Comparative life cycle impact assessment of offshore support vessels powered by alternative fuels for sustainable offshore wind operations using machine learning

Abstract

Offshore Anchor handling tug supply (AHTS) vessels, a type of offshore support vessel, are critical for the operations related to handling anchors of offshore structures, oil rigs, and wind turbines, towing them to remote deep-sea locations, and securing them in place. Amidst growing concerns regarding the environmental footprints of carbon-based fuels and impending carbon taxation, the International Maritime Organization, policymakers, classification societies, shipping firms, and stakeholders are seeking cleaner alternatives. LNG (Liquefied natural gas) and Green ammonia as energy vectors are considered among the top contenders for future clean alternatives for offshore vessels. This study evaluated the environmental performance of newly built AHTS vessels powered by LNG and Green ammonia as marine fuels designed for offshore operations. This environmental impact assessment study uses IPCC and Environmental footprint methodologies. Considered broad impact groups: Global warming, human toxicity, eutrophication, ecotoxicity, and atmosphere-related impacts, and analyzed the process impacts. This study uses Supervised machine learning algorithms such as the Random forest, Decision tree, and XGBOOST models for environmental performance evaluation and prediction. The study reveals that the recently manufactured AHTS vessel, utilizing conventional fuels like Heavy fuel oil, Marine diesel oil, and LNG, exhibits significantly increased GTP 100 and GWP 100 emission levels per tonne-kilometer when compared to green ammonia, with a 44 % and 10.6 % rise compared to Heavy fuel oil, respectively. Furthermore, the XGBOOST regression model outperformed the Random forest and Decision tree models in predictive accuracy for GWP 100. It is analyzed and proposed that effectively managing unsustainable processes would minimize environmental footprints and reduce carbon, nitrogen oxide, LNG, and ammonia-based emissions.