# Computational Fluid Dynamics (CFD) Investigation of Wind Turbine Nacelle Separation Accident over Complex Terrain in Japan

## Abstract

**:**

## 1. Introduction

## 2. Overview of Numerical Simulation Method

## 3. Outline of the Numerical Simulation Set-Up

_{in}h/ν) and was set to 10

^{4}in the present simulation. The characteristic length scales adopted for the simulation are shown in Figure 5. In the present study, h (=673.0 m) is the difference between the minimum and maximum surface elevations in the computational domain, U

_{in}is the wind velocity at the inflow boundary at the height of the maximum terrain in the computational domain, and ν is the coefficient of dynamic viscosity. The time step in the present simulation is set to Δt = 2 × 10

^{−3}h/U

_{in}.

## 4. Simulation Results and Discussion

_{scalar}/U

_{in}(=(u

^{2}+ v

^{2}+ w

^{2})

^{1/2}/U

_{in}), which were calculated at the hub heights (50.0 m above the ground surface, refer to Figure 3) of all wind turbines, WTs No. 1 to No. 6. In Figure 6, the horizontal axis indicates non-dimensional time (=T/(h/U

_{in})). For a hypothetical value of U

_{in}= 5.0 m/s for the actual wind velocity, the duration of time on the horizontal axis is approximately 45.0 min. An examination of Figure 6 reveals that an anomalous flow phenomenon is generated in the vicinity of the wind turbines, that is, the trends of the temporal change of the non-dimensional scalar wind speed are almost the same at all turbines, WTs No. 1 to No. 6. The wave pattern in the trends changes by alternating between low velocities and high velocities. As discussed in detail below, this wave pattern changes periodically, suggesting that terrain-induced turbulence is generated due to the topographic irregularities in the vicinity of the wind turbines passing through the wind turbines. Therefore, it can be speculated that all wind turbines, WTs No. 1 to No. 6, were subject to the effect of terrain-induced turbulence which originated from topographic irregularities, on a regular basis. Although it happened to be the nacelle of WT No. 3 that fell to the ground at the time of the accident, it may be claimed that this accident was bound to happen at one of the wind turbines on the wind farm and that it would have been no surprise even if the nacelle of a different wind turbine had fallen to the ground. Note that, in post-accident inspections, cracks similar to those on WT No. 3 were detected on all the turbines except for WT No. 1.

## 5. Conclusions

## Acknowledgments

## Conflicts of Interest

## References

- The Report of Wind Turbine Nacelle Separation Accident in the Taikoyama Wind Farm, Kyoto Prefecture, Japan. Available online: http://www.meti.go.jp/committee/sankoushin/hoan/denryoku_anzen/newenergy_hatsuden_wg/pdf/001_03_02.pdf (accessed on 26 December 2013). (In Japanese).
- Liu, Z.Q.; Ishihara, T. Numerical study of turbulent flow over complex topography by LES model. In Proceedings of the Sixth International Symposium on Computational Wind Engineering, Hamburg, Germany, 8–12 June 2014. [Google Scholar]
- Uchida, T.; Ohya, Y. Micro-Siting technique for wind turbine generators by using large-eddy simulation. J. Wind Eng. Ind. Aerodyn.
**2008**, 96, 2121–2138. [Google Scholar] [CrossRef] - Uchida, T.; Maruyama, T.; Ohya, Y. New evaluation technique for WTG design wind speed using a CFD-model-based unsteady flow simulation with wind direction changes. Model. Simul. Eng.
**2011**, 2011, 941870. [Google Scholar] [CrossRef] - Uchida, T.; Ohya, Y. Latest developments in numerical wind synopsis prediction using the RIAM-COMPACT CFD model. Energies
**2011**, 4, 458–474. [Google Scholar] [CrossRef] - Uchida, T. High-Resolution LES of terrain-induced turbulence around wind turbine generators by using turbulent inflow boundary conditions. Open J. Fluid Dyn.
**2017**, 7, 511–524. [Google Scholar] [CrossRef] - Uchida, T. Large-Eddy simulation and wind tunnel experiment of airflow over Bolund Hill. Open J. Fluid Dyn.
**2017**, 8, 30–43. [Google Scholar] [CrossRef] - Uchida, T. High-resolution micro-siting technique for large scale wind farm outside of japan using LES turbulence model. Energy Power Eng.
**2017**, 9, 802–813. [Google Scholar] [CrossRef] - Uchida, T. CFD Prediction of the airflow at a large-scale wind farm above a steep, three-dimensional escarpment. Energy Power Eng.
**2017**, 9, 829–842. [Google Scholar] [CrossRef] - Kim, J.; Moin, P. Application of a fractional-step method to incompressible Navier-Stokes equations. J. Comput. Phys.
**1985**, 59, 308–332. [Google Scholar] [CrossRef] - Kajishima, T.; Taira, K. Computational Fluid Dynamics: Incompressible Turbulent Flows; Springer: New York, NY, USA, 2016; ISBN 978-3319453026. [Google Scholar]
- Kawamura, T.; Takami, H.; Kuwahara, K. Computation of high reynolds number flow around a circular cylinder with surface roughness. Fluid Dyn. Res.
**1986**, 1, 145–162. [Google Scholar] [CrossRef] - Smagorinsky, J. General Circulation Experiments with the Primitive Equations, Part 1, Basic Experiments. Mon. Weather Rev.
**1963**, 91, 99–164. [Google Scholar] [CrossRef]

**Figure 2.**Still image of wind turbine accident, taken from a video by RKB Mainichi Broadcasting Corporation.

**Figure 6.**Temporal changes of scalar wind speed at the hub height (50.0 m). Duration of time shown on the horizontal axis is approximately 45.0 min (for U

_{in}= 5.0 m/s).

**Figure 8.**Temporal changes of the angle of the wind on the horizontal (x–y) and the vertical (x–z) cross-sections at the hub height (50.0 m above the terrain surface) in the case of WT No.3. Duration of non-dimensional time on the horizontal axis is approximately 45.0 min (for U

_{in}= 5.0 m/s).

**Figure 10.**Profiles of instantaneous streamwise wind velocity corresponding to Figure 9 (WT No.3).

**Figure 11.**Profiles of the standard deviations of the three components of the wind velocities at each of the wind turbine sites.

**Figure 12.**Wind characteristics in the vicinity of Taikoyama Wind Farm based on MSM-S GPV data for 2012.

**Figure 14.**Conceptual figure of wake interaction between two wind turbines by visualization of the Laplacian of the pressure field.

**Figure 15.**Results of a preliminary calculation of economic feasibility for the case of one wind turbine (hub height = 70.0 m), based on the MSM-S GPV data from the location labeled GPV1 in Figure 12.

© 2018 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Uchida, T. Computational Fluid Dynamics (CFD) Investigation of Wind Turbine Nacelle Separation Accident over Complex Terrain in Japan. *Energies* **2018**, *11*, 1485.
https://doi.org/10.3390/en11061485

**AMA Style**

Uchida T. Computational Fluid Dynamics (CFD) Investigation of Wind Turbine Nacelle Separation Accident over Complex Terrain in Japan. *Energies*. 2018; 11(6):1485.
https://doi.org/10.3390/en11061485

**Chicago/Turabian Style**

Uchida, Takanori. 2018. "Computational Fluid Dynamics (CFD) Investigation of Wind Turbine Nacelle Separation Accident over Complex Terrain in Japan" *Energies* 11, no. 6: 1485.
https://doi.org/10.3390/en11061485