Title :
Developing the Theory of a Model-Based Dynamic Recurrent Neural Network
Author :
Karam, Marc ; Zohdy, Mohamed A.
Author_Institution :
Dept. of Electr. Eng., Tuskegee Univ., AL
Abstract :
The theory lying behind a model-based dynamic recurrent neural network (MBDRNN) previously used to improve the linearized models of nonlinear systems is developed in this paper. The MBDRNN is initially based on the linearized system model, and then is trained to represent the system´s nonlinearities by adapting the weights of its nodes´ activation functions using back-propagation . The details of the various computations necessary for a successful operation of the MBDRNN are presented.
Keywords :
backpropagation; linear systems; mean square error methods; multivariable systems; recurrent neural nets; state-space methods; back-propagation; linearized system model; model-based dynamic recurrent neural network; node activation functions; system nonlinearities; Approximation algorithms; Approximation error; Computer networks; Equations; Frequency domain analysis; Neural networks; Nonlinear dynamical systems; Nonlinear systems; Recurrent neural networks; Systems engineering and theory;
Conference_Titel :
System Theory, 2007. SSST '07. Thirty-Ninth Southeastern Symposium on
Conference_Location :
Macon, GA
Print_ISBN :
1-4244-1126-2
Electronic_ISBN :
0094-2898
DOI :
10.1109/SSST.2007.352361
