Offshore Wind Energy plays a vital role in advancing the United Nations Sustainable Development Goals (SDG 13) by driving clean carbon-free energy. The high cost of offshore wind energy driven by conservative designs, steep operation and maintenance expense, and complex logistics due to high rate of failures poses a significant challenge. Hence, for offshore wind to compete with non-renewable sources, it is necessary to reduce Levelized Cost of Energy. Monopiles supporting offshore wind turbines (OWTs) share a significant portion of this cost. Thus, enhancing their reliability, anticipating potential damage, and optimizing inspection and maintenance is crucial. Monitoring damage parameters such as cracks, vibrations, and strains through sensors help failure detection, and use of failure data to train advanced ML algorithms aid in predictive maintenance. With a goal to enhance the reliability of OWT monopile foundations, this paper critically reviews the reliability studies and examines the role of monitored data, addressing existing challenges with proposed solutions. Given the limited availability of sensor-collected data, the use of Artificial Intelligence (AI) and Physics-Based Machine Learning (PBML) is recommended. As structural response and material strength degrade over time, updating the prediction model with real-time operational data becomes essential. Digital Twin (DT) facilitates these dynamic updates, enhancing prediction accuracy. The proposed Reliability-Centric DT framework effectively integrates the sensor data and physical information, using advanced AI, thereby creating a virtual representation of the actual support structure. Consequently, informed decisions regarding operation and maintenance can be made, leading to potential cost savings.
Khan et al. (2026) studied this question.