EU PVSEC Programme Online
EU PVSEC 2021, 6 - 10 September 2021
Presentation: 5CV.2.17 Artificial Intelligence Based Solutions for PV Plant Condition Monitoring and Diagnosis Using Unmanned Aerial Vehicle Images and on-Site Measurements
Type: Visual
Date: Wednesday, 8th September 2021
13:30 - 15:00
Author(s): A. Betti, M. Tucci, M. Piazzi, L. Gigoni, C. Lanzetta
Presenter / Speaker: M. Piazzi, I-EM, Livorno, Italy
Event: Conference Conference
Session: 5CV.2 Operation, Performance and Maintenance of PV Systems
Topic: 5. 3 Operation, Performance and Maintenance of PV Systems
Summary / Abstract: As solar PV installations continue to grow worldwide [1], efficient and low-cost Operation and Maintenance (O&M) tasks are required to keep high performances over time and increase PV system lifetime. Indeed, PV arrays are often subject to hard working conditions, and therefore affected by many types of faults, which may be mainly categorized in intermittent (e.g., panel soiling), permanent (e.g., short-circuit) and incipient due to panel aging [2]. Faults must be promptly and accurately identified to avoid severe system degradation. Therefore, Fault Detection and Diagnosis (FDD) methods are nowadays widely investigated and adopted in the industry to early detect and possibly prevent catastrophic PV system failures. FDD methods are mainly based on the use of either on-site electrical and environmental measurements or UAV thermal and visual imagery [3] [4] [5]. In particular, due to the challenging Big Data collected by on-site sensors and Unmanned Aerial Vehicles (UAVs), visual inspection is nowadays inefficient and costly, whereas Machine Learning (ML) solutions are recently emerging as promising and convenient alternatives for automatic maintenance strategies [6] [7].