Next Article in Journal
Correction: Grzegórska et al. Phytoremediation—From Environment Cleaning to Energy Generation—Current Status and Future Perspectives. Energies 2020, 13, 2905
Previous Article in Journal
Effect of Thickness on the Breakdown Characteristics of Organic Insulation Materials under Microsecond Pulse Voltage
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Energies
Volume 17
Issue 14
10.3390/en17143540
energies-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Analysis of Wind Resource Characteristics in the Ulanqab Wind Power Base (Wind Farm): Mesoscale Modeling Approach

by
Dong Xu
1,
Feifei Xue
2,*,
Yuqi Wu
3,
Yangzhou Li
2,
Wei Liu
1,
Chang Xu
2 and
Jing Sun
1
1
Northwest Engineering Corporation Limited, Xi’an 710065, China
2
College of Renewable Energy, Hohai University, Changzhou 213200, China
3
Eastern Construction Management Department of Construction Management Branch, China Three Gorges Renewables (Group) Co, Ltd., Nanjing 211100, China
*
Author to whom correspondence should be addressed.
Energies 2024, 17(14), 3540; https://doi.org/10.3390/en17143540
Submission received: 12 June 2024 / Revised: 7 July 2024 / Accepted: 15 July 2024 / Published: 19 July 2024
(This article belongs to the Section A3: Wind, Wave and Tidal Energy)
Browse Figures
Versions Notes

Abstract

:
With the rapid development and construction of large-scale wind power bases under the “Carbon Peaking and Carbon Neutrality Goals” target, traditional wind energy resource assessment methods typically rely on a limited amount of wind mast data, providing only limited wind resource analysis results. These methods are incapable of capturing the spatiotemporal distribution of wind energy resources throughout the entire base, thus failing to meet the construction requirements of wind power bases. In this study, the mesoscale WRF (The Weather Research and Forecasting Model) was employed for wind resource simulation in a large wind power base. Based on the terrain, meteorological observation data, and boundary conditions, high-resolution wind field simulation results were generated, providing more comprehensive spatiotemporal distribution information within the Ulanqab region’s wind power base. Through the analysis and comparison of measured data and simulation results at different horizontal resolutions, the model was evaluated. Taking the Ulanqab wind power base as an example, the WRF model was used to study the distribution patterns of key parameters, such as annual average wind speed, turbulence intensity, annual average wind power density, and wind direction. The results indicate that a 4 km horizontal resolution can simultaneously ensure the accuracy of wind speed and wind direction simulations, demonstrating good engineering applicability. The analysis of wind resource characteristics in the Ulanqab wind power base based on the mesoscale model provides reliable reference value and data support for its macro- and micro-siting.

1. Introduction

Wind power bases possess advantages not found in traditional wind energy developments [1]. They have high production capacity, economic benefits, and system stability, taking full advantage of geographic features to cluster numerous wind turbines and achieve large-scale wind energy utilization. Such scaled construction significantly enhances wind power generation capacity, increases electricity supply to the grid, and reduces the cost per unit of electricity, yielding better investment returns. The expansion of the base size also improves system stability; by operating multiple turbines in parallel, it can balance the fluctuations in wind energy, ensuring a continuous and stable power supply [2].
Planning for wind power bases requires extensive anemometric tower data, whereas traditional wind resource assessments mainly depend on meteorological stations or anemometric towers. These usually involve few observation sites with inconsistent observation times [3], which are inadequate for the construction needs of wind power bases. Compared to traditional methods, the mesoscale simulation approach WRF (The Weather Research and Forecasting Model) can obtain wind resource parameters for the entire base area, not limited to a few discrete sites. Integrating mesoscale simulation results with actual anemometric tower data enables a more accurate analysis of wind resource distribution characteristics, enhancing the wind resource data within the site [4].
The WRF is a mesoscale meteorological model developed by the U.S. Environmental Prediction Center and Atmospheric Research Center, among others. Its dynamical core uses a fully compressible, non-hydrostatic Eulerian model [5]; the vertical coordinate employs a terrain-following mass coordinate system; the horizontal coordinate uses an Arakawa-C grid; the time integration primarily utilizes a third-order Runge–Kutta scheme; spatial discretization ranges from second to sixth order [6]. Kine Solbakken [7] and others used WRF to assess wind simulations over Norway’s complex terrain, analyzing the impact of atmospheric input data and grid spacing on simulation results. They found that reducing grid spacing from 27 km to 9 km significantly decreases the error in ERA5 simulated wind speeds; however, reducing it further from 3 km to 1 km did not lessen the error. Kunal K.Dayal [8] and others conducted a high-resolution mesoscale wind resource assessment for Fiji, finding that the WRF model’s simulated wind resource parameters closely match measured results, providing reliable wind resource information. Xiangen Liu [9] conducted mesoscale wind field simulations in complex terrain forest areas, indicating that the Noah-MP scheme and YSU scheme are recommended for mesoscale wind simulations in such areas. Talam Enock Kibona [10] used the WRF model to forecast wind energy resources in Tanzania, demonstrating that the WRF model can predict wind characteristics over 72 h, and NCEP data can serve as a wind data source for areas with limited meteorological data.
Building on this research, this paper integrates mesoscale simulation results with actual conditions to better analyze the specific distribution of wind resources at wind power base sites, further refining the wind resource data within the Ulanqab wind power base. However, in practical engineering, an excessive pursuit of high resolution in grid simulation can diminish the impact of disturbances within the mesoscale model, leading to discrepancies between the assessed and actual values of wind resources. Therefore, it is particularly important to consider the relationship between changes in horizontal resolution and the accuracy of mesoscale simulation results. This study aims to conduct a comprehensive examination of the wind resource characteristics in the Ulanqab area using the WRF method, through systematic data collection, meticulous analysis, and precise modeling. It reveals the distribution patterns and trends of wind energy resources in the area, providing data support for the macro- and micro-siting of wind power bases.

