Typhoon Eye-Induced Misalignment Effects on the Serviceability of Floating Offshore Wind Turbines: Insights Typhoon SOULIK

Abstract

The northern Taiwan Strait, characterized by deep waters and high wind energy density, presents significant potential for developing floating offshore wind turbines (FOWTs). However, the region is prone to typhoons, with substantial variations in wind speed and direction during typhoon eye passages, posing challenges to FOWT safety and performance. This study investigates the serviceability of a 10 MW FOWT installed offshore of Hsinchu under typical wind and wave conditions during the eye of Typhoon SOULIK. Wind and wave data were sourced from the ERA5 reanalysis database. Simulations were conducted using OrcaFlex 11.4c, which enables fully coupled dynamic analysis of the entire FOWT system, including the mooring system, platform, tower, turbine, and nacelle, facilitating accurate predictions of system behavior in complex offshore environments. This study evaluated scenarios of maximum wind speed, significant wave height, wind–wave misalignment, and minimum wind speed during typhoon eye passage, considering both idle and power production modes in accordance with IEC TS 61400-3-2 requirements. The results indicate that platform yaw motion exceeds IEC limits during typhoon events, particularly in power production mode. This highlights the need for reducing platform motion. It is recommended to further develop control strategies or implement an active control system for the platform to ensure operational reliability. This research provides critical insights into FOWT design and operational challenges in typhoon-prone regions.

1. Introduction

To achieve net-zero carbon emission goals, there is a global push towards developing various forms of renewable energy. According to the reports from the IEA, GWEC, and EWEA [1,2,3,4], the global capacity of offshore wind is expected to reach 10.9 GW by 2030. In addition to being connected to the grid, the generated electricity can also serve as the energy source for alternative fuels such as hydrogen, ammonia, and methanol. A total of 80% of the potential offshore wind resources are located in waters deeper than 60 m. Consequently, current offshore wind development is shifting towards deeper waters, moving from fixed-foundation to floating wind turbines. This trend indicates that floating offshore wind farms represent a future development direction. Numerous studies have been conducted on floating offshore wind. For example, the IEA promotes a series of IEA Wind Tasks to discuss wind energy technologies. Among the Tasks, Tasks 23 and 30 [5,6,7,8] compare the integrated loads and motions of spar-type and semi-submersible floating wind turbines using various simulation codes such as FAST, Bladed, HAWC, OrcaFlex, etc. The results show that simulations from these codes exhibit good consistency.

Currently, offshore wind farm development in shallower waters in Taiwan is becoming saturated, and expansion is moving into areas with depths ranging from 50 to 200 m. Floating offshore wind farms are becoming a feasible option for such developments. In January 2023, Taiwanese authorities outlined plans for a demonstration project of a floating offshore wind farm, indicating that there will be floating offshore wind farms in Taiwan. According to research [9] by the Industrial Technology Research Institute as shown in Figure 1, the highest wind energy density is offshore of Hsinchu in the Taiwan Strait, where water depths range from 50 to 100 m as shown in Figure 1 [9,10], making it suitable for developing floating offshore wind farms. Authorities also consider this region a potential site for floating offshore wind farms [11]. Taiwan is prone to typhoons, with an average of three to four typhoons striking per year [12], and up to seven in extreme cases such as in 2001. Therefore, wind turbines established in Taiwan must be designed to withstand typhoon conditions. Due to floating offshore wind turbines (FOWTs) withstanding environmental loads from wind and marine forces via platforms and mooring systems, the load conditions are more complex compared to fixed-offshore wind turbines. Additionally, the motion of the floating wind turbine must be considered during design and O&M. Consequently, experiences on typhoons from fixed-offshore or onshore wind farms in Taiwan may not directly apply to floating offshore wind farms. Further research is necessary to address the issues mentioned above. In the Asia–Pacific region, most floating offshore wind farm projects need to address typhoon issues during design and O&M, including the effect of wind and marine conditions as well as motion and loads on FOWTs. Xu et al. [13] analyze wave conditions during typhoons by using buoy data in the South China Sea and found that wind directions and wave directions may not always align. Barj et al. [14] studied the impact of wind–wave misalignment on system load under normal operation of spar-type floating wind turbines and found that spar-type floating wind turbines will experience the highest ultimate load and fatigue load when wind and waves are aligned, and it is recommended that the wave directions aligned with the wind and 90° out of alignment were considered to accurately capture the extreme and fatigue characteristics of the spar-type platform. Li et al. [15] evaluate the motion and integrated load of a 5MW FOWT under typhoon conditions. They used the Holland model and the SWAN model to evaluate wind and ocean conditions during Typhoon Mangkhut and used the results to evaluate the impact of floating wind turbines outside the eye of the typhoon. They concluded that platform motion significantly impacts the structural loads of floating wind turbines. Tao et al. [16,17] studied the impact of yaw misalignment on the motion response of floating wind turbines and the fatigue of floating wind turbines for different yaw control optimization purposes. They found that the yaw control of wind turbines significantly affects the motion response of the FOWT platform, while optimizing the control system with maximum power production as the goal will reduce the life of wind turbine bolts by about 3 to 9 years.

