遗传算法在多目标优化应用中的对比研究

多目标优化应用研究在过程工程领域越来越受重视。本文首先给出了多目标优化问题的一般形式,指出多目标问题求解任务：引导搜索向整个的Pareto优化范围;Pareto优化前沿上保持解集的多样性。在简要论述遗传算法求解多目标技术的基础上,对应用了遗传算法求解多目标的两种方法进行了对比研究,并给出了线性加权遗传算法和一种多目标遗传算法的计算框图。指出线性加权法求解Pareto最优解时不能不能很好地处理非凸区域、均匀分布的权重值不能生成均匀分布的Pareto前沿等局限性,以及多目标遗传算法生成种群多样性及Pareto最优解均匀分布的优点,并用实例进行了验证说明。     

基于最佳进化方向的多目标遗传算法

该文模拟自然界中生物总是向着有利于自己的方向进化,即朝生物利益最大化的方向进化这一现象,给出了一种新的设计适应度函数的方法,并且结合多目标优化的Pareto最优解的概念,提出了求解多目标优化问题的一种新的算法———基于最佳基因的多目标遗传算法。数值实验表明,该算法不仅操作简单、鲁棒性强、速度快、且能够获得数量多而且广泛的Pareto最优解。     

采用遗传算法求解多跳分组无线网服务质量路由优化模型

采用多目标遗传算法来确定多跳无线网服务质量路由优化问题的Pareto最优解集。通过计算表明，多目标遗传算法能够在一次运行中搜索到优化问题的近似Pareto最优解集，这为决策者进行目标折衷决策提供了充分的依据，此算法是有效可行的。     

基于精英选择和个体迁移的多目标遗传算法

提出基于遗传算法求解多目标优化问题的方法,将多目标问题分解成多个单目标优化问题,用遗传算法分别在每个单目标种群中并行搜索.在进化过程中的每一代,采用精英选择和个体迁移策略加快多个目标的并行搜索,提出了控制Pareto最优解数量并保持个体多样性的有限精度法,同时还提出了多目标遗传算法的终止条件.数值实验说明所提出的算法能较快地找到一组分布广泛且均匀的Pareto最优解.     

一种新的求解约束多目标优化问题的遗传算法

由于采用罚函数法将有约束多目标优化问题转化为无约束多目标优化问题会使求解不合理,因此,文章首先在无约束Pareto排序遗传算法的基础上,提出了一个简单、实用的能分别考虑目标函数和约束函数,而又可以避免采用罚函数的全新排序方法。接着,针对小生境技术在遗传后期依旧会出现遗传漂移现象和共享半径不易确定等缺陷,提出了一种易于实现的超量惩罚策略来替代小生境技术,用以改进种群的多样性。此外,还采用了Pareto解集过滤器、邻域变异和群体重组等策略对算法的寻优能力进行改进,并最终形成了一种求解有约束多目标优化问题的Pareto遗传算法(CMOPGA),还给出了具体的算法流程图。最后采用两个数值算例对算法的求解性能进行了测试。数值试验表明,采用CMOPGA可方便地求得问题的Pareto前沿,并能使求得的Pareto最优解集具有可靠、均布、多样等特点。     

具有约束多目标优化的进化算法

针对带有约束多目标优化问题,提出一种多目标优化进化算法。在选择过程中,采用约束的Pareto支配和聚集距离定义适应值,根据适应值挑选出有代表性的个体。在变异过程中,沿着权重梯度方向搜索来寻找可行的Pareto最优解。最后,采用两个数值算例测草算法的性能,结果表明该算法能获得多目标约束优化问题的可行Pareto最优解并且具有较好的分散性。     

多目标优化问题的有效Pareto最优集

多目标优化问题求解是当前演化计算的一个重要研究方向,而基于Pareto最优概念的遗传算法更是研究的重点,然而,遗传算法在解决多目标优化问题上的缺陷却使得其往往得不到一个令人满意的解。在对该类算法研究的基础上提出了衡量Pareto最优解集的标准,并对如何满足这个标准提出了建议。     

