Development of a Fault Detection and Localization Algorithm for Photovoltaic Systems

Abstract

Photovoltaic systems provide electrical power with reduced emissions at competitive costs compared to legacy systems. A low or medium voltage dc distribution system is usually used for solar integration. In dc systems, parallel and series arc faults are a safety concern. Thus, reliable and timely detection and mitigation of arc faults are critical. DC arc detection methods typically use time or frequency spectrum variations of the circuit current or voltage to differentiate the arcing event from other system events. Since practical systems include power electronics and maximum-power-point tracking, any detection scheme must perform robustly in the electrical environment that these components establish in the dc power system. A capacitor placed in parallel with the main system is an effective sensor for series arc fault detection and localization applicable in this complex electrical environment. This article shows that the analysis of the amplitude, polarity, and spectrum characteristics of the capacitor current and voltage resulting from perturbations caused by the arc provides an effective method to identify and localize faults. The detection accuracy of the proposed approach is 98.3% and the localization accuracy rate is 100% for the correctly detected faults.