机械系统动力优化设计的自适应混合遗传算法

针对机械系统动力优化设计问题的特点，本文提出了一个自适应混合遗传算法。该方法将目适应控制参数调整策略、小生境技术和单纯形法有机地融入遗传算法，较为有效地解决了基本遗传算法的过早收敛和局部搜索能力弱的问题，提高了算法全局寻优的可靠性和收敛速度，为解决机械系统动力优化设计问题提供了一种可靠、有效和易用的优化方法。一个五自由度汽车悬挂系统的动力优化设计实例验证了算法的效率、可靠性和实用性。

Adaptive Hybrid Genetic Algorithm for Dynamic Optimization Design of Mechanical System

An adaptive hybrid genetic algorithm is proposed to solve the dynamic optimization problems, which are nonlinear and non-convex generally in mechanical design. The niche and adaptive crossover and mutation probability are used to maintain the diversity and sustain the convergence in the proposed method. Nelder-Mead's Simplex Methods are used as a local search operator of genetic algorithms to exploit local information in an efficient way. So the proposed method can obtain better results of global search and high quality solutions effectively. A numerical example of vehicle suspension system optimization is employed to demonstrate the efficiency,validity and availability of the algorithm.