基于遗传多目标优化算法的Pareto最优解研究

遗传算法可有效求解多目标优化问题中的Pareto最优解,并利用MATLAB进行了仿真验证。     

基于数据仓库的多目标优化遗传算法

基于数据仓库的多目标优化遗传算法为解决多目标优化问题提供了有效的途径。其基本思想是:为求Pareto最优解的多目标优化遗传算法建立一个数据仓库,将进化过程中所产生的每一代Pareto最优解放入数据仓库中,在每一代先对数据仓库中的所有个体进行求Pareto最优解运算,淘汰掉劣解,再进行个体间的欧氏距离运算,将小于指定值的其中一个个体作为劣解处理。大量的计算机仿真计算表明,这种算法不仅能够有效地避免交叉或变异操作对Pareto最优解产生的破坏,而且进化速度极快,算法稳定,一般只需20￣40代的运算,即可得到分布广泛的Pareto最优解。     

QoS全局最优的多目标Web服务选择算法*

针对现有方法的不足,提出一种基于QoS全局最优的多目标动态Web服务选择算法。在给出动态服务组合模型的基础上,以“抽象服务规划”为输入,以用户的非功能性需求为全局约束,将动态服务选择问题转换为一个带QoS约束的多目标服务组合优化问题;利用多目标蚁群算法,多个目标函数被同时优化并产生一组满足约束条件的Pareto优化解。通过运用实验与基于多目标遗传算法的Web服务选择算法进行对比,证明了该方法的可行性和有效性。     

Adaptive weighted-sum method for bi-objective optimization: Pareto front generation

This paper presents a new method that effectively determines a Pareto front for bi-objective optimization with potential application to multiple objectives. A traditional method for multiobjective optimization is the weighted-sum method, which seeks Pareto optimal solutions one by one by systematically changing the weights among the objective functions. Previous research has shown that this method often produces poorly distributed solutions along a Pareto front, and that it does not find Pareto optimal solutions in non-convex regions. The proposed adaptive weighted sum method focuses on unexplored regions by changing the weights adaptively rather than by using a priori weight selections and by specifying additional inequality constraints. It is demonstrated that the adaptive weighted sum method produces well-distributed solutions, finds Pareto optimal solutions in non-convex regions, and neglects non-Pareto optimal solutions. This last point can be a potential liability of Normal Boundary Intersection, an otherwise successful multiobjective method, which is mainly caused by its reliance on equality constraints. The promise of this robust algorithm is demonstrated with two numerical examples and a simple structural optimization problem.     

多目标协调进化算法研究

进化算法适合解决多目标优化问题,但难以产生高维优化问题的最优解,文中针对此问题提出了一种求解高维目标优化问题的新进化方法,即多目标协调进化算法,主要特点是进化群体按协调模型使用偏好信息进行偏好排序,而不是基于Pareto优于关系进行了个体排序,实验结果表明,所提出的算法是可行而有效的,且能在有限进化代数内收敛。     

多目标拆卸线平衡问题的Pareto细菌觅食算法*

拆卸线平衡问题的优化涉及多个目标,为克服传统方法在求解多目标拆卸线平衡问题时不能很好处理各子目标间冲突及易于早熟等不足,提出了一种多目标细菌觅食优化算法。算法采用Pareto非劣排序技术对种群进行分级,并结合拥挤距离机制评价同级个体的优劣。为提高算法收敛性能,在趋向性操作结束后引入精英保留策略保留优秀个体,并采用全局信息共享策略引导菌群不断向均匀分布的Pareto最优前沿趋近。通过不同规模算例的对比验证表明了算法的有效性与优越性。     

Approximating Pareto frontier using a hybrid line search approach

The aggregation of objectives in multiple criteria programming is one of the simplest and widely used approach. But it is well known that this technique sometimes fail in different aspects for determining the Pareto frontier. This paper proposes a new approach for multicriteria optimization, which aggregates the objective functions and uses a line search method in order to locate an approximate efficient point. Once the first Pareto solution is obtained, a simplified version of the former one is used in the context of Pareto dominance to obtain a set of efficient points, which will assure a thorough distribution of solutions on the Pareto frontier. In the current form, the proposed technique is well suitable for problems having multiple objectives (it is not limited to bi-objective problems) and require the functions to be continuous twice differentiable. In order to assess the effectiveness of this approach, some experiments were performed and compared with two recent well known population-based metaheuristics namely ParEGO and NSGA II. When compared to ParEGO and NSGA II, the proposed approach not only assures a better convergence to the Pareto frontier but also illustrates a good distribution of solutions. From a computational point of view, both stages of the line search converge within a short time (average about 150 ms for the first stage and about 20 ms for the second stage). Apart from this, the proposed technique is very simple, easy to implement and use to solve multiobjective problems.     

