Potential use of Bayesian Networks for estimating relationship among rotational dynamics of floating offshore wind turbine tower in extreme environmental conditions

Rostam-Alilou A. A., Zhang C., Salboukh F., Gunes O.

OCEAN ENGINEERING, vol.244, 2022 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 244
  • Publication Date: 2022
  • Doi Number: 10.1016/j.oceaneng.2021.110230
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Academic Search Premier, PASCAL, Aerospace Database, Applied Science & Technology Source, Aquatic Science & Fisheries Abstracts (ASFA), Communication Abstracts, Compendex, Computer & Applied Sciences, Environment Index, Geobase, ICONDA Bibliographic, INSPEC, Metadex, Civil Engineering Abstracts
  • Keywords: Floating offshore wind turbine, Bayesian network, Rotational dynamic, Dynamic response interactions, Extreme loading, Stability, SYSTEM, FRAMEWORK
  • Istanbul Technical University Affiliated: Yes


The rotational stability of the floating offshore wind turbine tower plays a vital role in the structural stability of the whole system. So, the protection of the system from probabilistic structural damages and failures due to instability needs planning for robust operation and maintenance methods with different application principles such as structural health monitoring, early failure detection methods, and other novel structural control systems. Besides all the other operation and maintenance methods, early detection and diagnosis for probabilistic failures are effective approaches to save costs and maintain operational safety. This study outlines the potential use of Bayesian Networks (BNs) for the estimation of technical relationships among structural dynamic responses of the tower of a floating spar-type offshore wind turbine. In the method, rotational dynamic responses (rolling and pitching) of a floating turbine are recorded in six different loading scenarios considering two stages of normal operation and parked conditions due to extreme environmental excitations. Bayesian network models are created from these data to observe the probabilistic interactions with the effects of all involving parameters in the rotational vibration of the tower. Thus, obtained final BN models for both normal operational status and extreme environmental conditions can be introduced as feasible decision-support tools to be potentially used in the prediction of relationships and interactions among the dynamic responses of floating turbine's tower.