Machine Learning-Based Framework to Predict the Combined Effects of Climate Change and Floating Photovoltaic Systems Installation on Water Quality of Open-Water Lakes

Floating photovoltaic (FPV) systems represent a promising advancement in renewable energy technology; however, a comprehensive understanding of their environmental impacts is essential. The effects of FPV installation on lake water temperature remain unclear, potentially hindering the development of the technology due to associated negative implications for aquatic ecosystems. Furthermore, the rise in water temperature associated with climate change poses additional threats to open-water bodies. In this context, the current study endeavors to develop a machine learning (ML)-based framework to assess the combined impact of climate change and the installation of FPV systems on the water quality of open-water lakes. This framework involves the creation of three predictive models and a forecasting model utilizing various ML algorithms, concentrating on temperature and water quality predictions. The framework was applied to a case study assessing the impact of installing three distinct FPV systems on the water quality of Oostvoornse Lake in the Netherlands, employing water quality data available in the literature. The findings indicate a temporal increase in both air and water temperatures at the site, underscoring the ramifications of climate change. Additionally, the results suggest that FPV installations can influence lake thermal dynamics, leading to variations in water temperature and dissolved oxygen concentration, which presents both opportunities and challenges in addressing the impacts of climate change. The proposed framework will be an effective tool for evaluating the effects of FPV systems on water quality throughout their operational lifespan while addressing significant climate change issues.