摘要
提出了一种新的基于T-S模糊模型的建模方法,首先通过一种局部线性聚类算法,自适应确定模糊规则数目及初始T-S模型的前提和结论参数,建立相应的一阶T-S模糊神经网络.并用梯度下降和递推最小二乘混合算法训练网络参数,从而提高建模精度.最后,通过两个仿真实例验证了本文方法的有效性.*
Based on Takagi-Sugeno fuzzy model, a new modeling method is proposed. Firstly, a local linear clustering algorithm is used to determine the number of fuzzy rules and the T-S fuzzy model initial and last parameters, eonsequently the eorresponding 1st order T-S fuzzy neural network is established. Then a hybrid algorithm consisting of a gradient deseent algorithm and reeurrent least square algorithm is implemented to tune the network paramters, so as to improve aeeuraey of the fuzzy modeling. Finally, the effectiveness of the proposed method is demonstrated with two simulation examples.
出处
《信息与控制》
CSCD
北大核心
2006年第5期588-592,599,共6页
Information and Control
