Search the EPA Research Database

Project title

A Machine Learning Approach for Minimizing the Levelized Cost of Energy in Irish Floating Offshore Wind Farms

Primary Funding Agency

Taighde �ireann-Research Ireland

Co-Funding Organisation(s)

Lead Organisation

University of Galway (UG)

Lead Applicant

Not listed

Project Abstract

Ireland is committed to reducing its annual Greenhouse Gas (GHG) emissions by an average of 7% annually from 2021 to 2030, achieving the ambitious goal of net-zero emissions by 2050. Electricity generation is a significant contributor to GHG emissions and is responsible for substantial growth in the coming decade, with an estimated increase of 50% in demand. Consequently, decarbonizing electricity generation is a critical component of emissions reduction strategies. Ireland possesses substantial potential for offshore floating wind energy as a transformative opportunity for green electricity production. Nevertheless, developing offshore floating wind turbines comes with significant installation, maintenance, and operational costs, necessitating the development of highly efficient wind farms with minimized Levelized Cost of Energy (LCOE). Central to this efficiency is the precise positioning of wind turbines within a wind farm, known as wind farm layout optimization. An optimal layout is essential for mitigating the adverse effects of turbulent wakes on power generation and the lifespan of wind turbines within a wind farm. However, designing an optimal offshore floating wind farm layout is a complex, multidisciplinary endeavour. Current optimization methods rely on simplified models unable to capture the intricate dynamics of floating wind turbines under the influence of wind, waves, and currents. The dynamic nature of floating platforms further complicates the optimization process, making it computationally intensive. This project seeks to develop an advanced optimization framework for offshore floating wind farms, surpassing current methodologies. The proposed project will leverage data-driven surrogate models, incorporating physics-informed machine-learning techniques. These models will establish relationships between wind farm layout parameters and LCOE, facilitating the identification of optimal layout configurations through genetic algorithm utilization. The proposed research aims to advance Ireland's renewable energy sector, ensuring efficient offshore floating wind turbines, contributing significantly to the nation's climate and energy goals.

Grant Approved

�124,000.00

Research Hub

Climate Change

Research Theme

1. Carbon Stocks, GHG Emissions, Sinks and Management Options

Start Date

06/01/2025

Initial Projected Completion Date

31/05/2029