基于区间的不确定多目标优化方法研究

基于非线性区间优化,提出了一种不确定多目标优化方法.基于区间序关系和区间可能度,把不确定多目标的目标函数和约束转化为确定性的目标函数和确定性的约束.对于复杂的工程优化问题,为了提高效率,采用拉丁方试验设计方法,构建响应面近似模型,并基于近似模型进行不确定多目标优化,从而形成了非线性区间优化方法与近似模型相结合的高效不确定多目标优化方法.数值算例表明了该方法的有效性和工程实用性.     

改进的遗传算法在连续体结构多目标拓扑优化中的应用

连续体结掏多目标拓扑优化是结构优化领域中的一个较难的研究课题.本文提出了一种改进的SPEA2多目标优化算法.该算法中采用数学形态学中的四方向链码的编码方式进行结构拓扑表达,使产生的拓扑结构清晰且无异议,完全消除了模糊的拓扑边界和棋盘格现象.提出了"质量向量"的概念,用以量化两个结构拓扑之间的相似性,从而使得不同结构拓扑之间的相似性比较成为可能.同时还将机器学习中的范例学习的思想融入到新算法中,利用前面的有限元分析结果对后来的结构分析进行指导,剔除了很多不必要的重复的计算,从而使算法的总计算量大幅度地减少.将新算法应用于悬臂和简支两类深梁的多目标拓扑优化,获得了高质量的Pareto最优解,且具有很好的分布.     

FLNG低温软管内衬波纹管多目标优化设计

低温柔性管道是海上浮式液化天然气生产储卸装置系统中输送液化天然气的核心装备之一。U型波纹管作为低温柔性管道最内层结构,直接影响低温管道的安全性。本文选取波径和环板长度作为独立结构设计参数进行结构分析,解决了在传统的U型波纹管灵敏度分析中,波高会随波距的改变而改变的耦合计算问题。构建了径向基函数神经网络代理模型,采用遗传算法对U型波纹管结构开展了多目标优化,获得以拉伸刚度最大和弯曲刚度最小为优化目标时的帕累托最优解集,为U型波纹管在实际工程应用中提供有益的设计参考。

     

仿生多胞薄壁管耐撞性分析及优化

为提高薄壁管结构耐撞性,以雀尾螳螂虾螯为仿生原型,结合仿生学设计方法,设计一种含正弦胞元的多胞薄壁管结构。以初始峰值载荷、比吸能和碰撞力效率为耐撞性指标,通过有限元数值模拟分析了不同碰撞角度（0º、10º、20º和30º）条件下,仿生胞元数对薄壁管耐撞性的影响,通过多目标的复杂比例评估法获取仿生薄壁管的最优胞元数。基于不同碰撞角度权重因子组合,设置了4种单一角度工况和3种多角度工况,采用多目标粒子群优化方法获取了不同工况下薄壁管结构最优胞元高宽比和壁厚。复杂比例评估结果表明,胞元数为4的薄壁管为最优晶胞数仿生薄壁管。优化结果表明,单一角度工况下,最优结构参数高宽比的范围为0.88～1.50,壁厚的范围为0.36～0.60 mm,碰撞角度为0º和10º的最优高宽比明显小于碰撞角度为20º和30º的;多角度工况下,最优高宽比范围为1.01～1.10,壁厚范围为0.49～0.57 mm。

     

Extended Fuzzy C-Means and Genetic Algorithms to Optimize Power Flow Management in Hybrid Electric Vehicles