Considering that floating wind farms must be designed based on site conditions and that there are few studies on the serviceability of floating wind turbines installed in the Taiwan Strait under the environmental conditions during a typhoon eye, therefore, this study discusses the influence of wind and marine conditions during a typhoon eye on the serviceability of FOWTs in the Hsinchu offshore area, which is a potential site for developing floating offshore wind farms in the Taiwan Strait. The analysis is based on the integrated load and motion simulation results of FOWTs. Based on the requirement of IEC TS 61400-3-2 [18] and practical design considerations, the RNA acceleration, wind turbine inclination, and floating platform motion of FOWTs are discussed in this study.

This study examines FOWTs designed for the environmental conditions of the target sea area [19]. The wind turbine is mounted on a semi-submersible platform with an equilateral triangular shape when viewed from above. Each of the three vertices is equipped with two catenary mooring lines. The wind turbine installation position uses the innovative and mature off-center tower design first employed in the WindFloat project [20]. This design facilitates onshore installation operations, allows personnel access during operation and maintenance, and reduces construction costs. This platform is designed based on the DTU 10MW wind turbine [21], a reference wind turbine frequently used in offshore wind energy research. For more details, please refer to Section 3.1 of this study.

To achieve the goal of discussing the serviceability of the FOWT in the Hsinchu offshore area during a typhoon eye, information about the typhoons that have struck the area was collected first. The information includes the name of the typhoon, warning period, invading path, radius of the storm, and the wind and wave data. Subsequently, the distribution and characteristics of the wind speed, wind direction, significant wave height, wave direction, and wave period during the typhoon were analyzed. Based on this analysis and in accordance with IEC 61400 series standards, the load cases analyzed in this study are specified. Then, motion simulations of the FOWT are conducted using tools that meet the requirements of the standards. Finally, the simulation results are analyzed to explore the serviceability of the FOWT during typhoons. The related process is shown in Figure 2.

2. Environmental Conditions During Typhoon

2.1. Selection of Typhoons for Analysis

This study analyzes the wind and wave conditions of the typhoons that struck Taiwan and whose centers passed through the target area (offshore of Hsinchu) from 2013 to 2023 to identify the wind and wave characteristics that the FOWT constructed in the target area will experience during a typhoon. A typhoon whose center is most likely to pass through the target sea area is defined by the CWA as a category 2 or 7 typhoon in terms of the typhoon path impacting Taiwan, as shown in Figure 3 [22]. Among the 50 typhoons that struck Taiwan during this period, the ones with invading path categories 2 or 7 were SOULIK, DUJUAN, NESAT, HAITANG, and DOKSURI. The detailed information and invading paths of these typhoons are shown in

Table 1. Information on typhoons whose invading paths passed through the target area from 2013 to 2023 [23].

Typhoon	Warning Period (UTC + 8:00)	Path Category	Intensity 1
SOULIK	08:30 on 11 July 2013~23:30 on 13 July 2013	2	Intense
DUJUAN	08:30 on 27 September 2015~17:30 on 29 September 2015	2	Intense
NESAT	08:30 on 28 July 2017~10:15 on 30 July 2017	2	Moderate
HAITANG	11:30 on 30 July 2017~08:30 on 31 July 2017	7	Severe Tropical Storm
DOKSURI	08:30 on 24 July 2023~17:30 on 28 July 2023	7	Moderate

¹ The intensity of a typhoon is defined by the CWA according to the maximum average wind speed near the typhoon center as a “Severe Tropical Storm” when it is 7.2 to 32.6 m/s, “Moderate Typhoon” when it is 32.7 to 50.9 m/s, and “Intense Typhoon” when it is equal or greater than 51.0 m/s.

and Figure 4 [23]. DUJUAN and DOKSURI were typhoons whose centers did not pass through the target area; NESAT and HAITANG struck Taiwan during the same period, resulting in a more complex fluid field. SOULIK was the only typhoon that struck Taiwan during the period of 2013 to 2023, with its center passing through the target area. The wind speed distribution and the changes in wind direction during the passage of Typhoon SOULIK’s eye over the target area align with the characteristics observed in other studies, such as those by Ishihara et al., who observed the conditions during Typhoon Maemi’s eye passage over Miyakojima Island [24], or the investigation report from Taiwan Power Company who observed the wind conditions during Typhoon Dujuan’s eye passage over Taichung [25]. Therefore, this study takes Typhoon SOULIK as the case study to discuss the wind and wave characteristics of the target area during a typhoon.