2. Overview of the Ulanqab Wind Power Base and Experimental Design

2.1. Basic Information about the Wind Power Base

The terrain of the site is diverse, and Figure 1 shows the elevation map of the simulated area. The central geographic coordinates of the site are (112.57573°, 42.01994°), with a length of approximately 97,170 m and a width of about 56,600 m. The site covers a very wide area, but wind measurement data are only sourced from four anemometric towers within the base. The measured data cover the period from June 2010 to May 2011 (T3961 is missing data for September 2010). Two of the towers, T3943 (112.3373866°, 41.86181172°) and T3961 (112.4067652°, 42.05032825°), are located in relatively complex areas; the other two, T3730 (112.488165°, 41.87955042°) and T3863 (112.4726586°, 41.83487133°), are in relatively flat areas. Their locations are shown in the satellite image in Figure 2. The comparison results of measured data and preliminary simulation data from the four anemometric towers within the base are shown in Appendix A, Table A1. The most significant errors were observed in December; hence, subsequent studies on parameter impacts mainly focus on this month.

2.2. Selection of Parameterization Schemes

For this study, the WRF model was initialized and driven using data from the NCEP FNL Operational Model Global Tropospheric Analyses [11] (NCEP-FNL), at 6 h intervals. The land use categories in the simulation are based on the USGS (United States Geological Survey) [12] land cover classification, and the projection method used is Lambert. Table 1 summarizes the main physical parameterization schemes used in this WRF model implementation.

2.3. Experimental Design

This study utilizes the WRF model, version 4.3 [22]. Four statistical parameters are employed to perform an independence analysis of the grid: Mean Absolute Error (MAE), Root Mean Square Error (RMSE), Error, and Relative Error, which are used to comprehensively quantify the discrepancies between simulated data and observed data. The Root Mean Square Error (RMSE) represents the differences (referred to as residuals) between predicted values and observed values, indicating the dispersion of the samples; the smaller the RMSE, the better the simulation performance. The Mean Absolute Error (MAE) represents the average of the absolute errors between predicted values and observed values. It is a linear score where all individual differences are weighted equally on the average. MAE directly computes the average of the residuals, whereas RMSE gives more weight to larger differences. The distinctions between these two metrics are attributed to their convergence, uncertainty, and sensitivity to outliers [23].
Mesoscale numerical simulations are conducted using four different horizontal resolutions: 0.5 km, 1 km, 4 km, and 10 km. The nesting scenarios for the four cases are illustrated in Figure 3, with related grid configurations detailed in Table 2, the yellow box in the figure represents the simulation area, and the number 2 represents the domian 2.
The analysis of wind energy resource characteristics at the Ulanqab wind power base primarily focuses on annual average wind speed, time-averaged turbulence intensity, wind power density, and wind direction frequency. Using mesoscale results, the annual average wind speed distribution at various measurement points within the wind power base is calculated to identify the main wind speed intervals. The spatial distribution of turbulence intensity is computed to reveal the impact of terrain and meteorological factors. The wind power density in different areas is calculated to determine the regions richest in wind energy, providing references for wind farm planning. The predominant wind directions are identified to explore the characteristics of wind direction distribution and understand the variations in the wind field.

3. Grid Horizontal Resolution Selection

Four different horizontal resolutions (0.5 km, 1 km, 4 km, 10 km) are used to simulate the region of the large-scale wind power base. The time span is from December 1 to December 31, 2010. To maintain data frequency consistency, the simulation results are chosen as 1 h instantaneous values. Data at a height of 70 m at the grid points corresponding to each anemometer tower are extracted. The wind speed cloud maps are shown in Figure 4.
Combining Figure 5a,b and Table 3, it is evident that in the flat terrain at anemometer towers T3730 and T3863, the increasing resolution consistently leads to the wind speed simulation errors trending towards negative values, meaning that the finer the grid, the more likely it is to underestimate the wind speed. For T3730, the wind speed MAE/RMSE at a 4 km resolution are the smallest, at 2.38 m/s and 3.09 m/s, respectively. The wind speed simulation results for T3863 are broadly consistent. For complex terrains, as shown in Figure 5c,d and Table 3, the wind speed simulation errors for towers T3943 and T3961 are also negative. Among them, T3943 at a 4 km grid precision shows the smallest wind speed errors, with MAE/RMSE of 2.50 m/s and 3.11 m/s, respectively. T3961 performs best at WRF-0.5 km and WRF-4 km, though the differences between the two are slight, with WRF-4 km having MAE/RMSE of 3.87 m/s and 4.97 m/s, respectively, compared to the significantly larger errors in the other two cases.
Regarding wind speed errors, among the four anemometer towers, the results corresponding to the WRF-4 km scenario show the smallest errors for two of the towers, and relatively smaller errors for the other two towers in the WRF-4 km scenario. Increasing the horizontal resolution in the WRF model for calculating wind energy resources at large wind power bases does not necessarily optimize simulation effects, a finding echoed in previous studies [24,25]. This indicates that simulation accuracy is also related to terrain complexity and atmospheric stability [4].
The wind direction calculation results for flat terrains are shown in Figure 6a,b. Combined with Table 4, it can be seen that the wind direction simulation errors for anemometer towers T3730 and T3863 are normalized and achieve their minimum values under WRF-4km, at 0.96% and 1.49%, as well as 0.94% and 1.78%, respectively.
For wind direction simulations in complex terrains, an analysis of Figure 6c,d and Table 4 shows that the wind direction simulation performance of T3943 under WRF-4 km is only second to that under WRF-0.5 km; for T3961, the MAE under WRF-4 km is only 1.99%, and this condition has the smallest RMSE among the four cases.
In summary, conducting mesoscale simulations of wind power bases at a 4 km horizontal resolution can adequately balance the calculation accuracy of both wind speed and direction, while requiring fewer computational resources, making it more suitable for practical engineering applications.

