No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
CFD approach to evaluation of wind energy in complex terrain
Abstract
To evaluate the wind energy distribution in a complex terrain which is impossible to be done by using traditional methods, two strong wind events with stable wind direction were selected from the field record of a real complex terrain and were used as the researching objects. Then, the wind field in the terrain was meshed into hexahedral grids with two resolutions of 40 m×40 m and 20 m×20 m, and two turbulence models, namely SST k-ω and RNG k-ω, were adopted to simulate the wind speed field distribution. The simulation results of the wind indicate that SST k-ω model is superior to RNG k-ω model because it is of higher accuracy of the grid with a horizontal space of 20 m, and that, in the case of SST (20 m), the mean relative errors between the simulated and the measured wind speeds in 157° and 83° directions are respectively 6.46% and 5.50%. The authors also proposed a full-direction wind energy evaluation method applicable to complex terrains, which takes into consideration both the wind speeding ratio distribution obtained from CFD simulation and the local normal meteorological data.
... Relative to mesoscale modeling, CFD modeling has a higher spatial resolution (minimum horizontal grid spacing can be reduced to 10 m), so the model can describe more realistic terrain in a finer scale. Some experimental studies have demonstrated that CFD is capable to construct realistic fine-scale surface layer wind structure over extremely complex terrain [10][11][12] . ...
... This study used the commercial CFD software FLUENT as simulation tool. The authors demonstrated in a series of previous studies that FLUENT is capable of describing fine-scale wind structure over complex terrain [10][11][12] . For details on equations and physical models setting used in FLUENT simulations, refer to Li et al. [11] . ...
Leveraging the commercial CFD software FLUENT, the fine-scale three-dimensional wind structure over the Paiya Mountains on the Dapeng Peninsula near Shenzhen, a city on the seashore of South China Sea, during the landfall of Typhoon Molave has been simulated and analyzed. Through the study, a conceptual wind structure model for mountainous areas under strong wind condition is established and the following conclusions are obtained as follows: (1) FLUENT can reasonably simulate a three-dimensional wind structure over mountainous areas under strong wind conditions; (2) the kinetic effect of a mountain can intensify wind speed in the windward side of the mountain and the area over the mountain peak; and (3) in the leeward side of the mountain, wind speed is relatively lower with relatively stronger wind shear and turbulence.
... Symmetrical boundaries are adopted at the top and both sides of the fluid domain, while non-sliding walls are adopted for the mountains and greenhouse at the bottom of the fluid domain. The Realizable k- turbulence model, which has been widely used to simulate the wind pressure in valley areas (Xiao et al., 2009;Zhu et al., 2011;Yao, 2014), is adopted. A scalable wall function can produce consistent results for grids of arbitrary refinement and is always used in the simulation. ...
As the wind pressure distribution on plastic greenhouses in valley areas is different from that of greenhouses on plains, it is necessary to find out the variation trend of the wind pressure coefficient on greenhouses in the valley areas. Based on the ‘Realizable k-e turbulence’ model, the wind pressure characteristics on plastic greenhouses located in valley and plain areas are studied by using the verified numerical simulation method. The wind pressures on greenhouses and the corresponding fitting formula are obtained for different distances between the greenhouse and the foot of the mountain with a 0-degree wind angle. The results show that, with an increase of distance to the mountain, the positive wind pressure on the windward side of a plastic greenhouse increases and the negative wind pressure on the leeward side, roof surface and crosswind sides of the greenhouse decreases. The results from the proposed formulae are very close to those derived from the numerical simulation method, and the relative errors are all within 10%. The influence of canyon wind on the wind pressure distribution on plastic greenhouses should be considered in the design. This research can provide a reference for the wind resistance design of plastic greenhouses built in valley areas.
Plastic greenhouse; valley area; wind pressure distribution coefficient; numerical simulation; fitting formula
... It is the key procedure of the physical power prediction approach, and it directly determines the accuracy of the predicted power [3][4][5] . Computational Fluid Dynamics (CFD) model can be used to gain precise air flow field, and then more precise wind velocities on the hub height of wind turbines can be obtained according to NWP data [6][7][8] . Thus the accuracy of the wind power prediction can be increased. ...
