基于改进的LSSVM辨识动态非线性时变系统  

Identification of dynamic nonlinear systems based on modified LSSVM

作　　者：杜志勇[1] 王鲜芳[2] 郑丽媛[3] 

机构地区：[1]河南机电高等专科学校,河南新乡453002 [2]江南大学通信与控制工程学院自动化研究所,江苏无锡214122 [3]辽宁工程技术大学电气与控制工程学院,辽宁葫芦岛125105

出　　处：《计算机工程与应用》2010年第21期238-241,共4页Computer Engineering and Applications

基　　金：国家高技术研究发展计划(863)No.2006AA020301-11;河南省教育厅自然科学研究项目(No.2007510006)~~

摘　　要：针对非线性时变系统难以辨识的问题,提出了一种基于改进最小二乘支持向量机的辨识新方法。该方法在加权最小二乘支持向量机的基础上,引入用矢量基学习和自适应迭代相结合的方式得到一个小的支持向量,同时采用加权方法确定权值系数以减小训练样本中非高斯噪声的影响。通过对动态非线性时变系统的仿真,结果表明该算法具有较好的鲁棒性、支持向量稀疏性和动态建模实时性。According to the problem about difficult to identify the nonlinear time-varying system,a new identification meth-od is proposed based on modified Least Squares Support Vector Machines（LSSVM）.During the algorithm＇s training process,the vector base learning and automatic iterative procedures are introduced based on weighted least squares support vector machines,and a small support vector set can be obtained adaptively.Meanwhile the weights are determined by a robust method in order to reduce the effect of the outliers.Simulating for a nonlinear time-varying system,the result shows that the proposed method has a better robustness,sparseness of support vector and a real-time performance.

关 键 词：矢量基 加权最小二乘支持向量机 支持向量稀疏性 自适应迭代算法 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]