Title :
ANN based designing system for industrial induction motors
Author :
Hiyama, Takashi ; Ikeda, Masahiro
Author_Institution :
Dept. of Electr. & Comput. Eng., Kumamoto Univ., Japan
Abstract :
This paper presents an artificial neural network based designing system for industrial induction motors. Based on the actual design data for various types of induction motors, cascaded artificial neural networks have been trained to give better solutions for the fundamental quantities and parameters of individual induction motor. After the training, the selected quantities have been estimated quite accurately from the specification of the individual induction motor on the cascaded artificial neural networks. The proposed neural network based designing system is quite efficient to reduce the designing process and also to have the sub-optimal design even in a restricted time period without the knowledge of experts.
Keywords :
design engineering; induction motors; learning (artificial intelligence); neural nets; power engineering computing; ANN based designing system; cascaded artificial neural network; industrial induction motors; neural network training; power engineering computing; Artificial neural networks; Compressors; Cooling; Fans; Induction motors; Neural networks; Nonlinear equations; Parameter estimation; Rotors; Voltage;
Conference_Titel :
Machine Learning and Cybernetics, 2004. Proceedings of 2004 International Conference on
Print_ISBN :
0-7803-8403-2
DOI :
10.1109/ICMLC.2004.1378512