Identification of Pre-existing/Undetected Line-to-Line Faults in PV Array Based on Preturn on/off Condition of the PV Inverter

Abstract

A major challenge in a photovoltaic (PV) system is to identify the line-to-line faults that occurred under low irradiance conditions, during day-to-night/night-to-day transitions in the presence of blocking diodes. Due to active maximum power point tracker, these faults may not be identified and continue to be hidden even if the system returns to a high irradiance condition, since the magnitude of fault current may be less than the rating of the protection device (fuse). Those uncleared faults may subsequently lead to reduced peak output power, reliability issues, and may even cause fire risks. To address these issues, a method is proposed to identify the pre-existing/undetected faults using the transitory conditions that occur during the preturn on/off condition of the PV inverter (checking mode). From these transitory conditions, the I–V curve of the PV array is exploited without the need for additional components or sensors. Additionally, there is no need to disconnect the PV array from the setup. Moreover, the local minima is tracked, to discriminate the faults and partial shading conditions. The proposed algorithm is tested on a small-scale grid-connected PV system and implemented in LabVIEW. The results demonstrate the efficacy of proposed algorithm in detecting the pre-existing/undetected faults.