2.2. Wind and Wave Data Resources

To discuss the wind and wave conditions of the target area during Typhoon SOULIK’s impact on Taiwan, this study utilizes data from the ECMWF Reanalysis v5 (ERA5) [26] provided by the European Centre for Medium-Range Weather Forecasts (ECMWF). The ERA5 reanalysis dataset is produced by model forecasts of the ECMWF Integrated Forecast System (IFS) and using 4D-Var data assimilation techniques to integrate global observational data. It provides hourly estimates for atmospheric, ocean wave, and land surface quantities, including 10 m and 100 m wind components, significant wave height, peak wave period, wave direction, sea surface temperature, mean sea level pressure, etc. The spatial resolution of the dataset is 0.5° for ocean waves and 0.25° for atmosphere.

Due to the spatial resolution of the dataset and the invading path of Typhoon SOULIK as shown in Figure 5, this study uses wind and wave data from ERA5 at the data point of 25° N 120. 5° E (ERA5). The wind and wave data analyzed include wind speed and wind direction at 10 m and 100 m above sea level, significant wave height, wave direction, and wave period.

2.3. Characteristics of Wind and Wave Conditions During Typhoon SOULIK

The intensity variation of Typhoon SOULIK is depicted in Figure 4. It was classified as “Intense” before making landfall in Taiwan, “Moderate” while crossing Taiwan, and later downgraded to a “Severe Tropical Storm” after moving over mainland China. The typhoon’s track, sourced from the Typhoon Database of the CWA, indicates the estimated distance between the typhoon center and the target area, along with a radius of 25 m/s wind speed, as shown in Figure 5 and Figure 6. Between 00:00 and 03:00 on 13 July 2013, Typhoon SOULIK’s center was closest to the target area, with a 50 km radius of 25 m/s wind speed.

To examine wind direction (${\theta }_{wind}$), wave direction (${\theta }_{wave}$), and wind–wave misalignment during the typhoon, this study adopts the Earth coordinate system that defines north as 0°, with positive values indicating a clockwise direction and negative values indicating counterclockwise direction within the range of −180° to +180°, as shown in Figure 7. Wind and wave directions represent the direction from which they come, and wind–wave misalignment is defined as the angle difference between wind and wave direction of the smaller angle. The variations in wind direction, wave direction, and wind–wave misalignment during the Typhoon SOULIK warning period are shown in Figure 8.

The wind direction ranged from north–northwest to east–northeast 24 h before the center of the typhoon was closest to the target area. Subsequently, it shifted to the south or south–southwest during the period when the typhoon center passed through the target area between 00:00 and 03:00 on 13 July 2013. Finally, it maintained a south or south–southwest direction until the typhoon warning was lifted. The variation in wave direction during the typhoon warning period follows a pattern similar to that of wind direction, lagging behind in its changes. This results in noticeable wind–wave misalignment, especially as the typhoon’s center moves away from the target area. For instance, at 01:00 on 13th July 2013, the wind–wave misalignment reaches around 128°.

The hourly average wind speeds at a height of 10 m above sea level ($V_{10m, hour}$) and the significant wave height ($H_s$) during the warning period of Typhoon SOULIK are shown in Figure 9. Due to the typhoon center passing over the target area, the eye wall of the typhoon passes the area twice. This resulted in two peaks in wind speed and significant wave height in the target area during the typhoon warning period. One happens around 3 or 6 h before the time when the typhoon center is closest to the target area and another happens around 3 h after the time when the typhoon center is closest to the target area. During the passage of the typhoon’s eye through the target area between the two peaks, the wind speed and significant wave height decreased, dropping below the levels observed during the typhoon warning period. The detailed characteristics of wind and wave conditions during Typhoon SOULIK are shown in

Table 2. The characteristic values of the wind and wave conditions of the target area during Typhoon SOULIK.

Time (UTC)	${\mathit{V}}_{10\mathit{m}, \mathit{h}\mathit{o}\mathit{u}\mathit{r}}$ (m/s)	${\mathit{V}}_{\mathit{h}\mathit{u}\mathit{b}, \mathit{h}\mathit{o}\mathit{u}\mathit{r}}$ (m/s)	${\mathit{\theta }}_{\mathbf{wind}}$ (°)	${\mathit{H}}_{\mathit{s}}$ (m)	${\mathit{\theta }}_{\mathbf{wave}}$ (°)	${\mathit{T}}_{\mathit{p}}$ (s)	Note
12 July 2013 20:00	24,79	33.82	−9.85	5.81	8.15	8.83	Maximum wind speed before the center of Typhoon SOULIK passed through the target area
12 July 2013 21:00	22.53	31.07	−2.42	6.56	11.11	9.41	Maximum wave height before the center of Typhoon SOULIK passed through the target area
13 July 2013 01:00	12.71	15.65	−66.24	6.54	4.11	9.82	Minimum wind speed during the period when the center of Typhoon SOULIK passed through the target area
13 July 2013 02:00	16.81	20.66	−106.70	4.98	−14.94	9.08	Maximum wind and wave misalignment during the period when the center of Typhoon SOULIK passed through the target area

.

