Publication detail
Neuron Models of Mechatronic Systems
KLÍMA, M. VEČERKA, J.
English title
Neuron Models of Mechatronic Systems
Type
conference paper
Language
en
Original abstract
This paper presents modeling of the induction machine by using artificial neural networks (ANN). The architecture of proposed model is optimized through the change of number of inputs and number of neurons in the hidden layer. This approach is computationally very time-consuming, but this problem was reduced using the Parallel Virtual Machine (PVM) software. This software allows integration of a computational power of several network-connected computers into one virtual machine. Due to fact that adaptation of neural network is an independent process, computational power of virtual machine is proportional to a number of used computers.
English abstract
This paper presents modeling of the induction machine by using artificial neural networks (ANN). The architecture of proposed model is optimized through the change of number of inputs and number of neurons in the hidden layer. This approach is computationally very time-consuming, but this problem was reduced using the Parallel Virtual Machine (PVM) software. This software allows integration of a computational power of several network-connected computers into one virtual machine. Due to fact that adaptation of neural network is an independent process, computational power of virtual machine is proportional to a number of used computers.
Keywords in English
Neural network, induction machine, stator current, optimization, neural network architecture
Released
01.01.2000
Publisher
Institute of Theoretical and Applied Mechanics, Academy of Science
Location
Praha
ISBN
80-86246-06-X
Book
Engineering Mechanics 2000
Pages count
6
BIBTEX
@inproceedings{BUT4134,
author="Michal {Klíma} and Jiří {Večerka},
title="Neuron Models of Mechatronic Systems",
booktitle="Engineering Mechanics 2000",
year="2000",
month="January",
publisher="Institute of Theoretical and Applied Mechanics, Academy of Science",
address="Praha",
isbn="80-86246-06-X"
}