
تعداد نشریات | 8 |
تعداد شمارهها | 414 |
تعداد مقالات | 5,480 |
تعداد مشاهده مقاله | 6,018,493 |
تعداد دریافت فایل اصل مقاله | 5,301,789 |
Integrating digital twin for optimizing autonomous aerial monitoring of photovoltaic systems | ||
Advances in Energy Sciences and Technologies | ||
دوره 1، شماره 1، شهریور 2025، صفحه 5-20 اصل مقاله (3.11 M) | ||
نوع مقاله: Original Article | ||
شناسه دیجیتال (DOI): 10.22060/aest.2025.5748 | ||
نویسندگان | ||
Mohammad Kolahi1؛ Sayyed Majid Esmailifar2؛ Mohammadreza Aghaei* 3؛ Amir Mohammad Moradi Sizkouhi4 | ||
1Department of Mechanical Engineering, University of Isfahan, Isfahan, Iran | ||
2Department of Aerospace Engineering, Amirkabir University of Technology, Tehran, Iran | ||
3Department of Ocean Operations and Civil Engineering, Norwegian University of Science and Technology (NTNU), Ålesund, Norway Department of Sustainable Systems Engineering (INATECH), Albert Ludwigs University of Freiburg, Freiburg, Germany | ||
4Department of Electrical and Computer Engineering, Concordia University, Montreal, Canada | ||
چکیده | ||
In this research, we integrate digital twin (DT) technology through a platform designed for simulating and managing the autonomous aerial monitoring procedure of photovoltaic (PV) power plants, known as Digital-PV. This innovative platform enables users to test various scenarios and configurations of PV power plants, allowing for an evaluation of their impact on the autonomous aerial monitoring process. By doing so, it reduces the risks associated with real-world experiments and helps pinpoint the most effective strategies for improving PV system monitoring. Digital-PV also provides a virtual environment for conducting tests of autonomous flights and missions, covering aspects such as boundary detection, path planning, and fault detection. It includes features for generating data that inform the development of data-driven monitoring and inspection models. The development process of Digital-PV involved creating a digital twin of a utility-scale PV plant within Unreal Engine, simulating aerial robot flight with AirSim, and expanding of application programming interface (APIs) to enable our platform to adapt to different scenarios for evaluating smart monitoring models and collecting datasets. Moreover, throughout the study, a dataset of synthetic aerial images was collected from Digital-PV, which was subsequently used to train an end-to-end segmentation model aimed at detecting bird droppings on PV panels. Ultimately, this platform evaluated a range of intelligent aerial monitoring models, providing valuable insights into their capabilities and potential effectiveness in real-world applications. | ||
کلیدواژهها | ||
Photovoltaics (PV)؛ Aerial robot؛ Artificial intelligence (AI)؛ Digital twin (DT)؛ Autonomous aerial monitoring (AAM) | ||
آمار تعداد مشاهده مقاله: 22 تعداد دریافت فایل اصل مقاله: 9 |