2.4. Wind and Marine Conditions for the Analysis Cases

The wind speed described in the design load case (DLC) specified in the international standard IEC 61400-3-1 is defined as the 10 min average wind speed at hub height. Therefore, this study converts the hourly average wind speed data from ERA5 to the 10 min average wind speed by dividing the correction factor (0.95) [27] provided in the IEC 61400-3-1 standard. The turbulence intensity of the wind will follow the model suggested in IEC 61400-1 [28] based on the design load case DLC 6.1, which is applied for the wind turbine operating in idle mode during extreme conditions, as 0.11. The power law model is recommended by IEC 61400-1 for describing the wind profile, and it could fit the wind profiles of typhoons below 1000 m appropriately based on Junyi He et al.’s research [29]. Therefore, the power law is used to estimate the wind speed at hub height in this study. The power law exponent for each hour is derived hourly through linear fitting of the average wind speeds at heights of 10 m and 100 m from ERA5 during Typhoon SOULIK. The results of the wind profile applied in this study are shown in Figure 10 and the following equation, where $\mathsf{\alpha }$ is the power law exponent, $V_0$ is the reference wind speed, and $z_0$ is the reference height:

$$V\left(z\right)=V_0{\left({\displaystyle \frac{z}{z_0}}\right)}^{\alpha }$$

(1)

This study explores the serviceability of FOWTs under environmental conditions during a typhoon by simulating the cases specified in

Table 3. Environmental conditions of the simulation load cases in this study, including the 10 min average wind speed at hub height, wind direction, significant wave height, wave period, wind–wave misalignment, and current velocity at z = 0. FOWTs operating in idle mode are noted as “I” and those in power production mode as “PP”.

No.	${\mathit{V}}_{\mathit{h}\mathit{u}\mathit{b},10\mathbf{min}}$ (m/s)	${\mathit{\theta }}_{\mathbf{wind}}$ (°)	${\mathit{H}}_{\mathit{s}}$ (m)	${\mathit{T}}_{\mathit{p}}$ (s)	MIS (°)	${\mathit{U}}_{\mathit{w}}\mathbf{\left(}0\mathbf{\right)}$ (m/s)	Operating Mode
A	35.60	−9.85	5.81	8.83	18.00	0.36	I
B	32.71	−2.42	6.56	9.41	13.53	0.33	I
C	21.75	−106.70	4.73	9.08	91.76	0.22	I
D	21.75	−106.70	4.73	9.08	91.76	0.22	PP
E	16.48	−66.24	4.98	9.82	70.35	0.16	I
F	16.48	−66.24	4.98	9.82	75.35	0.16	PP

. The environmental parameters and the operating mode specified in Table 3 are based on the analysis results mentioned in Section 2.3 of this study and the requirements specified in IEC TS 61400-3-2 and IEC 61400-3-1. While the wind speed is within the range for power production of the wind turbine, despite it being during the period that the typhoon strikes, the operator could either operate the wind turbine in the parked mode to reduce the risk of damage to the wind turbine or in power production mode to generate more electrical energy. Therefore, if the wind speed is within the power production range of the reference wind turbine, both idle mode and power production mode are considered in this study. The load cases in this study include the parked FOWT under extreme environmental conditions and the power-generated FOWT under a severe sea state. The former condition could correspond to the DLC 6.1 defined in IEC 61400-3-1 and the latter could correspond to DLC 1.6. All the cases must consider the effect of the wind–wave misalignment as required by IEC TS 61400-3-2.

The environmental conditions for the load cases where the wind turbine is parked (or idle) during Typhoon SOULIK are defined as follows: the maximum wind speed (i.e., case A), the maximum significant wave height (i.e., case B), the maximum wind–wave misalignment (i.e., case C), and the minimum wind speed (i.e., case E) during the typhoon eye’s passage through the site. The wind speeds in the last two cases (cases C and E) fall within the power production range. Therefore, load cases D and F are defined as the FOWT operating in idle mode, but under the same environmental conditions as cases C and E, respectively, in order to explore the impact of operating mode on the FOWT during a typhoon.

