【作者】 姚坤；

【导师】 刘希玉；

【作者基本信息】 山东师范大学 ， 计算机软件与理论， 2007， 硕士

【摘要】 微粒群算法是上个世纪90年代提出的一种基于群体智能理论的优化算法,通过群体中粒子间的合作与竞争产生的群体智能指导优化搜索。相比于进化算法,微粒群算法保留了基于种群的全局搜索策略,采用简单的速度位移模式,避免了复杂的遗传操作,同时它特有的记忆使其可以动态跟踪当前的搜索情况以调整其搜索策略,具有全局收敛能力和鲁棒性,且不需要借助问题的特征信息。微粒群算法的特点吸引了大量学者对其进行研究,并被广泛的应用到了各个领域。协同进化算法是近十几年来在协同进化论基础上发展起来的一类新的进化算法。协同进化算法与传统进化算法的区别在于:协同进化算法在进化算法的基础上,考虑了种群与种群之间、种群与环境之间在进化过程中的协调。由于协同进化算法的诸多优越性,越来越多的学者对它进行了研究,协同进化已成为当前进化计算的一个热点问题。随着市场的日趋饱和,产品消费节奏明显加快,顾客的需求越来越多样化、个性化,购买行为更加具有选择性。在这种情况下,许多企业都已经意识到产品设计将成为抢占市场的决定性因素,因而试图通过产品的创新性来获得核心竞争力。创新性设计能够满足求新和多变的市场需求,有效地提高产品的市场竞争力。因此,创新概念设计也就成为近年来计算机辅助设计领域的热点之一。近年来,进化计算技术的兴起,为我们探索一种支持进化的创新概念设计途径提供了很好的支持。在此基础上产生的进化设计已成为最重要的创新概念设计技术之一。本文致力于研究将微粒群算法和协同进化算法相结合,建立一个支持创新概念设计的多微粒群协同进化模型,为分布式环境下的设计人员的协作和创新思路的开拓提供支撑平台。本文的主要工作为:1.提出了一个多微粒群协同进化算法尽管传统微粒群算法因其在迭代初期的有效性及实现简单等特点在很多领域中得到了广泛的应用,但是在寻找近似问题的最优解的效率和精度方面还是遇到了很大的困难。这表明,传统微粒群算法损失了粒子的多样性——所有的粒子被迄今为止找到的最好的微粒而吸引。本文在传统微粒群算法中加入了协同进化的思想,提出了一种多微粒群协同进化算法,保持了粒子的多样性,改善了微粒群算法的性能,仿真实验证明了该算法的有效性。2.构造了一个支持创新概念设计的多微粒群协同进化模型针对创新概念设计的独特性,本文在多微粒群协同进化的基础上,构造了一个支持创新概念设计的多微粒群协同进化模型。根据一般造型手机的特点,设计了相应的适应度评价函数,通过屏幕子群体和键盘子群体的协同进化,实现了一般造型手机的屏幕和键盘在外壳上的智能布局。3.在手机外观创新设计系统中实现了多微粒群协同进化模型在手机外观创新设计系统中,将利用该模型实现了一般造型手机的智能布局。该系统以HOOPS/NET作为整体架构、三维引擎ACIS为造型内核,采用SQL Server 2000数据库系统,利用VC++.NET 2003在WindowsXP平台上开发完成。在最后一般造型手机部件组装过程中使用了该模型实现了智能布局,结果令人满意。更多还原

【Abstract】 Particle Swarm Optimization(PSO) was presented in 1990s and it is an optimization algorithm based on theory of swarm intelligence. In PSO, swarm intelligence guides the optimization research, which is produced by cooperation and competition between particles. Comparing with evolution algorithm,Co-evolution algorithm has been developed as a kind of new evolution algorithm extensively on the base of co-evolution theory for a decade. The difference between co-evolution algorithm and traditional evolution algorithm is that co-evolution algorithm takes the adjustment of populations as well as that of populations and environment into account in evolving process. For the advantages of co-evolution algorithm, more and more scholars contribute to this field, which makes it a hotspot in evolution computation.With the saturation of commodities and the acceleration of consuming speed, the consumers’demands tend to be more and more various, characteristic and selective. Facing this reality, many enterprises begin to realize the importance of product design and pay more attention to creative products. Creative design can satisfy the novel and ever-changing market demands and help to make the product more competitive. Thus, creative design has been the popular research item in computer-aided design field. In recent years, the rise of the evolution computation and swarm intelligence provide us a new approach for conceptual creative design; therefore, it has become one of the most important techniques for creative conception design.Much attention has been paid to the combination of particle swarm optimization and co-evolution, then to create a multi-particle swarm co-evolution model which supporting creative concept design in this paper. So it can provide a supporting platform of collaboration and creation for distributed designers. The main work is as follows:1. Propose a multi-particle swarm co-evolution algorithmAlthough traditional PSO has been widely applied in many fields because of its simple realization and validity in early of iteration, it has many difficulties on efficiency and precision of finding the optimum of approximate problem. This indicates that traditional PSO loses the variety of particles– all particles are attracted by the best particle which is found as so far. Co-evolution is added into traditional PSO in this paper and a multi-particle swarm co-evolution algorithm is proposed which keeps variety of particles and improves performance of PSO. Numeric experiments prove the validity of this algorithm.2. Construct a multi-particle swarm co-evolution algorithm supporting creative concept designIn view of the uniqueness of creative concept design, a multi-particle swarm co-evolution model is constructed which supporting creative concept design base of previous algorithm. According to the characters of general appearance combined set, a proper fitness value function is designed, through co-evolution of sub-population of screen and sub-population of keyboard, intelligent layout of screen and keyboard on shell of general appearance combined set is realization.3. Apply multi-particle swarm co-evolution model in system of creative design of appearance of combined setIn the system of creative design of appearance of combined set, the model is applied in implementing intelligent layout of general appearance combined set. With HOOPS/NET adopted as the main skeleton, and entity modeling engine ACIS utilized as the sculpt kernel, the system is designed on WindowsXP platform with VC++.NET 2003. In addition, SQL Server 2000 is used as database system. In final stage- components assembling, the model is used to intelligent layout, the design result is satisfied.更多还原