Title :
Robustness test of an incipient fault detector artificial neural network
Author :
Chow, Mo-Yuen ; Yee, Sui Oi
Author_Institution :
Dept. of Electr. & Comput. Eng., North Carolina State Univ., Raleigh, NC, USA
Abstract :
Addresses the issue of robustness in artificial neural networks subject to small input perturbations. The robustness in artificial neural networks is studied using the concept of input-output sensitivity analysis applied to an incipient fault detector artificial neural network (IFDANN). The IFDANN was designed to detect winding insulation faults and bearing wear in single-phase squirrel-cage induction motors. Modification of the IFDANN, with the intention of increasing its robustness to input noise during real-time applications, is discussed. Analytical and simulation results are presented to show the significant improvement in robustness of the modified IFDANN for operation with noisy measurements
Keywords :
computer testing; electrical engineering computing; fault location; insulation testing; machine windings; neural nets; sensitivity analysis; squirrel cage motors; wear; IFDANN; artificial neural networks; bearing wear; incipient fault detector; input perturbations; input-output sensitivity analysis; noisy measurements; real-time applications; robustness; single-phase squirrel-cage induction motors; winding insulation; Artificial neural networks; Electrical fault detection; Fault detection; Induction motors; Neural networks; Noise measurement; Noise robustness; Sensitivity analysis; Testing; Working environment noise;
Conference_Titel :
Neural Networks, 1991., IJCNN-91-Seattle International Joint Conference on
Conference_Location :
Seattle, WA
Print_ISBN :
0-7803-0164-1
DOI :
10.1109/IJCNN.1991.155152