Intelligent-inspired framework for fatigue reliability evaluation of offshore wind turbine support structures under hybrid uncertainty

• Type: Article in International Scientific Journal
• Date: 2024

Authors

Meng, DB (Author) Other; Yang, SY (Author) Other; Yang, HF (Author) Other; Abilio M P De Jesus (Author) FEUP; Correia, J (Author) FEUP; Zhu, SP (Author) Other

Journal

Title: Ocean Engineering

Vol. 307

ISSN: 0029-8018

Publisher: Elsevier

Indexing

ISI Web of Knowledge - 30 Citations

Scopus - 0 Citations

Other information

Authenticus ID: P-010-E2X

DOI: 10.1016/j.oceaneng.2024.118213

Abstract (EN): Offshore wind energy has become a key strategy in the global transition to sustainable and renewable energy. This emphasis has significantly propelled the advancement of offshore wind turbine technology. However, fatigue of offshore wind turbine support structures remains a significant challenge due to long-term exposure to harsh environmental conditions. In this study, an intelligent optimization-inspired Support Vector Regression (SVR) modeling strategy is proposed to construct a high-precision approximation model. Compared with original SVR modeling method, the modeling strategy proposed in this study has higher accuracy in building models of complex engineering structures. Additionally, a fatigue reliability evaluation framework considering hybrid uncertainty is presented. The new reliability evaluation framework can be combined with a variety of heuristic algorithms to unleash the potential of heuristic algorithms. Finally, this study applies the proposed framework to an example of fatigue reliability evaluation of offshore wind turbine support structures considering the effects of multiple uncertainties. The performance of different heuristic algorithms under the proposed framework is compared in this study. Furthermore, this study mainly finds that the MAE index of the established SVR model using the intelligent optimization algorithm is 31.2% higher than that of the SVR model that does not use the intelligent optimization algorithm. Compared with the fatigue reliability evaluation that only considers random uncertainty, the failure probability of the fatigue reliability evaluation using hybrid uncertainty is 0.64% higher. The fatigue reliability evaluation framework considering hybrid uncertainties proposed in this study is more accurate and conservative.

Language: English

Type (Professor's evaluation): Scientific

No. of pages: 15