Machine learning–based long-term degradation and LCOE Analysis of floating PV with custom pontoon design

Roby Mohajon, Md Rabiul Islam Polash, Md Nabil Shahriar, Anupom Bhowmick, Pranab Kumar Mondol, Hrittik Mutsuddi, Jihanul Haque Jihan and Nur Mohammad

PLOS ONE, 2026, vol. 21, issue 7, 1-31

Abstract: Floating Photovoltaic (FPV) systems are a feasible alternative for solar energy utilization in areas where land is scarce. However, issues related to performance degradation, ecological effects, and economic viability have restricted the widespread acceptance of this technology. This study examines the eco-compatibility of a 5 MW FPV solar plant, which is suitable for a wetland ecosystem in Bangladesh, and examines the technical feasibility, environmental sustainability, and economic viability of the plant. In this study, a model was developed using detailed system-level simulations, and the model validated the structural feasibility of a specially designed light-permeable annular pontoon by performing hydrostatic buoyancy and stability tests. Ecological compatibility is assessed using a light-transmission-based photosynthetic viability model, whereas the long-term degradation of the performance ratio and energy yield over a 25-year lifetime is predicted using a climate-aware machine learning framework that incorporates irradiance, temperature, humidity, and system aging effects. Furthermore, an economic model sensitive to inflation was used to examine the levelized cost of electricity (LCOE) in the case study. The results show an average performance ratio of 82.4%, lifecycle energy outputs of approximately 222 GWh, and an LCOE of 0.0315 USD/kWh. In addition, approximately 70% of the aquatic photosynthesis potential is retained, and approximately 104,149.5 tCO₂ is eliminated.

Date: 2026
References: Add references at CitEc
Citations:

Downloads: (external link)
https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0342926 (text/html)
https://journals.plos.org/plosone/article/file?id= ... 42926&type=printable (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:plo:pone00:0342926

DOI: 10.1371/journal.pone.0342926

Access Statistics for this article

More articles in PLOS ONE from Public Library of Science
Bibliographic data for series maintained by plosone ().