4. Results and Discussion

Based on the study of grid horizontal resolution, the grid horizontal resolution for wind resource calculations is set at 4 km as shown in Figure 7. Within the site, 45 points are selected as the analysis locations for wind resource characteristics, conducting a year-long mesoscale simulation of the Ulanqab wind power base. The main selection criterion involves avoiding the boundaries to prevent the boundary effects from influencing the simulation results; each point represents the wind resource conditions at 70 m height covering a 12 km × 12 km area surrounding it. The parameters included are annual average wind speed, turbulence intensity, annual average wind power density, and various wind directions. The location information for the analysis points of mesoscale wind resource characteristics in the Ulanqab area is shown in Appendix A, Table A2, and the mesoscale simulation wind resource analysis data are shown in Table A3 and Table A4 in Appendix A.
According to Table A3 and Figure 8, Figure 9 and Figure 10, the annual average wind speed in the Ulanqab wind power base area ranges from 7.327 to 9.131 m/s; turbulence intensity ranges from 0.2877 to 0.3516; and the simulated values for annual average wind power density range from 400 to 727.83 W/m2, with small fluctuations, indicating a relatively stable wind field [26]. In conjunction with related wind resource assessment data, the wind power density level at the base is rated between 5 and 6. The wind speed is at a higher level with minimal fluctuation, showing great potential for wind energy development and suitability for the operation of wind energy equipment.
Table A4 provides the distribution of wind direction frequencies at different azimuth angles in the Ulanqab area under mesoscale simulation, with the predominant wind directions being mainly northwest (NNW), west (W), west-southwest (WSW), and north (N). To enhance the utilization rate of wind energy resources, placing wind turbines in the direction of the prevailing winds should be prioritized, and the layout of turbines should consider seasonal changes in wind direction.
In flat terrains, the wind farm can be more easily expanded, maximizing the covered area and enhancing the overall efficiency of wind energy capture. Thus, it is advisable to use an equidistant layout, meaning the distance between each wind turbine is equal. Additionally, it is necessary to optimize the layout of the wind turbines according to the local wind direction patterns, ensuring that turbines capture the maximum wind energy during seasonal wind shifts.
In complex terrains, adjusting the height of wind turbines is often more crucial. By increasing the height of the turbines, the significant impact of terrain obstacles on wind speed can be avoided, reducing the blocking effect between turbines and ensuring that they fully capture wind resources. Flow simulation and wind tunnel experiments can be conducted to understand turbulence, vortices, and wind speed variations within the wind field, determining the optimal layout of the wind turbines.

5. Conclusions

This study used the mesoscale WRF model to investigate the simulation effects of wind energy resources at the Ulanqab wind power base under different horizontal resolutions, specifically analyzing the distribution characteristics of wind resources. The main conclusions are as follows:
  • Overly fine horizontal resolution can lead to underestimations of wind speed. Conducting mesoscale simulations of large wind power bases at a 4 km horizontal resolution can adequately balance the calculation accuracy of both wind speed and direction while requiring fewer computational resources, making it more suitable for practical engineering applications.
  • In the Ulanqab wind power base area, the annual average wind speed ranges from 7.327 to 9.131 m/s; turbulence intensity ranges from 0.2877 to 0.3516; the measured values for annual average wind power density range from 400 to 727.83 W/m2; and the wind power density level is between 5 and 6. The wind speed is at a high level with minimal fluctuation, showing great potential for wind energy development and suitability for the operation of wind energy equipment.
  • In the Ulanqab wind power base, an equidistant layout is recommended for flat terrains to maximize the wind farm’s coverage area. In complex terrains, adjusting the height of wind turbines is more crucial. Increasing the height appropriately can avoid significant impacts of terrain obstacles on wind speed, reduce blocking effects between turbines, and ensure that turbines fully capture wind resources.
Future research could consider using more meteorological data sources to further improve simulation accuracy. For example, incorporating temperature and humidity to gain a more comprehensive understanding of wind resources, refining numerical simulation methods, and enhancing simulation accuracy, especially under complex terrain and variable meteorological conditions. Based on the simulation results and measured data, the planning and layout of wind farms can be optimized to maximize the utilization of available wind resources.

Author Contributions

Methodology, C.X.; Project administration, F.X.; Resources, D.X., W.L. and J.S.; Writing—original draft, Y.W.; Writing—review and editing, Y.L. All authors have read and agreed to the published version of the manuscript.

Funding

This work was supported by the China Power Construction Corporation Research Project, under grant number XBY-ZDKJ-2020-05, the China Power Construction Corporation Research Project under grant number DJ-ZDXM-2020-52, the Fundamental Research Funds for the Central Universities under grant number B240201171, and the National Natural Science Foundation of China under grant number 52106238.