Incorporating the findings from Huang et al. [30] on the wave spectrum of the Taiwan Strait during a typhoon, the JONSWAP spectrum with a gamma value of 1.315 is applied to characterize the wave conditions during a typhoon. According to the requirements of IEC 61400-3-1, sea currents shall be considered in addition to the wind shear, wind speed, wind direction, and wave parameters when analyzing the design load case. This study applies the wind-generated current model recommended by IEC 61400-3-1, which is based on the hourly average wind speed at 10 m height above sea level ($V_{10m,hour}$) and the following equation to describe the sea current velocity ($U_w$). The sea current direction is assumed to be consistent with the wind direction as recommended by IEC 61400-3-1. The results of the sea currents are shown in Table 3.

$$U_w\left(z\right)=\{\begin{array}{cc}0.01V_{10m,hour}\left(1-{\displaystyle \frac{z}{20}}\right)& , z\le 20 m\\ 0& , z>20 m\end{array}$$

(2)

3. Simulation Methods

3.1. Target Model

The DeltaFloat platform is a shallow-draft semi-submersible platform. It consists of three pontoons connected to three columns, forming an equilateral triangle when viewed from above, as illustrated in Figure 11. The dimensions of the three columns are identical, each with a diameter of 12.5 m and a height of 35 m. The distance from each column to the center of the platform is 40 m. The dimensions of the three pontoons are also identical, with each being about 9.4 m wide and 7 m high. The platform design is based on an off-center tower system, where the wind turbine and tower are positioned on one of the three columns rather than at the center of the platform. In this study, the column on which the wind turbine and tower are installed is referred to as the main column, designated as C1. The other two columns are designated as C2 and C3, respectively. The platform is designed with a draft of 20 m, a freeboard of 15 m, and a displacement of approximately 19,000 tons. The main specifications of the platform are detailed in

Table 4. Main parameters of the DeltaFloat platform [19].

Column to Center (L) (m)	Column Diameter (D) (m)	Height of Pontoon (H) (m)	Height of Column (m)
40	12.5	7	35
Draft (T) (m)	Displacement (ton)	GMT/GML (m)
20	About 19000	8.54

.

The mooring system of DeltaFloat consists of six mooring lines, with two lines attached to each of the three columns. The mooring chain has a nominal diameter of 127 mm and a breaking strength of 12,171 kN. According to the requirements of the IEC 61400 series standards and relevant guidelines for mooring line strength, such mooring systems are considered redundant systems. The safety factor of the complete system is 1.67, resulting in a maximum allowable tension of 7288 kN. Additionally, clumps are deployed, with one placed every 2 m from the touchdown point, totaling 16 clumps per mooring line. Each clump has a dry weight of 5000 kg. The horizontal distance from the anchor point to the center of the column is 494 m. The main specifications of the mooring system are detailed in

Table 5. Main parameters of the mooring system [19].

Mooring Type	Material	Number of Lines	Clump Number per Line (pcs/mooring)
Catenary	R3 Studless Chain	6	16
Dry Line Density (kg/m3)	Unstretched Length (m)	Nominal Chain Diameter (mm)	Clump Weight (kg/pc)
20	500	127	5000

.

3.2. Reference Wind Turbine

In this study, the DeltaFloat was analyzed in conjunction with the DTU 10 MW reference wind turbine from the Technical University of Denmark (DTU). The main specifications of the wind turbine are detailed in

Table 6. Main parameters of the DTU 10 MW reference wind turbine [20].

Wind Regime	Rotor Mass (ton)	Nacelle Mass (ton)	Tower Mass (ton)
IEC Class I A	228.0	446.0	628.4
Number of blades	Rated power (MW)	Rated wind speed (m/s)	Rotor speed (rpm)
3	10	11.4	6.0~9.6
Cut-in wind speed (m/s)	Cut-out wind speed (m/s)	Rotor diameter (m)	Hub height (m)
4	25	178.3	119.0

, while the thrust and power of the wind turbine at different wind speeds are shown in Figure 12. This wind turbine is designed to achieve its rated power of 10 MW at a rated wind speed of 11.4 m/s. Additionally, the wind turbine control system is set such that the rotor of the wind turbine will generate a maximum thrust of approximately 1507.4 kN at a wind speed of 11 m/s.

3.3. Coordinate and Orientation

The model coordinate system is defined with the origin at the center of the platform’s still water plane. The X direction points towards the main column C1, and the Z direction is vertically upwards. The inflow direction angle is 0° in the positive X direction. When viewed from above, a counterclockwise rotation of the wind turbine is considered a positive angle. The model coordinate system is illustrated in Figure 11 and 13.

