Diagnosis and Protection of IPM Motors Using Wavelet Packet Transform

This paper presents the practical implementation of a new algorithm for the detection and diagnosis of stator electrical faults in three-phase interior permanent magnet (IPM) motors. Motor current signature analysis is performed off-line on the collected data to extract the most significant features from the fault currents. It has been found that magnitudes of harmonics for motor fault currents can be used to discriminate between different faulted and normal unfaulted conditions. The proposed algorithm utilized the multiresolution property of the WPT, and is based on the identification of fault harmonics using the selected mother wavelet `db3´. The novel wavelet packet transform (WPT) based algorithm for the protection of IPM motors is implemented and tested on-line on the laboratory 1-hp and 5-hp IPM motors. The single phasing fault, single line to ground (L-G) fault and line-to-line (L-L) fault are investigated. In all the tests carried out, the faults are detected, and the trip signal is initiated almost at the instant or within one cycle of the fault occurrence