login

Search documents

Browse topics

Document details

 
Title:
 
Automated Module Failure Identification and Proposal of Repowering in Operating Solar Plants for Continuous Optimum Operation
 
Author(s):
 
H.-J. Rodríguez San Segundo, J.S. Sánchez Serrano, R. Imbert, C. Gonzalo-Martín, V. Robles, A. Calo López, C. de Vicente Suso
 
Keywords:
 
Artificial Intelligence, PV Plant, PV Module Failure, Aerial Inspection
 
Topic:
 
PV Systems - Performance, Applications and Integration
Subtopic: Operation, Performance and Maintenance of PV Systems
Event: 35th European Photovoltaic Solar Energy Conference and Exhibition
Session: 6BO.7.4
 
Pages:
 
1569 - 1572
ISBN: 3-936338-50-7
Paper DOI: 10.4229/35thEUPVSEC20182018-6BO.7.4
 
Price:
 
 
0,00 EUR
 
Document(s): paper
 

Abstract/Summary:


Solar PV power plants are composed of thousands of solar modules. It is a known fact that 2% of them will fail after year 10 of operation, causing losses as high as 27% of total income. In a market of plants over 10 years old of 10 Gigawatts peak (GWp) in 2018, and 100 GWp by 2025, this leads to global losses of EUR 750 million and EUR 7.5 billion, respectively. Despite these figures, this problem is currently not being tackled, because, to recover the losses, a continuous repowering of the PV plant must be performed, i.e., a continuous detection and substitution of the failed modules for new ones. This implies at least an annual inspection, and an ulterior analysis of the inspection results in order to make the best decision on the repowering configuration. While the inspection will soon start reducing costs by the increasing penetration of drones, the ulterior analysis is still carried out manually by qualified technicians (supervising, one by one, the thousands of images taken by the drone). This fact escalates costs and makes the a continuous repowering unaffordable, i.e., not worth the recovery of the losses. The only way to make continuous repowering viable is to automate the image processing, analysis and decision-making. The authors, through our experience in the field, have calculated that automation will save up to 90% of current costs, making repowering profitable. Therefore, in the frame of a Spanish R&D project, we are currently developing Optimized Solar Repowering (OSR), a software that carries out an automated image analysis of the images taken by the drone, in which failures are identified, geo-localized and classified, and, also, automatically proposes the optimum repowering configuration, by means of own developed algorithms embedded into a user-friendly software architecture. First results and advances of the project are presented.