基于遗传算法的桥式起重机结构进化设计

发布时间：2019-04-26 17:00

【摘要】：随着生物进化论的发展,进化技术成为一种通用的问题求解技术,越来越受到人们的青睐。通过学习进化的过程,解决生活中的复杂问题,生物进化理论便被完美地应用于实际。在所有进化算法中,遗传算法是发展最迅速,研究最热门的一种。遗传算法采用研究目标种群的方式提取基因,组织搜索解空间的多个区域,进行遗传操作与变异,适应度分析,产生较优秀的子代,多次遗传,直至产生理想的最优解。其具有自组织、自学习、自适应等特征,特别适合大规模并行计算,而且进化效率高,操作简单,通用性强。有限元理论已广泛应用于工程实践之中,不但提高了结构分析的精度,节省结构设计的时间,提高设计效率,而且实现了程序化与参数化设计。将进化设计理论与结构有限元分析结合起来,应用于桥式起重机的金属结构设计中,具有显著的科研和经济价值,且能更好的指导工程实际。论文首先深入研究国内外有关遗传算法和结构进化设计的文献资料,概述了结构进化设计的研究历史与发展现状,有限元理论在大型金属结构设计中的重要作用与工程应用；然后运用有限元分析软件ANSYS的APDL语言针对桥式起重机进行参数化建模,并按照起重机设计规范进行工况分析,在此基础上,结合遗传算法建立数学模型,对金属结构已参数化的基因进行遗传进化操作,适应度分析,多代进化,最终得到满足要求且用料最省的最优解,即为最优结构尺寸；最后对论文的研究算法运用VC++6.0进行封装,形成简洁的用户界面,方便工程使用。本文从桥式起重机金属结构设计的实际出发,结合遗传算法与结构有限元分析的各自优点,既实现了金属结构在各种工况下的快速计算,又完成了结构设计基因多方向快速并行优化,最终得到满意最优解,更好地指导工程实际。
[Abstract]:With the development of biological evolution theory, evolutionary technology has become a general problem solving technology, which is more and more popular. By learning the process of evolution and solving the complex problems in life, the theory of biological evolution is perfectly applied to practice. Among all evolutionary algorithms, genetic algorithm is the fastest developing and the most popular one. The genetic algorithm uses the method of studying target population to extract genes, organize searching for multiple regions of solution space, carry on genetic operation and variation, analyze fitness, produce better offspring, inherit many times, and then produce ideal optimal solution. It has the characteristics of self-organization, self-learning, self-adaptation and so on. It is especially suitable for large-scale parallel computing. Moreover, it has high evolutionary efficiency, simple operation and strong generality. The finite element theory has been widely used in engineering practice, which not only improves the precision of structural analysis, saves the time of structural design, improves the efficiency of design, but also realizes programmed and parameterized design. The combination of evolutionary design theory and structural finite element analysis can be applied to the metal structure design of overhead crane, which has remarkable scientific research and economic value, and can better guide the engineering practice. Firstly, the paper deeply studies the literature about genetic algorithm and structural evolutionary design at home and abroad, summarizes the research history and development status of structural evolutionary design, the important role and engineering application of finite element theory in the design of large-scale metal structures. Then using the finite element analysis software ANSYS APDL language to carry on the parametric modeling to the overhead crane, and according to the crane design criterion to carry on the working condition analysis, on this basis, combined with the genetic algorithm to establish the mathematical model, The genetic evolution operation, fitness analysis and multi-generation evolution of genes with parameterized metal structure were carried out. Finally, the optimal solution was obtained, that is, the optimal structural size, which meets the requirements and uses the optimal material. Finally, the algorithm of the paper is encapsulated by VC 6.0.It forms a simple user interface and is easy to use in engineering. Based on the practical design of metal structure of overhead crane, combining the advantages of genetic algorithm and finite element analysis of structure, this paper not only realizes the fast calculation of metal structure under various working conditions, but also combines the advantages of genetic algorithm and finite element analysis of structure. Finally, the fast parallel optimization of structural design gene in multi-direction is completed, and finally the satisfactory optimal solution is obtained, which can guide the engineering practice better.
【学位授予单位】：武汉理工大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TP18;TH215

【参考文献】