Detection and Identification of Power Disturbance Signals Based on Nonlinear Time Series

Power disturbances are increasing with the proliferation of non-linear devices. This paper introduces a novel idea for analyzing disturbance signals by treating them as nonlinear time series. Fractal number measurement and phase space reconstruction are two kinds of time series analytical tools widely used in nonlinear systems. Eight particular disturbance signals are selected and then analyzed using these two tools. Fractal number measurements are applied to detect transient disturbances and locate arising time, and phase space reconstruction, to identify and classify various types of disturbance signals. The results show that the methods proposed in this work are effective and intuitive