|
|
Fuzzy neural network control method with compensation
for time-delay system |
PAN Hai-peng, LV Yong-song |
Institute of Automation, Zhejiang SciTech University, Hangzhou 310018, China |
|
|
Abstract The traditional fuzzy neural network (FNN) controller has the disadvantage of long setting time for timedelay system. A new method was proposed to obtain the compensation value aimed at the problem. The parameters for timedelay model were identified online based on the recursive least square (RLS) algorithm. Then the output of the system can be predicted, and the compensation value was obtained. An FNN controller of twodimension input with compensation was designed. Simulation results show that the method has the characteristic of short setting time, high control precision and obvious improvement compared with the traditional FNN controller.
|
Published: 01 July 2010
|
|
时滞系统的模糊神经网络补偿控制
针对传统的模糊神经网络(FNN)在线控制方法用于控制时滞对象时存在调节时间较长的问题,分析产生这一现象的原因,对传统的模糊神经网络在线控制进行改进,给出一种新的确定补偿量的方法.基于递推最小二乘(RLS)法在线辨识对象模型,通过时滞对象模型预测对象输出的变化,利用补偿方法得到控制量的补偿量.设计二维输入的带补偿的模糊神经网络控制器,进行实验与仿真研究.仿真结果表明,该补偿方法调节时间短,控制精度高,比传统的模糊神经网络的控制效果明显.
|
|
[1] 袁宇浩,张庆灵,陈兵.非线性广义系统时滞相关的模糊控制[J].自动化学报,2006,32(5): 824828.
YUAN Yuhao, ZHANG Qingling, CHEN Bing. Delaydependent fuzzy control for nonlinear descriptor systems [J]. Acta Automatica Sinica, 2006,32(5): 824828.
[2] 张建海,张森林,刘妹琴.离散时滞标准神经网络模型的鲁棒稳定性分析[J].浙江大学学报:工学版,2009,43(8): 13831388.
ZHANG Jianhai, ZHANG Senlin, LIU Meiqin. Robust stability analysis of delayed discretetime standard neural network [J]. Journal of Zhejiang University: Engineering Science, 2009,43(8): 13831388.
[3] 韩安太,王树青.基于LMI的一类非线性时滞关联大系统的分散模糊控制[J].控制与决策,2004,19(4): 416428.
HAN Antai, WANG Shuqing. Decentralized fuzzy control for a class of nonlinear interconnected largescale systems with timedelay based on LMI approach [J]. Control and Decision, 2004,19(4): 416428.
[4] 王立新.模糊系统与模糊控制教程[M].北京:清华大学出版社,2003: 70118.
[5] JANG J S R, SUN C T. Functional equivalence between radial basis function networks and fuzzy inference systems [J]. IEEE Transactions on Neural Networks, 1993, 4(1): 156158.
[6] CASTELLANO G, FANELLI A M, MENCAR C. A neuronfuzzy network to generate humanunderstandable knowledge from data [J]. Cognitive Systems Research, 2002, 3(2): 125144.
[7] 李明忠,王福利.基于递归神经网络的一类非线性无模型系统的自适应控制[J].控制与决策,1997,12(1): 6472.
LI Mingzhong, WANG Fuli. Adaptive control of black box nonlinear systems using recurrent neural networks [J]. Control and Decision, 1997,12(1): 6472.
[8] KORBICZ J, KOWAL M. Neuralfuzzy networks and their application to fault detection of dynamical systems [J]. Engineering Applications of Artificial Intelligence, 2007, 20(5): 609617.
[9] SHIEH H L, YANG Y K, CHANG P L, et al. Robust neuralfuzzy method for function approximation [J]. Expert Systems with Applications, 2009, 36(3): 69036913.
[10] 姜永森,田军,杨慧中.双率时变系统遗忘因子最小二乘参数估计[J].清华大学学报:自然科学版,2008,48(S2): 17231727.
JIANG Yongsen, TIAN Jun, YANG Huizhong. Parameter estimation for the forgotten factor least squares algorithm for dualrate timevarying systems [J]. Journal of Tsinghua University: Science and Technology, 2008,48(S2): 17231727.
[11] DING Feng, XIAO Yongsong. A finitedatawindow least squares algorithm with a forgetting factor for dynamical modeling [J]. Applied Mathematics and Computation, 2007, 186(1): 184192. |
|
Viewed |
|
|
|
Full text
|
|
|
|
|
Abstract
|
|
|
|
|
Cited |
|
|
|
|
|
Shared |
|
|
|
|
|
Discussed |
|
|
|
|