MS#06.12 Digital Twinning for Offshore Wind: Integrating Physics-Based, Data-Driven, and Hybrid Methods for Reliability and Performance Assessment

E. TRONCI¹, E. CHATZI², F. PIMENTA³, B. MOAVENI⁴
¹ Northeastern University|² ETH Zurich|³ Faculty of Engineering, University of Porto|⁴ Tufts University

Reliability, monitoring and sensing, O&M

Digital twinning has emerged as a transformative technology in offshore wind, offering unparalleled opportunities to enhance the reliability and performance of wind turbines through continuous monitoring and advanced analytics. This mini-symposium will explore the critical role of digital twins in bridging the gap between physics-based models, data-driven approaches, and hybrid methodologies to enable real-time performance assessment and structural health monitoring of key turbine components, with a focus on towers and substructures systems.
A key focus of this session will be leveraging digital twins to create dynamic, virtual replicas of offshore wind turbine structures that provide actionable insights into their operational health, structural integrity, and performance under varying environmental conditions. By integrating high-fidelity physics-based simulations with experimental data from sensor networks, digital twins offer a comprehensive approach to monitoring turbine behavior, predicting future maintenance needs, and understanding the degradation and fatigue of structural components. This approach not only improves reliability but also optimizes operational strategies, reduces downtime, and extends the service life of critical structural elements.
The session will showcase recent advancements, emphasizing the integration of experimental data and real-time data analytics with advanced computational models. Experts from academia and industry will present case studies and innovative solutions addressing structural reliability and performance, with a focus on towers, substructures, and other turbine components. Discussions will explore how physics-based models and data-driven methods can be combined to improve monitoring precision, enabling predictive maintenance and lifecycle optimization for offshore wind turbine structures. Contributions focusing on towers and substructures and using experimental data are strongly encouraged.
Topics accounted for in this session cover, but are not limited to:

• Physics-based models for structural reliability, model updating, and performance monitoring
• Data-driven approaches leveraging machine learning and data analytics for enhancing data characterization and predictive performance assessment
• Hybrid methods integrating physics-based simulations and machine learning for comprehensive monitoring strategies
• Life-cycle assessment and service life extension of towers and substructures using digital twin technology
• Probabilistic modeling for reliability assessment and decision-making under uncertainty
• Degradation models and variability in operational conditions affecting structural component performance