Data Availability Statement

The original contributions presented in the study are included in the article. Further inquiries can be directed to the corresponding authors.

Conflicts of Interest

Authors Dong Xu, Wei Liu and Jing Sun were employed by the company Northwest Engineering Corporation Limited. Author Yuqi Wu was employed by the company China Three Gorges Renewables (Group) Co, Ltd. Eastern construction Management Department of Construction Management Branch. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Appendix A

Table A1. Comparison of daily average wind speed MAE, daily average wind speed RMSE, and monthly average wind speed relative error between preliminary simulation data and actual measurements for (a) T3943, (b) T3961, (c) T3730, and (d) T3863.
Table A1. Comparison of daily average wind speed MAE, daily average wind speed RMSE, and monthly average wind speed relative error between preliminary simulation data and actual measurements for (a) T3943, (b) T3961, (c) T3730, and (d) T3863.
(a)
MonthMAE/m·s−1RMSE/m·s−1Relative Error
2010-60.9720691.36900512.2970%
2010-71.4953901.74647418.6485%
2010-81.3087501.64996314.7526%
2010-91.6475142.06455623.2959%
2010-101.5624061.99841718.8081%
2010-114.3138194.89511735.9706%
2010-124.5515735.13445632.9202%
2011-12.8894093.51675624.5647%
2011-22.3082892.79717828.2564%
2011-32.6707123.47111525.6852%
2011-42.0077642.46614320.1443%
(b)
MonthMAE/m·s−1RMSE/m·s−1Relative Error
2010-60.9758751.1387583.1128%
2010-71.1037371.3635438.1692%
2010-81.2732531.56889412.5209%
2010-101.3069621.6867628.1646%
2010-113.4322623.97680626.3689%
2010-124.3802365.21180630.7107%
2011-12.4235283.13941323.3169%
2011-21.9299702.72165418.9256%
2011-31.5440192.0986999.5772%
2011-41.5549722.0369028.4051%
2011-51.7582122.37402711.8054%
(c)
MonthMAE/m·s−1RMSE/m·s−1Relative Error
2010-61.1499261.44070713.7023%
2010-70.8964651.0855597.8385%
2010-80.8118151.0184588.4693%
2010-90.9016531.1719476.5620%
2010-100.7368680.8475176.9692%
2010-110.8592071.1199495.2778%
2010-121.4144891.9696617.1658%
2011-11.0552691.2443863.6111%
2011-20.8672021.0177386.5577%
2011-30.8654301.0993652.5069%
2011-40.7390830.9642270.0835%
2011-51.0051881.3074211.3272%
(d)
MonthMAE/m·s−1RMSE/m·s−1Relative Error
2010-60.8289181.0405760.2558%
2010-70.969571.2257484.5730%
2010-80.8301481.0884856.4043%
2010-91.0476211.3151226.4676%
2010-100.9969891.2251638.4948%
2010-112.3209542.62175520.9120%
2010-123.0757263.44182923.6107%
2011-12.2442882.67895619.2064%
2011-21.3554171.66542112.9593%
2011-31.8506452.18796717.2335%
2011-41.178281.45561111.1761%
2011-51.4632661.91680612.0390%
Table A2. Analysis of the characteristics of the medium-scale wind resources in the Ulanqab region.
Table A2. Analysis of the characteristics of the medium-scale wind resources in the Ulanqab region.
Point NumberLongitude/°Latitude/°
1112.002641.77696
2112.001541.88849
3112.000442.00002
4112.000942.11265
5111.998242.22314
6111.998242.22314
7111.998242.22314
8111.998242.22314
9112.149842.11224
10112.14942.22381
11112.301741.77807
12112.301241.88959
13112.300742.00114
14112.300242.11269
15112.299742.22426
16112.451341.77831
17112.451141.88984
18112.450942.00139
19112.450642.11293
20112.450442.22451
21112.600941.77834
22112.60141.88987
23112.60142.00143
24112.601142.11298
25112.601142.22455
26112.750541.77818
27112.750941.88971
28112.751242.00124
29112.751542.1128
30112.751842.22437
31112.900141.77779
32112.900741.88932
33112.901342.00087
34112.901942.11242
35112.902542.22397
36113.049741.77721
37113.050641.88873
38113.051542.00027
39113.052442.11182
40113.053342.22339
41113.199341.7764
42113.200441.88792
43113.201641.99947
44113.202842.11102
45113.20442.22258
Table A3. Mid-scale wind resource table for Ulanqab region (annual average wind speed, turbulence intensity, annual average wind power density).
Table A3. Mid-scale wind resource table for Ulanqab region (annual average wind speed, turbulence intensity, annual average wind power density).
Point NumbeAnnual Average Wind Speed/m·s−1Turbulence IntensityAnnual Average Wind Power Density/W·m−2
18.10411010.3229286513.6839459
28.07675850.3199952499.2297909
38.39985240.3186319560.1277942
48.4835430.319104582.9049902
58.4729750.320163585.9553334
68.4719180.334064587.5160964
78.3784920.319339548.9150254
88.4215470.327882579.4736091
98.8964530.346759700.437858
108.7279210.33253654.3782222
118.6514240.343526636.4947688
129.1318440.340107727.8269875
138.8225810.334211675.0139863
148.2519270.312326536.0354253
158.3279870.32769564.9023567
168.4505910.343339595.3066167
178.4886630.328766597.6369231
188.8725230.34792713.2900586
198.885020.351634708.592505
208.3886990.330233592.538197
217.9113850.328983483.8715135
228.808340.325949651.0879919
238.4570090.325289605.6354951
247.7738240.337327477.0047165
258.3020270.325018562.9418033
267.4871730.31269411.9261171
277.4229820.314516410.5165645
