小生境技术在动力学系统参数辨识中的应用研究  

Application and Research of Parameter Identification of Dynamics System Using Niche Genetic Algorithm

作　　者：周长峰[1] 贾倩[1] 张恒海[1] 

机构地区：[1]山东交通学院汽车工程系,山东济南250023

出　　处：《山东交通学院学报》2008年第3期1-4,10,共5页Journal of Shandong Jiaotong University

摘　　要：提出了基于小生境技术的动力学系统参数辨识方法。针对普通遗传算法收敛速度慢的特点,采用确定性选择机制、自适应交叉和变异算子以及归一化海明距离提高算法的运行效率,参数辨识和函数优化的实例表明小生境自适应遗传算法的搜索和优化能力得到很大提高。提出的参数辨识方法具有良好的多峰值寻优能力,可广泛应用于黑箱问题识别和系统优化的相关研究。A niche adaptive genetic algorithm is introduced to resolve the parameter identification and functional optimization problem. The deterministic sampling method, adaptive crossover and mutation operator are utilized to ensure the stable convergence, and the niche technology is used to enhance the multimodal search ability around every peak value. The study on parameter identification of rubber isolator and multimodal function optimization shows that the proposed method can obtain better results of global search and high quality solutions effectively. The method proposed in this article can be applied in the parameter identification of black box and structural optimization design widely.

关 键 词：小生境 遗传算法 参数辨识 动力学系统 

分 类 号：O313[理学—一般力学与力学基础] TH113[理学—力学]