1. Introduction
At the 80th Marine Environment Protection Committee (MEPC 80), governments adopted the IMO’s revised greenhouse gas (GHG) strategy [
1]. And the objective is to reach net-zero GHG emissions from international shipping close to 2050. Offshore wind energy, especially offshore wind turbines, has attracted considerable attention in power and renewables [
2]. According to DNV’s Energy Transition Outlook 2023, the installed global floating offshore wind capacity will have grown from today’s 200 MW to almost 270 GW by 2050. The floating offshore wind turbine (FOWT) has a large structure and size, and the structural motion response is complex. In order to study the dynamic responses of the wind turbine, researchers have conducted a series of studies using coupling analysis software, aiming to accurately assess the characteristics of the structure.
In as early as 2005, the National Renewable Energy Laboratory (NREL) developed the popular 5 MW Reference Wind Turbine (RWT) [
3], followed by the Technical University of Denmark’s (DTU) 10 MW RWT [
4]. The NREL, using numerical simulation and model experiment methods, developed the OC3-Monopile and OC4-DeepC platforms to support the 5 MW RWT [
5,
6]. Meanwhile, a braceless steel platform was designed by the Norwegian Research Centre for Offshore Wind Technology to support the 5 MW Reference Wind Turbine (RWT) [
7]. It consists of a central column, three side columns, and three pontoons; it was developed using numerical analysis methods. Additionally, the NAUTILUS-10 floating substructure [
8] was provided by the LIFE 50+ project to support the DTU 10 MW RWT. In Japan, the FORWARD Semi-Sub 2 MW FOWT and V-shape Semi-Sub 7 MW FOWT have been developed and installed on a four-column floating platform [
9,
10]. An 8.4 MW wind turbine has been designed for installation on the WindFloat platform as part of the WindFloat Atlantic (WFA) project [
11]. A concrete multi-column floating platform has been conceptually designed by Yokohama National University [
12]. Moreover, several barge-type platforms have been designed and experimented with to support wind turbines [
13,
14]. With the development of deep water turbines and large turbines, elevated requirements are put forward for the stabilization of the floating wind turbines. When the wind turbines are operated under extreme conditions, they are a highly nonlinear system under floating body motions and extreme wave–structure interactions. The motion characteristics of the whole offshore wind structures are crucial for the generation of power, as well as their safety.
Yang et al. [
15] developed a coupling framework based on FAST and AQWA through modifications on source codes of the user_force.dll of AQWA, obtaining the motion of multi-body platforms under the action of wind and waves. The framework has the ability to simulate multi-body platforms and to observe unique results, which contrasts with traditional tools. Guo et al. [
16] developed a multibody modeling framework, namely TorqTwin. The framework can predict the motion response of the floating platform by calling on the source code of OpenFAST through Python library. However, the portability of the code is limited. Domene et al. [
17] incorporated a direct air capture (DAC) system into the IEA 15 MW Reference Wind Turbine, analyzed the motion response, and compared it to the initial design. Chen et al. [
18] used a coupling analysis software based on OrcaFlex to conduct a time domain simulation for 15 MW RWTs. The analysis considered three frequency regions (low frequency, wave frequency, and high frequency) and revealed that the tension response of the mooring line would be significantly affected by high-frequency motions originating from the superstructure. Using the computational fluid dynamics (CFD) method, Xue et al. [
19] utilized the FOWT-UALM-SJTU solver to calculate the motion response of OC4 FOWT under different wave conditions and found that wind turbines with larger initial inclination angle recovered faster. Tran et al. [
20] used CFD to analyze the motion responses of the OC4 semisubmersible FOWT based on an overset grid technique, and found that the motion of the floating platform has good consistency compared with the calculation results of FAST. Xu et al. [
21] conducted a secondary development based on OpenFAST, analyzed the dynamic response of an FOWT in different stages of typhoon transit, and found that yaw errors would lead to an increased mooring load. Guo et al. [
22] conducted an experimental study based on a 12 MW semisubmersible floating wind turbine, obtaining its motion and mooring response, and found that the motion response and mooring tension in the 60° and 90° directions were greater than those in the 0° direction. Xiang et al. [
