基于支持向量回归机和粒子群算法改进协同优化方法.docVIP

基于支持向量回归机和粒子群算法改进协同优化方法 　　摘 要： 研究基于支持向量回归机和粒子群算法的改进协同优化方法.阐述了协同优化方法和支持向量回归机方法基本原理，为有效解决系统级优化协调困难问题，改善收敛性能，提高收敛速度，采用支持向量回归机构造系统级约束条件的近似模型，引入粒子群算法求解系统级和学科级优化问题.仿真计算结果表明，设计的协同优化方法可有效求解多学科设计优化问题，和基本协同优化方法相比，求解精度高，优化迭代次数少，稳定性好.可为多学科设计优化研究提供理论参考. 关键词： 协同优化；支持向量回归机；粒子群算法 　　 中图分类号：V412.1 文献标识码：A 　　 　　Improved Collaborative Optimization Based on Support Vector Regression and Particle Swarm Optimization 　　YANG Xixiang, YANG Huixin, JIANG Zhenyu, ZHANG Weihua 　　(Colloge of Aerospace and Material Engineering, National Univ of Defense Technology, Changsha, Hunan 410073,China) 　　Abstract:Improved collaborative optimization based on support vector regression and particle swarm optimization algorithm was researched. The basic principle of collaborative optimization and support vector regression was represented, and in order to resolve the difficulty in systemlevel coordination, improve convergence performance and efficiency, approximate models of constraint conditions in systemlevel were constructed using support vector regression, and particle swarm optimization algorithm was introduced to the systemlevel optimization and disciplinarylevel optimization. Simulation results show that the improved collaborative optimization can effectively resolve multidisciplinary design optimization problems, and compared to standard collaborative optimization, optimization accuracy is higher, systemlevel iterative operation is less, and the stability is better. All those can provide theoretical reference for the research of multidisciplinary design optimization. 　　 Key words: collaborative optimization; support vector regression; particle swarm optimization 　　 　　协同优化方法（Collaborative Optimization，CO）是应用广、效果好的多学科设计优化方法，由斯坦福大学Kroo教授于1994年首先提出.CO按学科对设计优化问题进行分解、协调，每个学科优化计算都具有很好的自制性，不必考虑其余学科影响，本质上是一种分布式并行优化决策过程，可大大缩短设计周期；学科间解耦、协调问题由系统级一致性体现，避免了求解非线性耦合方程组困难.CO的上述显著优点使其近年来在多学科设计优化领域受到高度关注和广泛应用[1-2]. 　　CO消除了复杂系统分析过程，但系统级优化的约束条件由各学科优化结果决定，系统级优化过程需要不断调用各学科优化器以计算其约束条件，同时，当子系统最优解靠近系统级期望最优解时，系统级约束的梯