Title :
Modeling and control based on a new neural network model
Author :
Quan, Yongbing ; Zhang, Huaguang ; Cai, Lilong
Author_Institution :
Dept. of Autom. Control, Northeastern Univ., Boston, MA, USA
Abstract :
We propose a new neural network model - the hyperbolic network model (HNM). Although it is a nonlinear model by nature, we may determine the initial values of the weights by experience and design the controller based on linear system theory. A method of designing a stable controller for the HNM is proposed and some simulation results are given
Keywords :
control system synthesis; modelling; neural net architecture; neurocontrollers; stability; controller design; hyperbolic network model; initial weight value determination; linear system theory; neural network model; nonlinear model; simulation; stable controller design; Algorithm design and analysis; Automatic control; Design methodology; Learning systems; Linear systems; Matrix converters; Mechanical engineering; Neural networks; Nonlinear control systems; Robust stability;
Conference_Titel :
American Control Conference, 2001. Proceedings of the 2001
Conference_Location :
Arlington, VA
Print_ISBN :
0-7803-6495-3
DOI :
10.1109/ACC.2001.946022
