Vibration suppression of multi-component floating structures via passive TMDs and Bayesian ascent

• Published in: Ocean engineering Vol. 259; p. 112088
• Main Authors: Zhang, Xiantao, Lu, Da, Dong, Hongyang, Zhao, Xiaowei, Brennan, Feargal, Liang, Yibo
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.09.2022
• Subjects: Bayesian ascent; Hydrodynamics; Multi-component floating structure; Tuned mass damper; Vibration suppression; Multi-component floating structure; Hydrodynamics; Vibration suppression; Tuned mass damper; Bayesian ascent
• ISSN: 0029-8018; 1873-5258
• DOI: 10.1016/j.oceaneng.2022.112088

This paper aims to achieve vibration suppression for multi-component floating structures under ocean waves. To this end, a multiple-passive-TMD (tuned mass damper) structure is employed, and a numerical model is developed incorporating both the dynamics of a hinged floating foundation and passive TMDs. A data-driven parameter optimization method is developed to search for optimal TMD parameters. This method is built upon Bayesian Ascent (BA) – an advanced sequential searching strategy for optimizing black-box functions, combining the merits of both the Bayesian Optimization method and the gradient-free trust-region algorithm. Simulation results under different wave conditions verify the effectiveness of the BA-based parameter selection method, showing that the resulting passive TMDs can significantly reduce the vibration of the whole floating structure. •A novel data-driven optimization method is designed for vibration suppression.•A multi-component floating structure with TMD modules is modeled and analyzed.•The parameters of TMDs are optimized by a Bayesian-based data-driven method.•The proposed method can reduce vibrations subject to different wave conditions.