DETECTION OF SYNCHRONOUS MOTOR INTER-TURN FAULTS BASED ON SPECTRAL ANALYSIS OF PARK'S VECTOR

This paper describes detection of synchronous motor inter-turn faults based on symptoms contained in stator phase currents. Armature short circuit, caused by insulation degradation are quite commonly occurring defects in electrical machines. Initially, short circuit comprises mostly single coils, causing the temperature rise due to higher value of current, which can reach up to tens times of the rated value. At the same time the phase current does not increase significantly. Increased temperature leads to rapid damage of the insulation and shorting the adjacent coils spreading the fault to the entire winding in a short time. Thus, it is very important to detected this type of fault in its early stage. Unfortunately currently used motor protection devices are insensitive to short-circuits of a small number of turns, because they cause too small quantitative changes in the phase currents. Phase currents begin to rise to the level detectable by protection devices when a large part of the winding is already covered by a fault. Therefore, there is a need for research on diagnostics of this type of damage. For the purpose of this paper a stepped short circuit fault of one coil group in the stator phase winding is performed. Shorting resistance values are chosen so that the short fault is diagnosed in its early stage. Spectral analysis of stator phase currents is carried out followed by spectral analysis of stator currents Park's vector. Comparison of the results of both studies shows that the signal of stator current Park's vector is more suitable in diagnostics of this type of faults.