The wind data recorded by the metocean observation tower at the Formosa 1 offshore wind farm in Hsinchu from May 2017 to April 2018 were analyzed. The results reveal that the main wind direction in the Hsinchu offshore area of Taiwan is 30° [31]. Moreover, as noted in the study by Tzeng et al. [32], for the DeltaFloat FOWT operating in power production mode under different inflow directions, the platform’s yaw motion is around 0° and the platform’s translation in the horizontal (X-Y) plane is smaller when the inflow direction is 180° (model coordinates). Based on the above information and considering the reduction in floating wind turbine cable translation and maximizing power generation with minimal rotor–nacelle assembly (RNA) yaw system action, this study aligns the main column (C1) of the floating wind turbine to face 30° in the Earth coordinate system, as shown in Figure 13.

3.4. Numerical Model and Validation

This study utilizes the offshore marine dynamic analysis software OrcaFlex 11.4c, which enables the calculation of blade forces using blade element momentum theory, platform motion using frequency domain potential flow diffraction–radiation theory, wind loads on the tower and drag forces on mooring lines using the Morrison equation, and tension in each mooring line using the finite element method, and even incorporate wind turbine control systems. OrcaFlex 11.4c is simulating the deformation of the wind turbine tower and blades by the linear static structural analysis through FEM and coupling with the results of the aerodynamic module. The software integrates hydrodynamic and aerodynamic forces, wind turbine control systems, and mooring system responses in the time domain to solve the motion equations, as represented by Equation (3), conducting comprehensive dynamic simulations of the FOWT system:

$$\left(M+A\right)\ddot{X}+B\dot{X}+CX=F$$

(3)

where F represents external forces, including wind, waves, currents, and other loads. M and A represent the system mass and added mass coefficients, respectively. B represents the damping coefficient and C represents the restoring stiffness coefficient. $X$, $\dot{X}$, and $\ddot{X}$ represent system position, velocity, and acceleration, respectively. While OrcaFlex 11.4c performs time domain simulation, the coupling effects between motion and load are iteratively calculated at each time step until a predetermined convergence criterion is reached, ensuring that the interaction between the two is within an acceptable range, and then the next time step is entered. The numerical model of the floating wind turbine established in this study by using OrcaFlex 11.4c is shown in Figure 14.

To validate the aerodynamic performance of the wind turbine simulated in this study against DTU’s design, simulations were conducted with a fixed wind turbine operating at steady cut-in to cut-out wind speeds. The results are illustrated in Figure 13. The simulated wind turbine power (blue dashed line) closely corresponds with DTU’s data (red solid line). However, the simulated rotor thrust (orange dashed line) only displays a similar trend to DTU’s data (green solid line). When the wind speed ranges from 11 m/s to 25 m/s, the average deviation between simulated results and those of DTU is approximately 5%. However, for wind speeds below 11 m/s, the simulated results are lower than DTU’s data.

The natural period of motion of this study is compared with the scaled experimental results from the DeltaFloat design team to verify the motion simulation results of the FOWT by the numerical model of this study. As shown in

Table 7. Validation of natural period of motion [19].

	Heave (s)	Roll (s)	Pitch (s)
DeltaFloat design team	21.8	33.7	34.2
This study	21.5	30.5	31.0

, the simulation results of this study and the results of the design team are similar. The natural periods of the roll and pitch motions are slightly lower than the results of the DeltaFloat design team.

According to IEC TS 61400-3-2, the integrated load analysis of the floating offshore wind turbine shall be conducted through time domain analysis. For scenarios where the wind turbine operates in idle mode under extreme environments, DLC 6.1, as specified by IEC 61400-3-1, is applicable. The analysis of DLC 6.1 shall be performed using either a continuous one-hour time domain simulation or six separate 10 min simulations. In this study, the latter approach is applied, as varying the temporal distribution with consistent environmental parameters allows for the consideration of a broader range of environmental condition combinations. This methodology leads to more comprehensive and reliable analysis results. For time domain simulation, establishing time series environmental conditions is essential. In this study, the National Renewable Energy Laboratory’s (NREL’s) tool, TurbSim, is utilized to generate the time series wind condition with specified mean wind speed and turbulence intensity, employing the Kaimal wind spectrum and a random seed. Similarly, OrcaFlex 11.4c is used to create the time series wave condition with defined significant wave height and peak wave period, utilizing the JONSWAP wave spectrum and a random seed.

4. Results and Discussion

According to the requirement specified in IEC TS 61400-3-2 for the serviceability analysis, the motion of FOWTs shall be carefully considered to avoid damage to the installed equipment or the neighboring facilities due to exceeding the limiting value related to FOWT motion. Therefore, the acceleration of the RNA and inclination of the tower caused by the motion of the platform are discussed in this study based on the simulation results from the cases listed in Table 3. The wind and wave conditions of the simulation model gradually increase to the set values from −300 s to 0 s. The data are used for analysis once the simulation results are stabilized. The average wind speed during the analysis period and the results of RNA acceleration and platform motion are presented in

Table 8. The average wind speed during the analysis period.