23] discussed the influence of different mooring lengths and counterweights on the motion response of semisubmersible wind turbines, and found that the motion of the floating platform in the pitch direction was affected. Zhang et al. [
24] used different mooring models to compare the motion performances and mooring tension of FOWTs, and found that the quasi-static method still has a better calculation accuracy in most cases, but the finite element method is more suitable for fatigue calculations. Lozon et al. [
25] conducted a dynamic analysis of shared mooring for floating wind farms and found that shared mooring does not cause an increase in dynamic response. Gao et al. [
26] discussed the differences in the dynamic responses of FWOTs when the platform is considered as either rigid or flexible. The results show that accounting for the flexibility of the tower and platform leads to a significant increase in both the motions in the pitch direction and the tower-base force. Huang [
27] constructed a shell-based Finite Element Model (FEM) to simulate motion responses under seismic loads, and demonstrated the flexible foundation of FOWTs presents a lower stress. Rinker et al. [
28] compared the aeroelastic responses from HAWC2 and OpenFAST for the IEA 15 MW RWT and found that the aerodynamic loads agree quite well. Most recently, the IEA 22 MW ultra-large-scale RWT and a floating platform have been designed in IEA Wind TCP Task 55 [
29]. While some significant and useful conclusions have been obtained, these studies are inadequate for understanding the motion responses of ultra-large-scale floating wind offshore turbines (FWOTs) under extreme conditions. Therefore, it is necessary to conduct a more in-depth analysis of wind turbine motion responses in extreme conditions.
In the present work, a classical semisubmersible floating offshore wind turbine, which consists of four columns, a heave plate, and three catenary chain mooring lines, developed by the International Energy Agency is studied. The hydrodynamic and motion characteristics under extreme conditions are crucial. The main focus is to utilize a software tool to predict the motion characteristics of floating offshore wind turbines and the hydrodynamic forces excited by highly nonlinear environmental loads.
The remainder of this paper is organized as follows:
Section 2 describes the methodology, while
Section 3 presents the model data of the FOWTs. In
Section 4, the wind and wave loads utilized in the simulation are discussed, and a comparison with the results of the free decay is made to demonstrate the accuracy of the model. The results are presented in
Section 5. Finally, the conclusions are drawn in
Section 6.
5. Results and Discussion
In this section, the motion characteristics, nacelle acceleration responses, and mooring forces of the IEA 15 MW Reference Wind Turbine are studied under different wind and wave combination conditions.
5.1. Motion Responses
The motion responses under different conditions are calculated and the simulation time is set to 3500 s. Then, the influence of different conditions on the motion response is obtained.
Figure 5 shows the time history curve of the pitch, heave, and surge motions of the floating platform under different conditions. It can be seen from
Figure 5 that the surge and sway motion are the main motion forms.
Transient effects can be observed in approximately the first 200 s and are not significant in the heave direction. The motion in all three directions starts from an initial position of 0 m. The most substantial change was observed in the surge direction, where the motion increased from 0 m to approximately 25 m under Ultimate 1 and Extreme conditions, and from 0 m to 15 m under the Operational 1 condition, before beginning to fluctuate. Similarly, in the pitch direction, the motion increased from 0 m to approximately 6 m under the Ultimate 1 and Extreme conditions, and from 0 m to 2 m under the Operational 1 condition, before beginning to fluctuate.
Table 6 describes the statistical parameters of motion response in each degree of freedom direction under different conditions. Here, ‘Max’, ‘Min’, and ‘Mean’ denote the maximum, minimum, and average values of motion response, respectively. ‘STD’ represents the standard deviation, and ‘Range’ indicates the difference between the maximum and minimum values (maximum–minimum). The values presented here are derived from data between 500 s and 3500 s, with transient results excluded from the statistical analysis.
