Optimization Problems

In this paper, we focus on the study of evolutionary algorithms for solving multiobjective optimization problems with a large number of objectives. First, a comparative study of a newly developed dynamical multiobjective evolutionary algorithm (DMOEA) and some modern algorithms, such as the indicator-based evolutionary algorithm, multiple single objective Pareto sampling, and nondominated sorting genetic algorithm II, is presented by employing the convergence metric and relative hypervolume metric. For three scalable test problems (namely, DTLZ1, DTLZ2, and DTLZ6), which represent some of the most difficult problems studied in the literature, the DMOEA shows good performance in both converging to the true Pareto-optimal front and maintaining a widely distributed set of solutions. Second, a new definition of optimality (namely, L-optimality) is proposed in this paper, which not only takes into account the number of improved objective values but also considers the values of improved objective functions if all objectives have the same importance. We prove that L-optimal solutions are subsets of Pareto-optimal solutions. Finally, the new algorithm based on L-optimality (namely, MDMOEA) is developed, and simulation and comparative results indicate that well-distributed L-optimal solutions can be obtained by utilizing the MDMOEA but cannot be achieved by applying L-optimality to make a posteriori selection within the huge Pareto nondominated solutions. We can conclude that our new algorithm is suitable to tackle many-objective problems.      

基于平衡搜索策略的多目标粒子群优化算法*

鉴于平衡全局和局部搜索在多目标粒子群优化算法获取完整均匀Pareto最优前沿方面的重要性,设计平衡全局和局部搜索策略,进而提出改进的多目标粒子群优化算法(bsMOPSO).文中策略在局部搜索方面设计归档集自挖掘子策略,通过对归档集中均匀分布的部分粒子进行柯西扰动,使归档集涵盖整个前沿面的局部搜索.在全局搜索方面设计边界最优粒子引导搜索子策略,以边界最优粒子替换部分粒子的全局最优解,引导粒子向各维目标的边界区域搜索.选取4种对比算法在ZDT和DTLZ系列的部分测试函数上进行实验,结果表明bsMOPSO具有更快的Pareto最优前沿收敛效率和更好的分布性.     

Dominance-Based Multiobjective Simulated Annealing

Simulated annealing is a provably convergent optimizer for single-objective problems. Previously proposed multiobjective extensions have mostly taken the form of a single-objective simulated annealer optimizing a composite function of the objectives. We propose a multiobjective simulated annealer utilizing the relative dominance of a solution as the system energy for optimization, eliminating problems associated with composite objective functions. We also propose a method for choosing perturbation scalings promoting search both towards and across the Pareto front. We illustrate the simulated annealer's performance on a suite of standard test problems and provide comparisons with another multiobjective simulated annealer and the NSGA-II genetic algorithm. The new simulated annealer is shown to promote rapid convergence to the true Pareto front with a good coverage of solutions across it comparing favorably with the other algorithms. An application of the simulated annealer to an industrial problem, the optimization of a code-division-multiple access (CDMA) mobile telecommunications network's air interface, is presented and the simulated annealer is shown to generate nondominated solutions with an even and dense coverage that outperforms single objective genetic algorithm optimizers.     

面向多目标优化的适应度共享免疫克隆算法

多目标优化的日标在于使得解集能够快速的逼近真实Pareto前沿.针对解的分布性问题,以免疫克隆算法为框架,引入适应度共享策略,提出了一种新的具有良好分布性保持的多目标优化进化算法;算法建立外部群体以保存非支配解,以Pareto优和共亨适应度作为外部群体更新与激活抗体选择的双重标准.为了增强算法对决策空间的开发能力,引入...     

一种新的多目标粒子群优化算法

论文提出了一种基于拥挤度和动态惯性权重聚合的多目标粒子群优化算法,该算法采用Pareto支配关系来更新粒子的个体最优值,用外部存档策略保存搜索过程中发现的非支配解;采用适应值拥挤度裁剪归档中的非支配解,并从归档中的稀松区域随机选取精英作为粒子的全局最优位置,以保持解的多样性;采用动态惯性权重聚合的方法以使算法尽可能地逼近各目标的最优解。仿真结果表明,该算法性能较好,能很好地求解多目标优化问题。