Case	A	B	C	D	E	F
Average wind speed (m/s2)	35.93	33.02	22.00	22.00	16.69	16.69

,

Table 9. The maximum RNA acceleration during the typhoon, where $a_{total}$, $a_H$, and $a_V$ are the total RNA acceleration, horizontal RNA acceleration, and vertical RNA acceleration.

Case	A	B	C	D	E	F
$a_{total}$ (m/s2)	2.12	1.95	1.49	1.66	1.47	1.62
$a_H$ (m/s2)	2.11	1.92	1.48	1.66	1.45	1.61
$a_V$ (m/s2)	0.52	0.63	0.37	0.41	0.48	0.49

,

Table 10. The maximum displacement of the platform in the horizontal plane (${\Delta }_H$) and the maximum inclination of the FOWT (${\theta }_{inclination}$). The inclination of the FOWT is defined as the angle between the tower and the vertical axis due to the motion of the platform.

Case	A	B	C	D	E	F
${\Delta }_H$ (m)	3.98	3.49	2.22	3.64	2.04	4.12
${\theta }_{inclination}$ (°)	4.76	4.23	1.60	6.33	1.28	5.57

and

Table 11. The translation range of the platform along the Z axis and the rotation range of the platform with respect to the Z axis. The translation of the platform in the Z direction is defined as heave and the rotation of the platform with respect to the Z axis is defined as yaw.

Case		A	B	C	D	E	F
Heave (m)	Max	1.63	2.01	1.24	1.23	1.53	1.24
	min	−1.30	−1.95	−1.03	−1.14	−1.33	−1.41
Yaw (°)	Max	10.66	9.15	4.40	8.34	4.41	11.09
	min	0.49	−0.42	−0.60	−1.46	−0.65	2.05

. The extreme value of the results of each case in this study is defined as the average of the extreme value of six 10 min simulation results.

4.1. The RNA Acceleration of the FOWT During the Typhoon

Generally, protective measures, such as shutdown, are triggered when RNA acceleration exceeds the limiting value to prevent damage to the FOWT. Therefore, it is crucial to determine the maximum RNA acceleration of the FOWT at the specified site during the design phase. According to the simulation results of this study, the maximum RNA acceleration in case A is the highest, approximately 0.22 times the gravity acceleration. The simulation results from cases A, B, C, and E were analyzed to identify the characteristics of RNA acceleration in the FOWT operating in idle mode. As shown in Figure 15, the maximum RNA acceleration of the FOWT operating in idle mode during a typhoon increases as the wind speed increases.

To access the impact of the FOWT’s operating mode, including idle and power production, on RNA acceleration, the simulation results of cases D and F were compared with those of cases C and E. The maximum RNA acceleration in the horizontal plane for the FOWT operating in power production mode is approximately 10 to 15% higher than that in idle mode, as shown in Figure 16. The difference in maximum RNA acceleration along the vertical direction (Z-axis) for the FOWT operating in power production mode is approximately 2 to 11% higher than that in idle mode.

4.2. The Inclination of the FOWT During the Typhoon

Due to the limitations on the inclination of wind turbines, which are in place to ensure the safety and operating life of the wind turbine, the maximum inclination of the FOWT is analyzed in this study to understand its behavior during the typhoon. According to the simulation results, the maximum inclination of the FOWT in case A is the largest among all the cases operating in idle mode, approximately 4.76°, as shown in Figure 17. To explore the impact of wind speed and wind–wave misalignment on the inclination of wind turbines, the simulation results of cases where only the wind speed or wind–wave misalignment of case C as shown in

Table 12. Environmental conditions of the control group with altered wind speed or wind–wave misalignment in case C.

No.	${\mathit{V}}_{\mathit{h}\mathit{u}\mathit{b},10\mathbf{min}}$ (m/s)	${\mathit{H}}_{\mathit{s}}$ (m)	MIS(°)	No.	${\mathit{V}}_{\mathit{h}\mathit{u}\mathit{b},10\mathbf{min}}$ (m/s)	${\mathit{H}}_{\mathit{s}}$ (m)	MIS (°)
C1	16.48	4.73	91.76	C3	21.75	4.73	37.21
C2	35.60	4.73	91.76	C4	21.75	4.73	18.00

are changed are analyzed. As shown in Figure 18, the maximum inclination increases with the increase in wind speed or the decrease in wind–wave misalignment.

To understand the influence of the FOWT operating in power production mode during the typhoon on the inclination of the FOWT, the simulation results of cases D and F are compared with cases C and E. The maximum inclination of the FOWT operating in power production mode is significantly greater than when operating in idle mode and could be approximately four times, as shown in Figure 19. Furthermore, the maximum inclination of case F is greater than in case D. This may be attributed to the thrust acting on the rotor in case F, which is larger than in case D.

