期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2013
卷号：48
期号：1
页码：336-341
出版社：Journal of Theoretical and Applied
摘要：This paper proposed a method to identify nonlinear systems via the fuzzy weighted least squares support machine (FW-LSSVM). At first, we describe the proposed modeling approach in detail and suggest a fast learning scheme for its training. Because the training sample data of independent variable and dependent variable has a certain error, and we obtain the sample which has a certain fuzziness from measuring, influencing the accuracy of model building, this paper would put the concept of fuzzy membership into the least squares support vector machine. Using the fuzzy weighted least squares algorithm for samples, each sample in the vector is introduced into the fuzzy membership degree, this improve the anti noise ability of least squares support vector machine. The efficiency of the proposed algorithm was demonstrated by some simulation examples.
关键词：Mixed Kernel Function; FW-LSSVM; Identification，Nonlinear Systems