287.3271780.322299400.671643
297.431280.295024390.4567518
307.6543530.301992438.9282381
317.3598710.302284385.6770955
327.7704680.308827453.2196023
337.8059370.293372449.7654077
348.4030740.304882556.9544909
357.9433440.31121495.1347244
367.7344930.28777430.8280079
377.5398790.293231408.1337379
388.2977720.302325539.747534
397.9163370.300147475.6981808
407.8911230.303338471.0177723
418.1652830.298124501.5179189
428.1121450.290225490.4749924
438.2187560.291404515.6343216
448.1476430.298013513.238947
458.0834550.306622512.1800006
Table A4. Mid-scale wind resource table for Ulanqab region (wind direction).
Table A4. Mid-scale wind resource table for Ulanqab region (wind direction).
NumberAzimuth Angle FrequencyNNNENEENE
10.140.1790.0980.047
20.1390.1910.10.045
30.1650.1880.0890.042
40.1920.170.0780.039
50.2090.1480.0720.042
60.170.2040.0990.044
70.1580.1980.1030.048
80.1690.1580.0750.041
90.2060.1750.0750.044
100.2230.1510.0790.046
110.1810.220.0910.043
120.1760.2080.1080.053
130.1970.1810.0850.051
140.1710.1690.0830.053
150.1970.1480.0870.042
160.1880.2210.0910.048
170.1680.1780.1060.057
180.2020.1890.0880.054
190.2010.1830.0970.052
200.2040.1640.0880.047
210.1860.1940.1040.055
220.1560.2090.120.062
230.1620.1920.1070.062
240.1650.1660.1040.059
250.1830.1910.0860.059
260.1640.1650.110.068
270.1240.1560.1410.086
280.1120.2020.1360.069
290.1430.1740.1060.062
300.1550.1690.0820.056
310.1340.1540.1360.083
320.1030.1710.1710.071
330.1140.2140.1160.059
340.1570.1770.1020.057
350.1660.1770.090.053
360.1210.150.1650.067
370.1130.1810.1270.061
380.1260.1880.1180.068
390.1370.1710.1190.062
400.1610.1770.090.053
410.1170.1740.1540.071
420.1180.1730.1350.07
430.1160.1770.1350.064
440.140.1840.1060.057
450.1590.1740.0920.058
NumberAzimuth Angle FrequencyNNNENEENE
10.0290.0260.0240.021
20.030.0260.0270.022
30.0320.0240.0240.028
40.0350.0240.0240.024
50.0320.0220.0260.028
60.0320.0250.0210.024
70.0350.030.0260.021
80.0350.0310.0270.025
90.0350.0270.0260.025
100.0340.0260.0280.029
110.0330.0280.0310.022
120.0390.0280.0280.02
130.0320.0240.0250.024
140.0360.0250.0250.021
150.0350.0250.0270.027
160.0430.0370.0210.02
170.0440.0430.0250.014
180.0380.0320.0280.021
190.0360.0310.0280.02
200.0340.0270.0310.024
210.0470.0380.0250.021
220.0470.0350.0250.018
230.0420.0340.0220.016
240.0420.0330.0270.018
250.040.0330.0290.023
260.060.0360.0210.014
270.0540.0330.020.012
280.0410.0290.0220.015
290.0420.0370.0290.019
300.040.0330.0320.022
310.0410.0280.0240.015
320.0410.0290.0260.016
330.0420.0340.0290.021
340.0440.0380.0290.022
350.0410.0340.0270.022
360.0470.0290.0230.015
370.0490.0330.0260.015
380.0520.0310.0250.014
390.0490.0310.0270.017
400.0430.0350.0280.025
410.0510.0310.0220.016
420.0520.030.0220.015
430.0510.030.0250.014
440.0490.0310.0280.017
450.0460.0320.0270.023
NumberAzimuth Angle FrequencyNNNENEENE
10.0210.0180.0180.017
20.0150.0180.0250.021
30.0220.0150.0160.019
40.0240.0210.0190.016
50.0240.0210.020.017
60.0210.0180.0230.019
70.0150.0170.0250.022
80.0180.0150.0160.02
90.0190.0220.0180.019
100.0240.0180.0210.022
110.0160.0180.0210.021
120.0140.0140.020.023
130.0190.0170.0210.022
140.020.0240.0220.021
150.020.0170.0210.023
160.0130.0150.020.023
170.0110.0110.0180.026
180.0140.0150.0160.023
190.0170.0190.0240.029
200.0170.0150.0220.034
210.0120.0150.0220.026
220.0110.0150.0210.03
230.0130.0160.0240.034
240.0110.0150.0230.05
250.0170.0140.0140.041
260.0090.0150.0240.035
270.0130.0120.0220.042
280.0110.0130.0220.074
290.0110.0130.0160.066
300.0190.0150.0110.028
310.0090.0170.0290.046
320.0130.0160.0240.047
330.0110.0160.0160.05
340.0120.0160.0140.03
350.0170.0170.0190.033
360.010.0150.0280.055
370.0110.0160.0210.05
380.010.0160.0170.041
390.0110.0150.0160.036
400.0160.0140.0130.034
410.010.0160.0250.054
420.0110.0140.0210.053
430.010.0160.0150.045
440.0150.0160.0140.035
450.0140.0180.0150.029
NumberAzimuth Angle FrequencyNNNENEENE
10.0310.0780.1210.132
20.0270.0710.1190.123
30.0310.0740.1140.116
40.0280.0730.1090.124
50.0240.0670.1060.141
60.0260.0530.1030.121
70.0270.0460.1090.12
80.030.0610.1310.148
90.0240.0430.1130.13
100.0220.0410.1010.134
110.0290.0430.0930.112
120.0270.0370.080.126
130.0280.040.0990.136
140.0250.0440.120.14
150.0230.050.1240.132
160.0290.0450.0830.103
170.0340.0440.0850.136
180.0390.0430.0760.12
190.0320.0310.0660.133
200.0360.0320.090.134
210.0270.0480.0870.094
220.0280.0390.0760.108
230.030.0450.0880.114
240.0370.0530.0790.12
250.0560.0470.0610.107
260.0390.0560.1020.082
270.0560.0580.0820.088
280.070.0460.0580.078
290.0710.0470.0670.097
300.0660.0760.0910.104
310.0470.0580.1010.08
320.0510.0540.0840.082
330.060.070.0750.074
340.0560.0740.0720.1
350.0490.0710.080.103
360.0540.050.0820.09
370.0610.0660.0890.081
380.0620.0640.0870.08
390.0670.0710.0880.084
400.0740.0720.0750.092
410.0610.0470.0640.087
420.0650.0540.0740.092
430.0670.0640.0880.082
440.0680.0730.0830.084
450.0670.0720.0810.093

