Title :
Research on nonlinear adaptive inverse control using FLANN data-together constructed by LS-SVM
Author :
Zhang, Jinmin ; Meng, Ping ; Wang, Siming
Author_Institution :
Dept. of Mechatron. Eng., Univ. of Lanzhou Jiaotong, Lanzhou, China
Abstract :
With the study on the constructive method of generic functional link artificial neural network(FLANN), a novel constructive method based on the least squares support vector machines (LS-SVM) is proposed and applied to building the nonlinear object model and inverse model. The nonlinear system identification technology of this method is applied to the adaptive inverse control to increase the self-adaptability of the nonlinear system and improve the dynamic properties. The adaptive inverse control system designed can not only get a better dynamic response but also reduce the disturbance to a minimum.
Keywords :
adaptive control; control system synthesis; least squares approximations; neural nets; nonlinear control systems; nonlinear dynamical systems; self-adjusting systems; support vector machines; FLANN data; LS-SVM; constructive method; dynamic property; dynamic response; generic functional link artificial neural network; inverse model; least square support vector machine; nonlinear adaptive inverse control system design; nonlinear object model; nonlinear system identification technology; nonlinear system self adaptability; Adaptation model; Adaptive systems; Artificial neural networks; Control systems; Heuristic algorithms; Mathematical model; Nonlinear dynamical systems; adaptive inverse control; functional link artificial neural network (FLANN); least squares support vector machines (LS-SVM);
Conference_Titel :
Electric Information and Control Engineering (ICEICE), 2011 International Conference on
Conference_Location :
Wuhan
Print_ISBN :
978-1-4244-8036-4
DOI :
10.1109/ICEICE.2011.5778236