Based on the calculation results, it is evident that when only the wave height and wave period are increased, the wind turbine can still operate normally, without a corresponding increase in wind speed (referred to as ‘Ultimate 1’). Furthermore, the responses in the surge and pitch directions exhibit minimal variation compared to the rated working condition (Operational 1). Specifically, the average motion response amplitude increased by 7% in the surge direction and decreases by 4% in the pitch direction.
Under the extreme environmental condition, occurring once in 50 years (referred to as ‘Extreme’), the average motion response amplitude is reduced by 33% in the surge direction, and is reduced by 106% in the pitch direction, compared to the rated working condition. This reduction is likely due to the pitched blades and braked generator.
However, the influence of waves on wind turbine response is most notable in the heave direction, where the ‘Range’ increased by a factor of 12 in ‘Ultimate 1’ and by a factor of 16 in ‘Extreme’ scenarios.
5.2. Nacelle Acceleration Responses
The acceleration of the nacelle has an important influence on the force exerted on the structure, and the nacelle acceleration responses serve as a reference for studying the tower responses. Therefore, the nacelle acceleration responses are obtained.
Figure 6 presents the acceleration responses in the
x and
y directions, while
Figure 7 displays the maximum (Max) values and standard deviation (STD) values of the response. The effects of transient phenomena are hardly noticeable in the acceleration response.
It can be seen from
Figure 6 that the nacelle acceleration responses are all between −2 m/s and 2 m/s in the
x direction, and are between −1 m/s and 1 m/s in the
y direction under different conditions. In
Figure 7, the maximum nacelle acceleration response is measured at 1.65 m/s
2. According to relevant regulations, when the nacelle acceleration response exceeds 6 m/s
2, the risk of structural failure is expected to rise. Therefore, the floating offshore wind turbine structure is within a safe range.
It can also be found that the acceleration response of the structure is notably affected by the wave height and wave period. And the acceleration response of structure in the windward direction (X direction) is significantly greater than that in the lateral direction (Y direction). There is little difference in the acceleration response between the Ultimate 1 and Extreme conditions in the X direction. However, the acceleration response is larger under both the Ultimate 1 and Extreme conditions compared to the rated working condition. Under the Extreme condition, the maximum acceleration response in the Y direction is approximately three times greater than that under rated conditions, with the standard deviation being approximately 3.6 times larger.
5.3. Mooring Force Responses
There is no doubt that the state of the mooring cable is particularly important for the safety of floating wind turbine structures. The force responses of three mooring lines can be calculated under different conditions.
Figure 8 shows the time history curve of fairlead tensions, and
Figure 9 shows the statistic characteristics of fairlead tensions.
Transient effects, observable within the initial 100 s, exhibit different trends in mooring cables distributed in two directions around the FOWT. Due to the presence of initial tension, the tension in the mooring lines initially changes from 2.5 MN.
The tension of fairlead 1 increased from 2.5 MN to approximately 5 MN under the Ultimate 1 and Extreme conditions, and from 2.5 MN m to 4 MN under the Operational 1 condition, before beginning to fluctuate. Similarly, the tension of fairleads 2 and 3 decreased from 2.5 MN to approximately 1.8 MN under the Ultimate 1 and Extreme conditions, and from 2.5 MN to 2 MN under the Operational 1 condition, before beginning to fluctuate.
From the results, it can be observed that the tension of the mooring cable increases with deteriorating environmental conditions. When the wind turbine is operational, the average tension of the cable on the windward side (fairlead 1) is greater than that on the leeward side. Under DLC3, the average tension of the cable on the windward side decreased, while the tension of the two mooring cables on the leeward side increased. In all conditions, the mooring cable tension is far below its breaking strength (22 MN [
34]), indicating a large safety margin.