摘要
在传统T-S模糊模型的基础上,提出一种高次多项式T-S模糊辨识模型.将传统T-S模型规则后件中的线性模型用简单多项式模型代替,并进一步利用微粒群优化算法辨识规则后件参数.数值仿真表明:用该方案辨识得到的T-S模糊模型同传统的具有线性后件的T-S模型相比,能够显著减少模型规则条数而保持辨识精度不变,同时辨识时间也相应地缩短;且随着输入变量个数的增加,这一优势将更加明显.
A novel approach for T - S fuzzy model identification was proposed based on the traditional T - S fuzzy model. In more detail, the consequent of the fuzzy rule of T - S model in this research is polynomial model instead of linear or affine ones which is the main feature of traditional T - S model. Based on this candidate T - S fuzzy model, the particle swarm optimization algorithms are employed to estimate the parameters in this model. Numerical simulations demonstrate that the number of fuzzy rules is significantly reduced while the model accuracy is still unchanged. Moreover, the time for identification is shorter. This advantage comes to be more prominent with the increase of input variables.
出处
《哈尔滨工业大学学报》
EI
CAS
CSCD
北大核心
2007年第5期700-702,共3页
Journal of Harbin Institute of Technology
关键词
系统辨识
T—S模糊模型
微粒群优化算法
system identification
T- S fuzzy models
particle swarm optimization algorithms
