A Hierarchical Intelligent Fault Detection and Location Scheme for DC Ring Bus Microgrid

Abstract

A hierarchical fault detection and fault location framework consisting of two levels is proposed for DC ring bus microgrids. In the primary level, a decentralized fault detection scheme based on back propagation (BP) artificial neural network (ANN) and discrete wavelet transform (DWT) is proposed to classify the pole-to-pole (P-P) and pole-to-ground (P-G) faults under measuring noises and load disturbances only relying on the information needed by the local double closed-loop controller, and the diagnosis speed and reliability of system are guaranteed. In the secondary level, a distributed line fault location method is presented by synthesizing the detection results of local and neighbor DG units with low bandwidth communication, and the accuracy of line faults is improved greatly without the central controller and additional sensors. Several simulation cases are analyzed under various fault configurations such as load mutations, different fault types, fault locations and fault impedance. The effectiveness and feasibility of the proposed method are verified.