Premium
Fault simulation and diagnosis for vector control of AC motor drive
Author(s) -
Meng Linghui,
Liu Zhigang
Publication year - 2016
Publication title -
ieej transactions on electrical and electronic engineering
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.254
H-Index - 30
eISSN - 1931-4981
pISSN - 1931-4973
DOI - 10.1002/tee.22279
Subject(s) - artificial neural network , fault (geology) , matlab , control theory (sociology) , reliability (semiconductor) , nonlinear system , backpropagation , multivariable calculus , support vector machine , engineering , control engineering , computer science , maxima and minima , control (management) , artificial intelligence , mathematics , power (physics) , mathematical analysis , physics , quantum mechanics , seismology , geology , operating system
An AC drive system is a multivariable, nonlinear, strongly coupled, and complex electromechanical system whose safety and reliability are extremely important in modern electric locomotives. Therefore, it is very important to carry out the converter's fault diagnosis. In this paper, the vector control of an AC motor drive system is modeled and simulated with MATLAB in order to obtain reliable failure data. Through comparative analysis, we extract the effective fault features, which are input to the neural network (NN) to complete the fault diagnosis. Finally, we compare the training and diagnosis results of the Levenberg‐Marquardt (LM)‐back propagation (BP) and BP NNs through simulation experiments and show that the LM‐BP NN has a higher efficiency and higher diagnosis accuracy. Also, it does not fall into local minima and is suitable for pattern recognition. © 2016 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc.