Self-evolving function-link interval type-2 fuzzy neural network for nonlinear system identification and control
Self-evolving function-link interval type-2 fuzzy neural network for nonlinear system identification and control
复制标题
DOI：
10.1016/j.neucom.2017.11.009
复制
发表时间：
2017-07
期刊：
影响因子：
6
通讯作者：
Chih-Min Lin;Tien-Loc Le;Tuan-Tu Huynh
中科院分区：
文献类型：
--
作者：
Chih-Min Lin;Tien-Loc Le;Tuan-Tu Huynh
确定模糊神经网络结构的网络规模是非常重要的，而且往往很难获得最合适的值。提出了一种自进化函数-链接区间二型模糊神经网络(SEFT2FNN)，该网络利用初始空函数和隶属度函数自主构造规则库。将函数链应用于区间二型模糊神经网络，以得到更精确的函数逼近。利用最陡下降梯度法推导出了系统的自适应律。利用Lyapunov函数方法保证了系统的稳定性。最后，通过对非线性系统辨识和时变对象控制的数值仿真，验证了所提系统的性能。
Determining a network size for a fuzzy neural network structure is very important, and it is often difficult to obtain the most suitable value. This study develops a self-evolving function-link interval type-2 fuzzy neural network (SEFT2FNN) that autonomously constructs the rule base with the initial empty and the membership functions. The function-link is applied to an interval type-2 fuzzy neural network to give a more accurate approximation of the function. The adaptive laws for the proposed system are derived using the steepest descent gradient approach. The stability of system was guaranteed using Lyapunov function approach. Finally, the performance of the proposed system is verified using the numerical simulations of the nonlinear system identification and the control of time-varying plants.