Back to the programme printer.gif Print




Delegates are invited to meet and discuss with the poster presenters in this topic directly after the session 'How does the wind blow behind wind turbines and in wind farms?' taking place on Tuesday, 11 March 2014 at 16:30-18:00. The meet-the-authors will take place in the poster area.

Takeshi Kamio The University of Tokyo, Japan
Co-authors:
Takeshi Kamio (1) F P Makoto Iida (1) Chuichi Arakawa (1)
(1) The University of Tokyo, Tokyo, Japan

Printer friendly version: printer.gif Print

Presenter's biography

Biographies are supplied directly by presenters at EWEA 2014 and are published here unedited

He is a post doctoral researcher at University of Tokyo. He studies Wind Energy and other renewable energy technologies. His presentation is about the wind simulation over complex terrain.

Abstract

Comparison of les and observation on the turbulence characteristics of a complex terrain

Introduction

The purpose of this study is the numerical prediction of the turbulent wind on the complex terrain and its turbulence characteristics with the Large Eddy Simulation (LES). In this study, the LES simulation of a complex terrain site in Japan was compared to the observation for the wind speed profile, turbulence intensity and the turbulence spectrum. The measurement at the site for 3 years was carried out by the project, which was supported by the New Energy and Industrial Technology Development Organization (NEDO) of Japan.

Approach

The LES code, which is called MSSG, was developed in the Earth Simulator Center of Japan Agency for the Marine-Earth Science and Technology (JAMSTEC). The code was developed as the atmospheric simulator, and the authors applied it to the wind simulation for the wind resource assessment. The model is based on the compressive navier-stokes equation and LES turbulence model with Smagorinsky parameterization. The numerical simulations of the complex site were done with the LES code. The spatial resolutions were 10m for horizontal and 3m for vertical in the minimum. The 10m resolution digital map was used as the topographical data input. The steady inflow or cyclic condition was used as the inlet condition, and the convective condition was used for the outlet. The wall function with the roughness model was used as the ground surface condition, and the slip condition was used for the side and top. The simulations would correspond to the neutral atmospheric stability condition. From the simulation, the time series wind field data of 600 seconds, which was outputted in every 1 second, was achieved. The mean wind speed and turbulence intensity was calculated from the data, and also the turbulence spectrum was calculated by the FFT spectrum analysis. In the discussion, the numerical result was compared with the observation. In the comparison of the spectrum, the spectrum was transformed to non-dimensional form, with calculating the length scale parameters. The length scale parameters and spectrum distribution was compared with the observation. This comparison would show the validity of the model, boundary condition and resolution of the simulation.

Main body of abstract

The purpose of this study is the numerical prediction of the turbulent wind on the complex terrain and its turbulence characteristics with the Large Eddy Simulation (LES). In this study, the LES simulations of a complex terrain site in Japan were compared to the observation for the wind speed profile, turbulence intensity and the turbulence spectrum. The measurement at the site for 3 years was carried out by the project, which was supported by the New Energy and Industrial Technology Development Organization (NEDO) of Japan. At the previous event, the authors presented the similar study, but, after the some modification of the post data processing, the better result in the comparison between the computational result and the measurement was achieved. In the previous study, only 1 case simulation for a wind direction was presented. In that case (Case-A), the target was sea wind and the steady inflow condition was applied for the assumption ‘wind on the sea contains less turbulence than the land.’ In this study, 1 more case for another wind direction will be presented. In this case (Case-B), the steady inflow was not appropriate, because the target was not a sea wind but a land wind. For the land wind, the moderate turbulence is need for the inflow. Although several turbulence generation method has been already proposed, the authors employed the simple cyclic boundary condition. In that condition, the inflow is given as the mix of the initial steady condition and the turbulent flow information at the outlet of the computational domain. The numerical simulations were done with the LES code, which is called MSSG and was developed in the Earth Simulator Center of Japan Agency for the Marine-Earth Science and Technology (JAMSTEC). The code was developed as the atmospheric simulator, and the authors applied it to the wind simulation for the wind resource assessment. The model is based on the compressive navier-stokes equation and LES turbulence model with Smagorinsky parameterization. The numerical simulations of the complex site were done with the LES code. The spatial resolutions were 10m for horizontal and 3m for vertical in the minimum. The 10m resolution digital map was used as the topographical data input. The steady inflow or cyclic condition was used as the inlet condition, and the convective condition was used for the outlet. The wall function with the roughness model was used as the ground surface condition, and the slip condition was used for the side and top. The simulations would correspond to the neutral atmospheric stability condition. From the simulation, the time series wind field data of 600 seconds, which was outputted in every 1 second, was achieved. The mean wind speed and turbulence intensity was calculated from the data, and also the turbulence spectrum was calculated by the FFT spectrum analysis. In the discussion, the numerical result was compared with the observation. In the comparison of the spectrum, the spectrum was transformed to non-dimensional form, with calculating the length scale parameters. The length scale parameters and spectrum distribution was compared with the observation. In the comparisons of the mean wind speed (), turbulence intensity () and the turbulence spectrum (), the good agreement was achieved (Figures in this abstract were ones of Case-A). There was 10 percent error for the mean wind speed and turbulence intensity. The length scale and distribution of the non-dimensional spectrum were similar. These results would show the appropriateness of the numerical model, assumption of the boundary condition and computational resolution.

