摘要
针对一类具有切换结构的混杂系统,提出一种基于仿射传播聚类的模型辨识方法。将模型辨识问题等价成对系统数据的分类和分类数据的回归问题。通过仿射传播聚类算法对样本数据进行聚类划分,并分别采用最小二乘支持向量机算法对子样本分别建立模型。仿真结果验证了该方法的有效性和实用性。
For a class of hybrid systems with the switched form, a model identification method based on affinity propagation clustering is presented. In this method, the model identification is- sue is equivalent to the problems of classification of the system data and regression of the classifi- cation data. The affinity propagation clustering algorithm is successively applied to group the da- ta into clusters, and least squares support vector machines is used to construct sub -model. Simulation results show the effectiveness and feasibility of the proposed method.
出处
《大连民族学院学报》
CAS
2013年第5期547-550,共4页
Journal of Dalian Nationalities University
基金
中央高校基本科研业务费专项资金资助项目(DC120103070)
关键词
切换系统
仿射传播聚类
最小二乘支持向量机
switched systems
affinity propagation clustering
least squares support vector ma-chines
