基于最小二乘支持向量机的T-S模型在线辨识
LSSVM-Based Online Identification for T-S Model
-
摘要: 提出一种基于时间窗最小二乘支持向量机的T-S模型在线辨识算法,包括结构辨识和参数辨识.该算法以时间窗内数据的势能作为结构辨识依据,同时采用最小二乘支持向量机辨识系统参数,具有辨识速度快、精度高的特点.仿真结果证明了算法的有效性.Abstract: An online identification method based on least square support vector machine (LSSVM) with time window for T-S model is proposed,including structure and parameter identifications.Structure identification depends on the potential of data in the time window,and parameter identification is based on LSSVM,resulting in high identification speed and precision.The simulation result illustrates the effectiveness of the proposed method.