A method for fault feature extraction and diagnosis of dry-type air core reactor is proposed based on a model with time-varying parameters

by analysis on the variation law of state parameters in early process of reactor turn to turn short circuit fault. First

a time-varying fault circuit model based on Mayr arc model is established

and the spectrum characteristics of the equivalent inductor of the reactor is analyzed by using Fourier transform. The equivalent inductor is no longer a constant but a periodic oscillating variate. Then

the transfer solution

zero-state response and frequency spectrum characteristics of the output current of the fault reactor under power frequency excitation voltage are analyzed by using the solution method of the time-varying circuit's state equation. It is proved that during the periodic short circuit of the interturn arc

the output current of the fault reactor will excite a certain number of harmonic components

analyzing the current harmonic characteristics can diagnose the interturn short circuit fault of the dry-type air air core reactor. By analyzing the CT and PT recorded data of a reactor fault

the feasibility and effectiveness of this proposed method for fault feature extraction and diagnosis of the dry-type air core reactor are confirmed.