Digital-PV: A digital twin-based platform for autonomous aerial monitoring of large-scale photovoltaic power plants

In this study, a novel digital twin-based solution called Digital-PV has been developed for the simulation and managed execution of autonomous aerial monitoring of photovoltaic (PV) power plants. Digital-PV empowers users to simulate different scenarios and PV power plant configurations and assess their impact on PV systems' autonomous aerial monitoring process. This procedure reduces the risk associated with real-world experimentation and helps identify the most effective strategies to improve PV system monitoring. It also provides a virtual testing platform for autonomous flights and missions, including boundary detection, path planning, and fault detection along with data generation capabilities for developing data-driven monitoring and inspection models. The solution involved creating a digital twin of an R&D utility-scale PV plant environment in Unreal Engine, aerial robot flight simulation using AirSim, and developing application programming interfaces (APIs) for running desired scenarios for collecting data, testing different monitoring models like plant boundary extraction, path planning, fault detection, etc. In addition, during this study, a dataset of synthetic aerial images was collected from Digital-PV and used to train an end-to-end segmentation model for detecting bird droppings on PV panels. Finally, we utilized this platform to evaluate various intelligent monitoring models, gaining valuable insights into their capabilities and potential performance in real-world scenarios.