4.3. The Platform Yaw Motion of the FOWT During Typhoon

According to IEC 61400-3-1, yaw misalignment shall be considered during the design phase. Therefore, the yaw misalignment of the FOWT during the typhoon was analyzed. Simulation results from cases A, B, C, and E were examined to identify the characteristics of the absolute maximum yaw motion of the FOWT operating in idle mode. As shown in Figure 20, the absolute maximum yaw misalignment in case A is the largest, approximately 10.66°, which exceeds the ±8° yaw misalignment between wind and RNA required by IEC 61400-3-1. It is recommended that the yaw misalignment of FOWTs should also consider the additional yaw caused by platform movement. Furthermore, the risk of exceeding the RNA yaw limit during typhoons could be mitigated by adjusting the frequency of yaw control actions or enhancing the stiffness of the mooring system through active control to reduce the range of platform yaw motion.

To access the impact of the FOWT operation mode, including idle and power production, on platform yaw motion, the simulation results of cases D and F were compared with those of cases C and E. As shown in Figure 21, the absolute maximum platform yaw motion of the FOWT operating in power production mode is significantly larger than the FOWT operating in idle mode and could be more than two times the magnitude. In particular, for the FOWT operating in power production mode, the absolute maximum platform yaw motion in case F, with higher wind speed, is almost twice case D according to the simulation results from this study.

4.4. The Platform Translational Motion of the FOWT During a Typhoon

The movement of the FOWT could cause the cable to undergo increased bending, making it more susceptible to wear and fatigue damage. Additionally, a significant proportion of insurance claims are due to cable failure [33]. For these reasons, this study analyzes the translation motion of the FOWT during the typhoon. According to the simulation results, the maximum horizontal displacement in case A is the largest, approximately 3.94 m. The maximum range of heave motion in case B is the largest, ranging from 2.01 to −1.95 m. As shown in Figure 22, the range of heave motion for the FOWT operating in idle mode during a typhoon is positively related to the significant wave height. For the FOWT operating in power production mode, the maximum horizontal displacement of the platform is higher with stronger wind speeds compared to the cases with lower wind speeds. Generally, the aerodynamical load acting on the rotor of the FOWT in power production mode is much larger than that in idle mode. However, based on the results of this study, this does not appear to impact the horizontal displacement or the heave motion of the platform, as shown in Figure 23. To understand the influence of wind and wave direction on the movement of the FOWT during the passage of Typhoon SOULIK’s eye over the target site, the initial position and the position at maximum horizontal displacement are compared to the direction of wind and wave as shown in Figure 24. It was found that the direction of the maximum horizontal displacement of the platform is primarily influenced by wind direction.

5. Conclusions

Based on the wind and wave conditions during Typhoon SOULIK at the target site (Hsinchu offshore) and the operational mode of FOWTs, this study specifies six load cases to account for the environmental conditions and operation modes of the wind turbine during the typhoon. Integrated simulations of the FOWT were conducted to investigate its motion characteristics during the typhoon. Based on the simulation results, the following conclusions regarding the serviceability of FOWTs under typhoon conditions are presented:

• The maximum RNA acceleration of the FOWT operating in idle mode during a typhoon is positively related to the wind speed and could achieve 2.12 m/s², which is approximately 0.22 times the gravity acceleration. Additionally, the maximum RNA acceleration for the FOWT operating in power production mode is higher than that in idle mode.
• The maximum yaw motion of the platform in case A is 10.66°, which exceeds the ±8° yaw misalignment required by IEC standard 61400-3-1. It is recommended that the yaw misalignment of FOWTs should also account for the additional yaw motion caused by the movement of the platform, especially for the off-center platform.
• The direction of the maximum horizontal displacement of the floating platform is primarily affected by the wind direction, and the magnitude of the maximum horizontal displacement is positively correlated with wind speed. When the floating wind turbine operates in power production mode, the horizontal displacement is greater than when it operates in idle mode.
• During a typhoon, the maximum RNA acceleration, inclination, and yaw of the FOWT operating in power production mode is larger than those in idle mode under the same environmental conditions.

During a typhoon, the motion of the floating platform results in adverse effects, such as additional RNA yaw misalignment, acceleration at the top of the wind turbine, and wind turbine inclination. To mitigate these effects and reduce the platform’s motion response, it is recommended to use active control to increase the mooring system’s stiffness to reduce the motion response of the platform. Moreover, when designing the RNA control system, it is recommended to incorporate the yaw motion of the platform and adjust the control frequency for typhoon scenarios. These strategies could enhance the stability of floating platforms in typhoon conditions and improve the overall operational performance of wind turbines.