Power cable incipient faults are single phase short circuits with short duration and little effect on system operation primarily, so detection of incipient faults is needed to keep them from developing into permanent faults. The over-current characteristics caused by different disturbances are compared and the over-currents characteristic quantities are extracted by multi-scale wavelet transform. Fault current high frequency detail coefficients, low frequency approximation coefficients and the maximum module are utilized as compound rules to detect the transient process of cable over-currents. Time-frequency eigenvectors of cable incipient faults are created and the gray correlation analysis is used to obtain the relevancy of fault currents between fault currents of under test and standard samples in order to identify the incipient faults.