ABSTRACT

In recent years, the technology for the floating offshore wind turbine (FOWT) has developed rapidly, with numerous countries preparing to establish floating offshore wind farms. Studying the wake model of FOWT is crucial for reducing wake effects and optimizing wind farm layouts. Most existing wake models are primarily suited for fixed wind turbines, while the motion of the FOWT platform can greatly affect its wake characteristics. Therefore, it is necessary to propose new wake models specifically for floating wind turbines. This study employs a machine learning-based symbolic regression (SR) method to predict the wake of FOWT. First, the FOWT-UALM-SJTU solver, which utilizes large eddy simulation (LES) coupled with unsteady actuator line model (UALM), is used to simulate multiple surge conditions of FOWT and obtain wake velocity field data. Then, a wake model is developed using symbolic regression. Finally, the proposed wake model is validated. The results show that the model accurately predicts the velocity distribution in the far-wake region of FOWT under surge motion.

INTRODUCTION

In recent years, wind energy has become a focal point of attention due to its advantages of being non-polluting, renewable, and abundant in resources (Rohrig et al., 2019). The application of wind energy is divided into onshore and offshore wind farms. Compared to onshore wind farms, offshore wind farms benefit from higher wind speeds, longer annual operational hours, and no land use (Li et al., 2020). Consequently, offshore wind farms have gradually become a focus of development for many countries. Additionally, wind resources in deep-sea areas are significantly superior to those in nearshore environments. However, in deep-sea regions, fixed wind turbines do not meet commercial requirements, and only FOWTs can be considered. The technology for floating offshore wind turbines (FOWTs) has rapidly developed in recent years, making it possible to build floating offshore wind farms (FOWFs) that harness the abundant wind energy resources in deep-sea areas (Chitteth Ramachandran et al., 2022).

This content is only available via PDF.
You can access this article if you purchase or spend a download.