Estimating Power Generation of Invisible Solar Sites Using Publicly Available Data

Abstract

Large-scale integration of invisible solar photovoltaic generation into power systems could significantly affect the system net load and pose new challenges in the operation of power systems. Invisible solar photovoltaic refers mainly to small-scale roof-top solar sites that are not monitored, and thus are invisible to utilities and system operators. Invisible solar generation affects the shape of system net electrical load and could make net load forecasting more challenging. In this paper, a methodology is proposed to estimate the power generation of invisible solar photovoltaic sites. The proposed method only uses the measured power generation data of publicly available sites. It uses real time data of a small subset of sites to estimate the aggregated power generation from known sites within a region. The proposed model is validated using actual invisible solar generation data of the California power system.