Title :
Continual on-line training of neural networks with applications to electric machine fault diagnostics
Author :
Tallam, Rangarajan M. ; Habetler, Thomas G. ; Harley, Ronald G.
Author_Institution :
Sch. of Electr. & Comput. Eng., Georgia Inst. of Technol., Atlanta, GA, USA
fDate :
6/23/1905 12:00:00 AM
Abstract :
An online training algorithm is proposed for neural network (NN) based electric machine fault detection schemes. The algorithm obviates the need for large data memory and long training time, a limitation of most AI-based diagnostic methods for commercial applications, and in addition, does not require training prior to commissioning. Experimental results are provided for an induction machine stator winding turn-fault detection scheme that uses a feedforward NN to compensate for machine and instrumentation nonidealities, to illustrate the feasibility of the new training algorithm for real-time implementation
Keywords :
asynchronous machines; automatic test software; computerised monitoring; fault diagnosis; feedforward neural nets; learning (artificial intelligence); machine testing; power engineering computing; real-time systems; commercial applications; electric machine fault diagnostics; fault detection schemes; feedforward neural net; induction machine stator winding turn-fault detection scheme; neural networks continual online training; real-time implementation; training algorithm; Current measurement; Electric machines; Fault detection; Feedforward systems; Induction machines; Mathematical model; Neural networks; Stators; Table lookup; Voltage;
Conference_Titel :
Power Electronics Specialists Conference, 2001. PESC. 2001 IEEE 32nd Annual
Conference_Location :
Vancouver, BC
Print_ISBN :
0-7803-7067-8
DOI :
10.1109/PESC.2001.954450