To meet the requirements of the wind power prediction of wind farms, the wind velocity at the wind turbines' hub height were predicted in this paper, taking an actual wind farm as the example. First, many steady CFD numerical calculations were conducted to simulate the air flow field above the wind farm. Standard k-εturbulence model was adopted in the calculation. Then, taking the mesoscale NWP parameters as input data, the wind velocities were predicted by the simulated flow fields during the period of the year 2010. Compared with the measured wind velocity, the yearly mean absolute error (MAE) of predicted wind velocity for each wind turbine is less than 2m/s, and as the MAE decreases, the number of wind velocity samples increases. It shows that wind velocity prediction by this CFD method is accurate. The research in this paper may provide support for the work of wind power forecasting.
Computational fluid dynamics(CFD)is an ambitious prospect for numerical calculation of wind flow over complex terrain. The RANS model and k-ε turbulence model are employed in the study of mean wind velocity distribution over Nan'ao island. Based on the condition of neutral stratification, the incoming wind profile, turbulence model parameters and wall functions are analyzed. The wind direction is divided into 16 sectors, and then carried out wind flow numerical simulation for each direction sector. Monthly(annual)averaged wind velocities are calculated from a set of measured time-serial wind data along with the numerical simulation results. There are 12 anemometers mounted on 6 wind masts in the field, and the measurements are used to verify the calculation results. The average deviation for all measured points is 4.41%, for those points at same height as the reference point, deviations are within 3.7%.
Wind power prediction is of great significance for the safe and economic operation when large-scale wind power is connected to the electricity grid. Forecasting the wind speed accurately is essential for wind power prediction. A novel approach for short-term wind speed forecasting was put forward which is based on the computational fluid dynamics (CFD) pre-calculated flow fields (CPFF). Firstly, it discretizes the inflow conditions, and pre-simulates the wind fields affected by wind farm's terrain and roughness using CFD model on various inflow conditions. Then the flow field characteristics are extracted from all the simulated flow fields to compose a database. Finally, by coupling the mesoscale NWP input data with the reference mast, the site-specific wind at the hub height of wind turbines can be predicted using the database. This approach was verified taking a wind farm located in north China for example and the results were compared to the measured wind speed. The annual RMSE of wind velocity at every turbine's hub is less than 2.5 m/s and the MAE is less than 2.0 m/s, besides, the larger the absolute error of predicted wind velocity, the smaller its appearing probability. It can be concluded that the forecasting approach is not only of high accuracy and stability, but also short time demanding and especially practical for the engineering projects because the complicated CFD calculations were done before forecasting.
The turbulent flow over a circular hill, having a cosine-squared cross section and a maximum slope of about 32°, was investigated using split-fiber probes designed for measuring flows with a high turbulence and separation. Profiles of the means and variances for the three velocity components are presented and compared with those in the undisturbed (no-hill) boundary layer. The turbulent boundary layer separated behind the crest and reattached just at the lee foot of the hill. The pronounced speed-up of flow occurs not only at the hilltop but also at the midway slope on its side. The maximum perturbations in the longitudinal and vertical velocity variances were observed at the height of the hill (z/h=1), corresponding to the separated recirculating flow on the lee slope of the hill, while the maximum in the lateral velocity variance appeared at the height of z/h=0.125 beyond the hill, corresponding to the low-frequency motion in the wall layer.
Wind energy as a viable source of renewable energy for the 21st century will be played an integral role in Chinese electricity strategy. The research of the assessment of wind energy resource on national or regional level will deal with the questions how to evaluate the distribution of the wind resource in data-sparse areas or in areas of complex terrain and how to hint the general areas where a high wind resource may exist. Aiming to develop a technique for estimating the magnitude and distribution of wind resource over a selected area, simulations were run with an improved 3-dimensional hydrostatic model for complex terrain which indicates the wind characteristics successfully. It seemed that numerical simulation technique can be utilized in performing the regional wind energy assessments and potential wind turbine sites evaluations in the areas that have high wind resource but where data were previously not available or were very limited.
Constructing the equilibrium atmosphere boundary layer is an important yet still unsolved problem in the numerical simulation based on Reynolds Averaged Navier-Stokes (RANS) method for Computational Wind Engineering (CWE). This problem is reinvestigated from a new viewpoint using turbulence models. A set of new inflow turbulence boundary conditions is presented based on the standard k-ε model after an approximate solution is derived from the Turbulence Kinetic Energy (TKE) equation. The fact that the suggested inflow turbulence boundary conditions may produce horizontally homogeneous equilibrium ABL in SST k-ω model is verified and validated. Therefore, the proposed approach is more suitable for numerical simulation of the bluff body flows. A new method of modeling equilibrium ABL is introduced and a set of inflow turbulence boundary conditions are presented for practical applications.