The need for personal transportation must be harmonized by considering the impact of so huge number of vehicles on the environment. The adoption of hybrid electric vehicles can provide a sensible improvement from an environmental viewpoint, but at the same time makes more difficult the definition and implementation of the overall powertrain control mechanism. In fact, powertrain control problems are known to be very complex due to conflicting requirements, and this difficulty augments in case of hybrid electric vehicles. Most of the features of the future hybrid electric vehicles are enabled by a new energy flow management unit designed to split the instantaneous power demand between the internal combustion engine and the electric motor, ensuring both an efficient power supply and reduced emissions. Classic approaches that rely on static thresholds, optimized on a fixed drive cycle, cannot face the high dynamicity and unpredictability of real-life drive conditions. The need to actually control a real vehicle stimulates the research of innovative methodologies for the real-time identification of the operating points of each energy source. This paper is framed into this context: after a brief discussion about a non-conventional formalization of the energy flows problem based on a multiobjective function, a knowledge-based control system for splitting the vehicle's power demand between the engine and motor is presented. The proposed approach exploits a fuzzy clustering criterion that combined with a genetic algorithm, permits to achieve better results, both in terms of a reduced computational effort and an improved efficiency of the control system over various driving cycles. To validate the proposed approach, simulation tests and comparisons with other energy management strategies are discussed.     

并行Pareto多目标演化算法

提出并实现了一个并行Pareto多目标演化算法(PPMEA)，PPMEA算法是一个通过演化算法来解决基于Pareto的多目标优化问题的并行计算模型。在这个算法中，采用了并行演化算法中常用的全局并行模型和粗粒度并行岛模型。构成每个岛的子群体以不同的杂交概率和变异概率各自独立的演化，但是每隔一定的代数它们要交换外部集中的个体。标准测试函数的数值实验结果表明，所提出的算法能够使找到的解集快速收敛到Pareto非劣最优目标域，并且解集沿着Pareto非劣最优目标域有很好的扩展性。     

Nonempty Intersection Theorems and Generalized Multi-objective Games in Product <Emphasis Type="Italic">FC</Emphasis>-Spaces

A new class of generalized multi-objective games is introduced and studied in FC-spaces where the number of players may be finite or infinite, and all payoff are all set-valued mappings and get their values in a topological space. By using an existence theorems of maximal elements for a family of set-valued mappings in product FC-spaces due to author, some new nonempty intersection theorems for a family of set-valued mappings are first proved in FC-spaces. As applications, some existence theorems of weak Pareto equilibria for the generalized multi-objective games are established in noncompact FC-spaces. These theorems improve, unify and generalize the corresponding results in recent literatures.     

On finding multiple Pareto-optimal solutions using classical and evolutionary generating methods

In solving multi-objective optimization problems, evolutionary algorithms have been adequately applied to demonstrate that multiple and well-spread Pareto-optimal solutions can be found in a single simulation run. In this paper, we discuss and put together various different classical generating methods which are either quite well-known or are in oblivion due to publication in less accessible journals and some of which were even suggested before the inception of evolutionary methodologies. These generating methods specialize either in finding multiple Pareto-optimal solutions in a single simulation run or specialize in maintaining a good diversity by systematically solving a number of scalarizing problems. Most classical generating methodologies are classified into four groups mainly based on their working principles and one representative method from each group is chosen in the present study for a detailed discussion and for its performance comparison with a state-of-the-art evolutionary method. On visual comparisons of the efficient frontiers obtained for a number of two and three-objective test problems, the results bring out interesting insights about the strengths and weaknesses of these approaches. The results should motivate researchers to design hybrid multi-objective optimization algorithms which may be better than each of the individual methods.     

多目标决策中的力学思想及其交互重心移动法

利用力学中的重心概念 ,提出求多目标规划问题满意解的交互重心移动法 .它不仅简便易行 ,而且可以有效地利用以前的信息 ,防止决策者前后判断不一致情况的出现 .     

基于多目标优化模型的旅游路线研究

旅游活动正在成为全球经济发展的重要动力之一,而旅游线路直接关系着旅游的时间、费用、旅游体验.为设计更合理的覆盖全国5A景区的旅游路线,本文使用聚类分析、贪婪算法、图论等方法并设计了合理的旅游体验函数,最终得到了从西安和北京出发的游客采用全程自驾或结合其它交通工具十年内游遍所有201个5A景区、费用最优、旅游体验最好的旅游线路.