1. NONLINEAR MODELING OF LABORATORY MODEL AMIRA DR300 BY RECURRENT NEURAL NETWORK.
- Author
-
Hubacek, Jiri, Bobal, Vladimir, and Brazdil, Michal
- Subjects
- *
ARTIFICIAL neural networks , *RECURSIVE sequences (Mathematics) , *COUPLINGS (Gearing) , *GENERATORS (Computer programs) - Abstract
Purpose of this paper is to design a nonlinear recurrent neural network model of a laboratory device AMIRA DR300 made by a company AMIRA in Germany. This apparatus consists of two direct-current engines, whose shafts are connected together by a fixed shaft coupling. A shaft coupling rotation speed is an output value of this laboratory model and it is possible to measure it either by a tachometer generator or an incremental position sensor. Recurrent neural network was trained to achieve similar properties to the first engine which is called as motor. The second engine can be used as a generator of a faulty measured value. [ABSTRACT FROM AUTHOR]
- Published
- 2011