Wind flow over complex and steep terrain was investigated by a wind tunnel experiment and a numerical simulation. The mean value and the turbulence of the wind flow over complex terrain in a coastal region of Japan were measured by split-fiber and X-wire probes. It was found that the split-fiber probe can give a reasonable accuracy while the conventional X-wire probe measurement contains a large error because of the existence of reverse flow and strong across wind in complex terrain. The applicability of the power law to the wind prediction over complex terrain was also examined. It was noticed that the application of the power law for the estimation of the wind speed or the turbulence intensity in complex terrain is difficult. Numerical simulations were carried out by the non-linear model developed by the authors and the conventional linear model. The wind field predicted by the non-linear model shows fairly good agreement with the experiment while conventional linear model tends to overestimate the mean wind speed and underestimate the turbulence.
Measurements of flow past simulated sinusoidal hills were taken in an atmospheric boundary layer wind tunnel (ABLWT) that modeled typical full-scale complex terrain for many wind turbine locations in the Altamont Pass, California, USA. Velocity profiles and speed-up factors for several model hills were determined. All hills modeled had the same height and sinusoidal cross-section, and length-to-width aspect ratios of infinity, four and one. Each of the three models was tested with approach wind directions from 0° to 90°, in 15° increments. It was observed that speed-up can vary significantly depending on the approaching wind direction. The effect of wind direction on speed-up was also investigated using field data from a site in the Altamont Pass. Average speed-up factor was found to vary significantly at the site in time, and as a function of atmospheric stability.
The flow solver “3DWind” is used to explore new aspects of the Askervein hill flow case. Previous work has investigated sensitivities to the grid, the inflow boundary profile, the roughness and the turbulence model. Several different linear and non-linear numerical models have also been validated by means of the Askervein hill case. This analysis focuses on the flow sensitivity to the grid spacing, the incident wind direction and the vertical resolution of topographic data. The horizontal resolution is found to be fine enough to cause only minor differences compared to a grid where every second node is removed. The vertical resolution dependence is mainly attributed to the wall functions. Simulations are performed for wind directions 200°, 205°, 210° and 215° at the reference station. The smallest directional biases compared to experimental values along a line through the hilltop are found for the directions 200° and 205°. There are larger wind direction changes along this line through the hilltop in the 200° case than in the 215° case. Still the simulation results give less veering than found in the experimental results, and this is maybe caused by a slightly stable atmosphere. The sensitivity to the vertical resolution of the topographical data is found to be particularly high close to the ground at the top of the hill; this is where the speed-up is most important. Differences decrease with the height from the ground. At higher levels the speed-ups are smaller and caused by terrain formations with larger scales.
Nowadays, environmental concerns on the causes of global warming have led to many countries to introduce renewable energy technologies like wind power. An appropriate selection of a suitable land for wind power plants can provide significant output of energy. The final goal of this study is to develop a numerical prediction model, based on computational wind engineering, as accurate as possible to predict wind energy distribution of a local area. It aims to develop a “local area wind energy prediction system” (hereafter LAWEPS). The works involved in this project are divided into three phases. In the initial phase of the study, a multi-step wind simulation with nesting method was designed. In the second phase of the work, each sub-model was coded and evaluated. Data of observation and experiment are obtained in parallel and used for verification with computation. At the present stage of the final phase of the project, the performance of the entire simulation system, LAWEPS, is tested and examined by comparing its results with measured data. Computational fluid dynamics (CFD) models for meteorological phenomena and building scale phenomena are developed within LAWEPS for large and small areas, respectively. Results obtained from LAWEPS are promising and stimulating. This paper reports on the current status of this project and highlights on the achievements obtained within this study so far.
- F A Castro
- J M Palma
- L A Silva
Castro, F. A., Palma, J. M. Silva, L. A. (2003). Simulation of the Askervein Flow.
Part 1: Reynolds Averaged Navier±Stokes Equations (k-e Turbulence Model).
Boundary-Layer Meteorology 107 (3): 501-530.
Engineering, Science, Construction, and Operations in Challenging Environments ©
- Space Earth
Earth and Space 2010: Engineering, Science, Construction,
and Operations in Challenging Environments © 2010 ASCE