A new perspective on multiobjective optimization by enhanced normalized normal constraint method

In industrial applications, several objectives are often managed simultaneously (e.g., minimizing the cost and the weight of a mechanical structure satisfying some constraints). Although lots of optimization studies deal with only one objective, this approach is often not realistic for engineering optimization. Therefore, improvements in multiobjective optimization methods are required. This paper presents the formulation of a new utopia hyperplane that improves the proposal of the original normalized normal constraint method using two approaches: a redefinition of the anchor points and an exact linear transformation between the design objectives space and the normalized space. Both approaches always produce a normalized space with equal scales that improves the even distribution of the solutions over the Pareto frontier. Examples of the method proposed are presented related with mechanical engineering and structure design including a challenging non-convex Pareto frontier. Partially supported by FEDER DPI2005-07835, FEDER DPI2004-8383-C03-02 projects (MEC—Spain) and GV06/26 (Generalitat Valenciana)     

The normalized normal constraint method for generating the Pareto frontier

The authors recently proposed the normal constraint (NC) method for generating a set of evenly spaced solutions on a Pareto frontier – for multiobjective optimization problems. Since few methods offer this desirable characteristic, the new method can be of significant practical use in the choice of an optimal solution in a multiobjective setting. This papers specific contribution is two-fold. First, it presents a new formulation of the NC method that incorporates a critical linear mapping of the design objectives. This mapping has the desirable property that the resulting performance of the method is entirely independent of the design objectives scales. We address here the fact that scaling issues can pose formidable difficulties. Secondly, the notion of a Pareto filter is presented and an algorithm thereof is developed. As its name suggests, a Pareto filter is an algorithm that retains only the global Pareto points, given a set of points in objective space. As is explained in the paper, the Pareto filter is useful in the application of the NC and other methods. Numerical examples are provided.     

基于线性权重最优支配的高维多目标优化算法

在高维多目标优化问题中,Pareto支配关系存在非支配解随优化目标数增加呈指数级增长和种群选择压力下降等问题。针对这些问题,基于线性权重聚合函数和支配关系两种比较多目标解方法的思想,提出一种线性权重最优支配关系（LWM-dominance）,并理论证明了LWM非支配解集是Pareto非支配解集的子集,同时保留了种群中重要的角解。进一步地,基于LWM支配关系,实现了一个高维多目标进化优化算法,基于该算法的实验验证了LWM支配关系的性质。在随机解空间中的实验结果表明LWM支配关系适用于5~15个目标的高维多目标优化问题,通过DTLZ1~DTLZ7高维多目标优化问题进化过程中LWM非支配解集与Pareto非支配解集规模的对比实验,结果表明优化目标数为10和15时非支配解的比例平均下降了约17%。     

Incremental user-interface development for interactive multiobjective optimization

Interactive multiobjective optimization (IMO) is a subfield of multiple criteria decision making. In multiobjective optimization, the optimization problem is formulated with a mathematical model containing several conflicting objectives and constraints depending on decision variables. By using IMO methods, a decision maker progressively provides preference information in order to find the most satisfactory compromise between the conflicting objectives. In this paper, we consider implementation challenges of IMO methods. In particular, we consider what kind of interaction techniques can support the decision making process and information exchange between IMO methods and the decision maker. The implementation of an IMO method called Pareto Navigator is used as an example to demonstrate concrete challenges of interaction design. This paper focuses on describing the incremental development of the user interface for Pareto Navigator including empirical validation by user testing evaluation.     

基于生态协同的多目标优化研究

在分析现有多目标优化技术的基础上,提出了一种基于生态协同的多目标优化算法.此算法借鉴生态学中的生态种群密度竞争方程来描述多目标间的复杂关系,可以同时从个体和种群层次指导多目标之间关联程度的调整.实验结果表明,此算法更易于寻找多目标优化问题的满意解.