References

  1. Wang, Q.; Luo, K.; Wu, C.L.; Zhu, Z.F.; Fan, J.R. Mesoscale simulations of a real onshore wind power base in complex terrain: Wind farm wake behavior and power production. Energy 2022, 241, 122873. [Google Scholar] [CrossRef]
  2. Prósper, M.A.; Otero-Casal, C.; Fernández, F.C.; Miguez-Macho, G. Wind power forecasting for a real onshore wind farm on complex terrain using WRF high resolution simulations. Renew. Energy 2019, 135, 674–686. [Google Scholar] [CrossRef]
  3. Sun, X.J.; Huang, D.G. An Explosive Growth of Wind Power in China. Int. J. Green Energy 2014, 11, 849–860. [Google Scholar] [CrossRef]
  4. Yuan, R.; Ji, W.; Luo, K.; Wang, J.; Zhang, S.; Wang, Q.; Fan, J.; Ni, M.; Cen, K. Coupled wind farm parameterization with a mesoscale model for simulations of an onshore wind farm. Appl. Energy 2017, 206, 113–125. [Google Scholar] [CrossRef]
  5. Laprise, R. The Euler equations of motion with hydrostatic pressure as an independent variable. Mon. Weather Rev. 1992, 120, 197–207. [Google Scholar] [CrossRef]
  6. Fang, F.; Xiaoning, Z.; Guishan, Y.; Yang, H.; Yuchen, H. Assessment of the Impact of Wake Interference Within Onshore and Offshore Wind Farms Based on Mesoscale Meteorological Model Analysis. Proc. CSEE 2022, 42, 4848–4859. (In Chinese) [Google Scholar] [CrossRef]
  7. Solbakken, K.; Birkelund, Y.; Samuelsen, E.M. Evaluation of surface wind using WRF in complex terrain: Atmospheric input data and grid spacing. Environ. Modell Softw. 2021, 145, 105182. [Google Scholar] [CrossRef]
  8. Dayal, K.K.; Bellon, G.; Cater, J.E.; Kingan, M.J.; Sharma, R.N. High-resolution mesoscale wind-resource assessment of Fiji using the Weather Research and Forecasting (WRF) model. Energy 2021, 232, 121047. [Google Scholar] [CrossRef]
  9. Liu, X.; Cao, J.; Xin, D. Wind field numerical simulation in forested regions of complex terrain: A mesoscale study using WRF. J. Wind Eng. Ind. Aerodyn. 2022, 222, 104915. [Google Scholar] [CrossRef]
  10. Kibona, T.E. Application of WRF mesoscale model for prediction of wind energy resources in Tanzania. Sci. Afr. 2020, 7, e00302. [Google Scholar] [CrossRef]
  11. National Centers for Environmental Prediction; National Weather Service; NOAA; U.S. Department of Commerce. NCEP FNL Operational Model Global Tropospheric Analyses, Continuing from July 1999; National Center for Atmospheric Research, Computational and Information Systems Laboratory: Boulder, CO, USA, 2000. [CrossRef]
  12. López-Vázquez, C.; Ariza-López, F.J. Global Digital Elevation Model Comparison Criteria: An Evident Need to Consider Their Application. ISPRS Int. J. Geo-Inf. 2023, 12, 337. [Google Scholar] [CrossRef]
  13. Thompson, G.; Field, P.R.; Rasmussen, R.M.; Hall, W.D. Explicit Forecasts of Winter Precipitation Using an Improved Bulk Microphysics Scheme. Part II: Implementation of a New Snow Parameterization. Mon. Weather Rev. 2008, 136, 5095–5115. [Google Scholar] [CrossRef]
  14. Hong, S.; Lim, J.J. The WRF single-moment 6-class microphysics scheme (WSM6). Asia-Pac. J. Atmos. Sci. 2006, 42, 129–151. [Google Scholar]
  15. Hong, S.-Y.; Lim, J.-O.J. The WRF single–moment 6–class microphysics scheme (WSM6). J. Korean Meteor. Soc. 2017, 42, 129–151. [Google Scholar] [CrossRef]
  16. Dudhia, J. Numerical study of convection observed during the winter monsoon experiment using a mesoscale two-dimensional model. J. Atmos. Sci. 1989, 46, 3077–3107. [Google Scholar] [CrossRef]
  17. Mlawer, E.J.; Taubman, S.J.; Brown, P.D.; Iacono, M.J.; Clough, S.A. Radiative transfer for inhomogeneous atmospheres: RRTM, a validated correlated-k model for the longwave. J. Geophys. Res. Atmos. 1997, 102, 16663–16682. [Google Scholar] [CrossRef]
  18. Paulson, C.A. The mathematical representation of wind speed and temperature profiles in the unstable atmospheric surface layer. J. Appl. Meteorol. Clim. 1970, 9, 857–861. [Google Scholar] [CrossRef]
