摘要
通常离线提取T-S模糊模型的规则后,规则数无法在模型使用中进行调整,而这成为表达非线性系统复杂性的一个瓶颈.针对这一问题,本文引入一种神经网络的生长和修剪方法,从实时数据中提取T-S模型的规则,并定义其对应局部模型对输出的影响,以此作为在线调整规则数的依据,从而更准确地表达了非线性系统的复杂性和运行中的变化.再加上基于竞争性EKF(Extended Kalman filter)的模型参数在线学习,T-S模型的建模精度也得到了保证.整个算法完全实现了T-S模糊模型的在线辨识,使模型的结构和参数具有很好的自适应能力.对CSTR(Continuously stirred tank reactor)系统的辨识,表明了该算法在处理非线性系统辨识问题上的优越性能.
Offline rule extraction for the T-S fuzzy systems usually gives a fixed number of fuzzy rules, which make it a bottleneck for revealing the complexity of nonlinear systems. Thus, due to a growing and pruning strategy of the neural network, in this paper the fuzzy rules are extracted from real-time data and their number is adjusted online by the impact degree of one local model, such that the rules vary with the system dynaxaically and more precisely reflect the character of nonlinear systems. Furthermore, the accuracy of the T-S model is guaranteed by the parameter learning based on a competitive extended Kalman filter （EKF）. The entire algorithm presents a completely online identification of the T-S model and gains a structural and parameter adaptability. An example for CSTR identification illustrates its good performance.
出处
《自动化学报》
EI
CSCD
北大核心
2007年第10期1097-1100,共4页
Acta Automatica Sinica
基金
国家自然科学基金(60474051)
高等学校博士点专项科研基金(20060248001)
国家教育部新世纪优秀人才计划资助~~
关键词
T—S模型
模糊规则
生长与修剪
平均响应
在线辨识
T-S model, fuzzy rule, growing and pruning, average response, online identification
