Abstract
This paper presents a wavelet-based method for transient disturbance analysis in oscillographic records of the transmission power systems. The proposed method can recognize faults and other disturbances with transients, such as voltage sags and switching transients in power systems. Developing faults and multiple transient disturbances in a single oscillographic record, such as a fault followed by an automatic line reclosing are also detected. The proposed method is based on the analysis of the wavelet coefficient energies of the phase voltages and currents achieved by sliding data windows with length equivalent to one cycle of the fundamental power frequency. These energy waveforms are normalized and records from a nonspecific line in a generic network with different sampling rates and rated voltages can be evaluated without distinction. The performance of the method was evaluated for both actual records from a Brazilian Power System and records simulated on Real Time Digital Simulator (RTDS) and good results were obtained.