  19. Chen, F.; Dudhia, J. Coupling an advanced land surface–hydrology model with the Penn State–NCAR MM5 modeling system. Part I: Model implementation and sensitivity. Mon. Weather Rev. 2001, 129, 569–585. [Google Scholar] [CrossRef]
  20. Hong, S.; Noh, Y.; Dudhia, J. A new vertical diffusion package with an explicit treatment of entrainment processes. Mon. Weather Rev. 2006, 134, 2318–2341. [Google Scholar] [CrossRef]
  21. Grell, G.A.; Freitas, S.R. A scale and aerosol aware stochastic convective parameterization for weather and air quality modeling. Atmos. Chem. Phys. 2014, 14, 5233–5250. [Google Scholar] [CrossRef]
  22. Skamarock, W.C.; Klemp, J.B.; Dudhia, J.; Gill, D.O.; Liu, Z.; Berner, J.; Wang, W.; Powers, J.G.; Duda, M.G.; Barker, D.; et al. A Description of the Advanced Research WRF Model Version 4.3. 2021, No. NCAR/TN-556+STR. Available online: https://opensky.ucar.edu/islandora/object/opensky:2898 (accessed on 14 July 2024).
  23. Schneider, P.; Xhafa, F. Chapter 3—Anomaly detection: Concepts and methods. In Anomaly Detection and Complex Event Processing over IoT Data Streams; Academic Press: Cambridge, MA, USA, 2022; pp. 49–66. [Google Scholar] [CrossRef]
  24. Marjanovic, N.; Wharton, S.; Chow, F.K. Investigation of model parameters for high-resolution wind energy forecasting: Case studies over simple and complex terrain. J. Wind Eng. Ind. Aerodyn. 2014, 134, 10–24. [Google Scholar] [CrossRef]
  25. Giannakopoulou, E.; Nhili, R. WRF model methodology for offshore wind energy applications. Adv. Meteorol. 2014, 2014, 319819. [Google Scholar] [CrossRef]
  26. Yang, J.; Zhang, C.; Sun, Q.Y.; Wang, G. Review of wind farm site selection. Acta Energiae Solaris Sin. 2012, 33, 136–144. [Google Scholar]
Energies 17 03540 g001
Figure 1. Topographic elevation map of the Ulanqab wind power base area.
Figure 1. Topographic elevation map of the Ulanqab wind power base area.
Energies 17 03540 g001
Energies 17 03540 g002
Figure 2. Satellite images of the wind measurement towers: (a) T3943; (b) T3961; (c) T3730; (d) T3863.
Figure 2. Satellite images of the wind measurement towers: (a) T3943; (b) T3961; (c) T3730; (d) T3863.
Energies 17 03540 g002
Energies 17 03540 g003
Figure 3. Nesting schematic for each case: (a) WRF-0.5 km; (b) WRF-1 km; (c) WRF-4 km; (d) WRF-10 km.
Figure 3. Nesting schematic for each case: (a) WRF-0.5 km; (b) WRF-1 km; (c) WRF-4 km; (d) WRF-10 km.
Energies 17 03540 g003
Energies 17 03540 g004
Figure 4. Mesoscale simulation of wind speed at 70 m (simulates a certain moment in time).
Figure 4. Mesoscale simulation of wind speed at 70 m (simulates a certain moment in time).
Energies 17 03540 g004
Energies 17 03540 g005
Figure 5. Comparison of wind speed between WRF results and observed data: (a) T3730; (b) T3863; (c) T3943; (d) T3961.
Figure 5. Comparison of wind speed between WRF results and observed data: (a) T3730; (b) T3863; (c) T3943; (d) T3961.
Energies 17 03540 g005
Energies 17 03540 g006
Figure 6. Comparison of wind direction between WRF results and observed data: (a) T3730; (b) T3863; (c) T3943; (d) T3961.
Figure 6. Comparison of wind direction between WRF results and observed data: (a) T3730; (b) T3863; (c) T3943; (d) T3961.
Energies 17 03540 g006
Energies 17 03540 g007
Figure 7. Illustrative wind map of resource analysis locations in the Ulanqab area; ✩: Site Center.
Figure 7. Illustrative wind map of resource analysis locations in the Ulanqab area; ✩: Site Center.
Energies 17 03540 g007
Energies 17 03540 g008
Figure 8. Illustrative map of annual average wind speed in the Ulanqab wind power base area based on mesoscale simulations.
Figure 8. Illustrative map of annual average wind speed in the Ulanqab wind power base area based on mesoscale simulations.
Energies 17 03540 g008
Energies 17 03540 g009
