Informatics Publishing Limited

Chaturvedi D. K. ¹, Devendra Singh ², Vikas Pratap Singh ³

Affiliations
1 Department of Electrical Engineering, Faculty of Engineering, Dayalbagh Educational Institute, Agra-282110, IN
2 M.Tech. Student in the Department of Electrical Engineering, Faculty of Engineering, Dayalbagh Educational Institute, Agra-282100, IN
3 Vikas Pratap singh, Energy Efficiency and Renewable Energy Division, Central Power Research Institute, Bangalore and IIT Jodhpur, IN

Abstract

A variety of fan faults occur in our day to day life such as electrical faults(winding faults), mechanical faults (broken rotor bars, eccentricity, bearing faults) etc. To detect the fault, many motor variables may be taken such as current, voltage, speed, sound, temperature and vibrations, so that the preventive action may be taken before the occurrence of faults in the fan. Current signature is useful for finding electrical faults such as stator faults etc. and acoustic signature is useful for finding mechanical faults such as rotor faults etc. In this paper, the on line current, voltage, rpm and temperature reading of faulty fan and healthy fan are recorded. These recorded signals are used to train a neural network so that it is able to detect the fault.

Keywords

Wavelet, ANN, Ceiling Fan, Fault detection

Full Text