Conclusion

In this study, the LES simulations of a complex terrain site in Japan were compared to the observation for the wind speed profile, turbulence intensity and the turbulence spectrum. From the simulation, 10 minutes averaged wind speed and turbulence intensity were calculated, and the non-dimensional spectrum was calculated by the FFT spectrum analysis and the non-dimensional transform. In the comparisons between the computational result and the observation, the good agreement was achieved. There was 10 percent error for the mean wind speed and turbulence intensity. The length scale and distribution of the non-dimensional spectrum were similar. These results would show the appropriateness of the numerical model, assumption of the boundary condition and computational resolution. The model was based on the compressive Navier-Stokes equation and the LES turbulence model with Smagorinsky parameterization. The boundary conditions were the steady inflow and cycle condition for the inlet, the convective condition for the outlet, the slip condition for the side and top, and the wall function with the roughness model for the ground surface. The spatial resolution was 10m for the horizontal and 3m for vertical. Then, the simulation in this study corresponded to the neutral atmospheric stability condition. This might mean the averaged wind data would be equal to the neutral atmospheric condition. However, the effect of the other conditions like the unstable condition should be studied. Because MSSG, the LES code in this study, is the atmospheric model, it is easier to expand the study to the unstable condition, and it may be future work.


Learning objectives
This study shows the appropriateness of the numerical model, assumption of the boundary condition and computational resolution for a complex terrain. And the result of this study would be useful for the design of a wind turbine and the wind resource assessment of a site, used with, or before, the observation.


References
1. Bechmann A., Sørensen N.N., Hybrid RANS/LES Applied to Complex Terrain, Wind Energy 2010; 13:36-50.
2. Takanori Uchida, Yuji Ohya, Large-eddy simulation of turbulent airflow over complex terrain, Journal of Wind Engineering and Industrial Aerodynamics 2003; 91:219-229.
3. Keiko Takahashi, Xindong Peng, Ryo Onishi, Takeshi Sugimura, Mitsuru Ohdaira, Koji. Goto, Hiromitsu Fuchigami, Multi-Scale Weather/Climate Simulations with Multi-Scale Simulator for the Geoenvironment (MSSG) on the Earth Simulator, Annual Report of the Earth Simulator Center, April 2007-March 2008, 2008; 27-33.
4. Smagorinsky, J., General Circulation Experiments With The Primitive Equations, I. The Basic Experiment, Monthly Weather Review, 1963; 91:99–164.
5. Lund, TS, Wu, XH, Squires, KD, “Generation of turbulent inflow data for spatially-developing boundary layer simulations ”, JOURNAL OF COMPUTATIONAL PHYSICS , Volume.140, Issue.2 (1998), pp.233-258
6. J. C. Kaimal, J. C. Wyngaard, Y. Izumi, O. R. Cote, Spectral characteristics of surface-layer turbulence, Q. J. R, Meteorol. , 1972; 98:563-589.
7. Yutaka Hasegawa, Hitoshi Suzuki, Hiroshi Imamura, Koji Kikuyama, Kunihiko Hatabara, Yusuke Majima, Analysis of Turbulence Characteristics for Surface Layer Over Complex Terrain : Evaluation of Power Spectrum and Coherency, Transactions of the Japan Society of Mechanical Engineering (B), 2005; 711:105-112.