A wavelet-based algorithm for power quality detection in electric arc furnace

During a fusion process, the electric arc furnaces present high consumption of electric power, unbalanced current signal waveform, with distortion due to harmonics distortion and the flicker effects; as well as sags and swells in the voltage signals. For the analysis of the power quality problems caused by arc furnaces, it is usually necessary to detect and classify them. With this purpose the technique of the Wavelet Transform will be used to non-stationary signals of an industrial plant with three arc furnaces. The papers presented so far deal exclusively with theoretical signals, containing only one kind of disturbance. This paper presents an alternative to the analysis of signals with multiple disturbances, often found in industrial electrical systems with electric arc furnaces.