Automating the maintenance of photovoltaic power plants

Photovoltaic (PV) plants experience aging over a time span that can lead to faults, consequently reducing the overall power production of PV plants. Thermo-diagnosis is one of the most effective methods for inspecting the photovoltaic plants for faults, such as hotspots. The inspection of individual PV panels for hotspots can be time consuming and expensive. Either the panels are located at a height on the rooftop of buildings or over a larger area, thus making the inspection process significantly complicated. This paper presents a solution to automate the process of PV plant maintenance by employing a Remotely Piloted Aircraft System (RPAS) equipped with a custom-designed payload that includes a thermal camera and a low-cost GNSS RTK receiver. The RPAS performs a flying mission over a PV plant to collect thermal images accurately geo-referenced by GNSS RTK receiver that allows to identify and locate the defective PV module without immediate human intervention. The paper presents end-to-end system design of the proposed solution and preliminary results of the proof-of-concept.