Output details

Fault diagnosis of motor drives using stator current signal analysis based on dynamic time warping

Dynamic time warping is a technique that has recently found considerable favour amongst acousticians for the processing of voice data. This paper takes this approach, modifies it quite extensively for use with high sample rate parallel data channels of non-acoustic data, and then applies it to the detection and diagnosis of faults in variable speed slip-compensated electric motor drives. The results permit a sensitivity and diagnostic sophistication far in excess of what is achievable with other approaches, and the method is presently being readied for industrial application in a collaborative research project with a multinational manufacturing company.