Figure 9. Illustrative map of turbulence intensity in the Ulanqab wind power base area based on mesoscale simulations.
Figure 9. Illustrative map of turbulence intensity in the Ulanqab wind power base area based on mesoscale simulations.
Energies 17 03540 g009
Energies 17 03540 g010
Figure 10. Illustrative map of annual wind power density in the Ulanqab wind power base area based on mesoscale simulations.
Figure 10. Illustrative map of annual wind power density in the Ulanqab wind power base area based on mesoscale simulations.
Energies 17 03540 g010
Table 1. Key physical parameterization schemes used in the current WRF model.
Table 1. Key physical parameterization schemes used in the current WRF model.
Scheme NameParameterization Scheme
mp_physicsThompson Scheme [13]
WRF Single–moment 6–class Scheme [14,15]
ra_lw_physicsDudhia Shortwave Scheme [16]
ra_lw_physicsRRTM Longwave Scheme [17]
sf_sfclay_physicsMM5 Similarity Scheme [18]
sf_surface_physicsUnified Noah Land Surface Model [19]
bl_pbl_physicsYonsei University Scheme (YSU) [20]
cu_physicsGrell–Freitas Ensemble Scheme [21] (dx ≥ 5000 m)
Table 2. Simulation parameter settings for four horizontal resolutions.
Table 2. Simulation parameter settings for four horizontal resolutions.
CaseGrid Scheme
(nx × ny × nz)		Horizontal Resolution/kmTime Step/sNumber of Computer Cores, Required Time/h
WRF-0.5 km160 × 120 × 40,
246 × 176 × 40		2500, 50015, 332, 160
WRF-1 km100 × 80 × 40,
146 × 116 × 40		5000, 100030, 632, 50
WRF-4 km36 × 30 × 40,
36 × 26 × 40		20,000, 4000120, 244, 10
WRF-10 km40 × 40 × 40,
28 × 28 × 40		30,000, 10,000180, 604, 5
Table 3. Wind speed error for grid independence study.
Table 3. Wind speed error for grid independence study.
Anemometer Tower NumberCaseMAE
/m·s−1		RMSE
/m·s−1		Monthly Average Wind Speed Error/m·s−1Monthly Average Wind Speed Relative Error/%
T3730WRF-0.5 km3.044.14−1.71−14.97
WRF-1 km3.124.28−1.91−16.72
WRF-4 km2.383.09−0.32−2.82
WRF-10 km2.653.380.665.74
T3863WRF-0.5 km3.094.12−2.25−16.83
WRF-1 km3.214.44−2.43−18.15
WRF-4 km2.753.43−2.06−15.41
WRF-10 km2.653.41−1.58−11.78
T3943WRF-0.5 km2.713.49−1.97−13.65
WRF-1 km3.124.28−1.91−16.72
WRF-4 km2.53.11−1.85−12.84
WRF-10 km2.843.58−2.23−15.50
T3961WRF-0.5 km3.494.7−1.42−9.82
WRF-1 km4.085.204−1.78−12.05
WRF-4 km3.874.97−1.61−10.89
WRF-10 km4.255.35−2.56−17.30
Table 4. Wind direction error for grid independence study.
Table 4. Wind direction error for grid independence study.
Anemometer Tower NumberCaseMAERMSE
T3730WRF-0.5 km2.29%4.13%
WRF-1 km1.91%3.39%
WRF-4 km0.96%1.49%
WRF-10 km1.09%1.96%
T3863WRF-0.5 km1.68%3.17%
WRF-1 km1.29%2.67%
WRF-4 km0.94%1.78%
WRF-10 km1.76%3.80%
T3943WRF-0.5 km2.76%5.37%
WRF-1 km3.38%6.59%
WRF-4 km3.16%6.47%
WRF-10 km3.83%7.05%
T3961WRF-0.5 km1.88%4.32%
WRF-1 km1.85%4.05%
WRF-4 km1.99%3.92%
WRF-10 km3.39%6.74%
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Xu, D.; Xue, F.; Wu, Y.; Li, Y.; Liu, W.; Xu, C.; Sun, J. Analysis of Wind Resource Characteristics in the Ulanqab Wind Power Base (Wind Farm): Mesoscale Modeling Approach. Energies 2024, 17, 3540. https://doi.org/10.3390/en17143540

AMA Style

Xu D, Xue F, Wu Y, Li Y, Liu W, Xu C, Sun J. Analysis of Wind Resource Characteristics in the Ulanqab Wind Power Base (Wind Farm): Mesoscale Modeling Approach. Energies. 2024; 17(14):3540. https://doi.org/10.3390/en17143540

Chicago/Turabian Style

Xu, Dong, Feifei Xue, Yuqi Wu, Yangzhou Li, Wei Liu, Chang Xu, and Jing Sun. 2024. "Analysis of Wind Resource Characteristics in the Ulanqab Wind Power Base (Wind Farm): Mesoscale Modeling Approach" Energies 17, no. 14: 3540. https://doi.org/10.3390/en17143540

APA Style

Xu, D., Xue, F., Wu, Y., Li, Y., Liu, W., Xu, C., & Sun, J. (2024). Analysis of Wind Resource Characteristics in the Ulanqab Wind Power Base (Wind Farm): Mesoscale Modeling Approach. Energies, 17(14), 3540. https://doi.org/10.3390/en17143540

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop