A novel method to predict Non-uniform illumination pattern on a large scale rooftop SPV array

Abstract

Incident irradiance plays a significant role in power generation on a stand-alone solar photovoltaics (SPV) array. Due to sun movement, non-uniform illumination (NUI) is an inevitable phenomenon on stand-alone rooftop SPV array. This results loss in power generation as well as affects the long term reliability of those SPV modules. It can be overcome, by incorporating solar tracking arrangement across each and every SPV module. But it would put a limit to the no of mounted SPV modules on a rooftop with a certain area. Hence stand-alone SPV array with integrated sensor network is recommended for monitoring the NUI and corresponding generated power on that SPV array. But these integrated sensors in each SPV modules comprise several challenges like localization of sensor nodes, power consumption, data communication by the sensor network etc. In this work alternatively a camera is mounted on an autonomous flying robot (an aerostat or a quadcopter) or on a pole is to monitor and predict the NUI pattern on that SPV array. It replaces large numbers of sensors by a single camera which assists in optical sensing and track the movement of NUI pattern on that rooftop. Simultaneously it helps in enabling the reconfiguration algorithm to the SPV array for optimized power generation.