Novel and comprehensive approach for power loss estimation of soiled photovoltaic modules

Abstract

The output power of photovoltaic (PV) generators is negatively affected by shading caused by surface soiling. Thus, it is crucial to evaluate the power loss level and plan cleaning schedules efficiently and economically, particularly in large-scale PV power plants. This paper proposes a novel and comprehensive solution that combines circuit topology analysis and image processing to estimate power loss in soiled PV panels. The performance evaluation is based on a wide range of soiling conditions, including sand, dust, and talcum powder, and is compared with a benchmarking system from prior studies. Compared with the best-performing data-driven-only method, the proposed scheme improves the estimation accuracy from 11.02% to 39.52% by exploring the circuit information.