期刊名称:International Journal of Advanced Research In Computer Science and Software Engineering
印刷版ISSN:2277-6451
电子版ISSN:2277-128X
出版年度:2012
卷号:2
期号:11
出版社:S.S. Mishra
摘要:Artificial neural networks have been studied for many years in the hope of achieving human like performance in various fields of engineering. This paper details a proposed scheme to identify the use of neural network model for the estimation of the speed of the rotating shaft driven by voltage controlled induction motor drive. The speed identification of rotating shaft is based on the conventional analog type of electro -mechanical sensors like tachometer, optical pyrometer etc. The conventional adaptive control schemes are complicated and need excessive computational effort for real time implementation. This paper proposes how the highly parallel building blocks that illustrate neural net components and design principles can be used to track more systems like in speed identification of induction motor drive. Computational elements or nodes are connected via weights that are typically adapted to improve performance. The decisions region required by any classification algorithm can be generated in a straight-line forward manner by three-layer feed-forward nets. It details about the preferable use of contact less ANN sensors over the conventional ones, which are mounted on the shaft of the rotating motor. Moreover, the paper gives relevant data about the simulation, training and testing of an ANN network. The verification of the proposed work through physical experimentation obviously suggests the use of neural networks to solve the above problems by mimicking the adaptive control architecture in human brain