Mechatronics, Electrical Power, and Vehicular Technology 05 (2014) 45-50 Mechatronics, Electrical Power, and Vehicular Technology e-ISSN: 2088-6985 p-ISSN: 2087-3379 Accreditation Number: 432/Akred-LIPI/P2MI-LIPI/04/2012 www.mevjournal.com © 2014 RCEPM - LIPI All rights reserved doi: 10.14203/j.mev.2014.v5.45-50 CFD AND WIND TUNNEL ANALYSIS FOR MOUNTED-WIND TURBINE IN A TALL BUILDING FOR POWER GENERATION Dany Perwita Sari a, *, Kang-Pyo Cho b a Research Center for Biomaterials, Indonesian Institute of Sciences (LIPI) Cibinong Science Center - Bogor, Jl. Raya Bogor Km. 46, Cibinong - Bogor 16911, Indonesia b CKP Wind Solutions 992-73 Dongjisan-ri, Cheongha-myeon, Gimje Si, Jeollabuk-do 576-893, Korea Received 08 January 2014; received in revised form 17 February 2014; accepted 19 February 2014 Published online 23 July 2014 Abstract A mounted wind turbine on the top of a tall building may provide high wind power in regions of high wind speed and low turbulence. The objective of this study is to evaluate wind speed on roof top models to optimize the wind turbine performance for power generation. Comparative analyses from three different roof top models were conducted. Computational Fluid Dynamics (CFD) simulation and wind tunnel testing were performed to evaluate the performance of wind turbine. Wind speed on the building model with a geometric scale of 1:150 was measured in CFD simulation then it was validated in wind tunnel test. Results presented in this paper suggest that an increase of wind speed could be achieved with ¼ circular shapes around the rooftop which can provide additional wind speed of 55.24%, respectively. Keywords: wind speed, roof shape, CFD, wind tunnel, tall building. I. INTRODUCTION In order to cope with global climate changes, it is urgently necessary to further develop new power energy system in which distributed energy is produced based on renewable energy source. Indonesia government created legislatively binding target to reduce fossil fuel consumption by 26% by the year 2020 [1] with the vision of energy mixes originated from four main sources: oil (30%), coal (22%), gas (23%) and new renewable energy (25%) [2]. Hence, tall building may use insulation and low energy devices to reduce energy and electricity consumption. Nowadays architects and designers are starting to seek out an innovative tall building design. They are important players behind the commercial greening movement [3]. In fact, green building is considered one of the most cost-efficient ways of reducing greenhouse gas emissions [4]. Mounted wind turbine on the top of a tall building is one of green buildings which on the rise. However, installations of a Building Integrated Wind Turbines (BIWT) are still limited because of low mean wind speeds, high levels of turbulence and relatively high aerodynamic noise levels generated by the turbine [5]. Abohela et al. [6] demonstrated that roof shapes could maximize energy harvesting from acceleration of the wind above the building. Wind turbine power of a wind generator [7] can be expressed as follows: 𝑃𝑡𝑢𝑟𝑏𝑖𝑛𝑒 = 1 2 𝜌𝐶𝑝𝐴𝑉 3 (1) where 𝑃𝑡𝑢𝑟𝑏𝑖𝑛𝑒 is the wind turbine power, 𝜌 is the air density (kg/m 3 ), 𝐶𝑝 is the coefficient of performance, A is the swept area of the blades (m2), and V is free wind speed (m/s). From equation (1), it is showed that the power of the wind increases with the cube of the wind speed. It is clear that wind speed has an important part to maximize the power generation. Wind turbine should be located at relatively windy site [8] and best roof shape. A few simulations using Computational Fluid Dynamics (CFD) has been undertaken to study wind flow close to building’s roof for applications of roof mounted wind turbine [9]. On the other hand, a wind tunnel can be used to conduct wind turbine performance test [10]. * Corresponding Author. Phone: +62-21-87914511 E-mail: dany.perwitasari@gmail.com http://dx.doi.org/10.14203/j.mev.2014.v5.45-50 D. P. Sari and K. P. Cho / Mechatronics, Electrical Power, and Vehicular Technology 05 (2014) 45-50 46 To maximize the wind turbine performance, three different roof shapes covering a tall building were simulated using CFD. In this study wind tunnel tests were carried out to validate the wind speed distribution and to determine turbulence flow. This model was tested in Seoul, South Korea. It rises about 64m from the ground. Each model was simulated and tested with scale of 1:150. The objective of this study is to find the best design of rooftop on a tall building which can maximize wind turbine performance. II. MODELLED PARAMETERS A mounted-wind turbine in a tall building is usually located in urban areas. This case study took place at Seoul, the capital city of South Korea. From the Korean Building Code (KBC) 2009 [11], Seoul is classified as terrain B which means urban and suburban areas with closely spaced residential or other buildings with height of 3.5m or so or scattered medium-rise buildings. A unique characteristic of wind is the variation of speed with height. The movement of the wind is controlled by the deviating force due to the earth’s rotation and the effects of friction of the surface of earth [12]. The wind velocity profile at certain area in atmospheric boundary layer could calculate using Power Law formula [12]. The urban terrain type specifies a mean wind speed profile with a power law exponent of 𝛼 = 0.22, exposure constant of Zb = 15 m and nominal height of the atmospheric boundary layer of Zg = 400 m. For specifying the optimum roof shape, the CFD commercial code Fluent 6.2.16 was used to simulate wind flow and to calculate wind speed above three different roof shapes above rectangular building shape whose edge height is 64m (high rise building ratio depth : width : height = 1:1:4). Wind turbine mounted building models are single tall building which were adopted and redesigned from mounted-wind turbine which currently exist (Fig. 1). This building tower was designed to increase wind speed to maximize wind turbine power. Using the wind turbine mounted building model in Figure 1, experimental results are shown as velocity coefficient. It is obtained from the velocity in a certain point (at horizontal coordinate x) divided by velocity approach from boundary layer in terrain B. 𝐶𝑣 = 𝑉(𝑥) 𝑉𝛼 (2) Wind turbine locations on Building A, Building B and Building C are in the top of the building in the height range between 50m to 70m from the ground. The average wind velocity is 10m/s so that the wind velocity approach 𝑉𝛼 for each model is normalized to 10m/s. III. CFD SIMULATION First of all, each building is analyzed using CFD. CFD is used for simulating wind flow between twin towers and to develop the models to find the best aerodynamic shape and the point for wind turbine to get maximum wind velocity. Building A, Building B and Building C are drawn using Gambit 2.3. (scale 1:150). This scale is used for similarity in wind tunnel test to validate the CFD analysis result. Each model uses boundary layer in terrain B with viscous K- epsilon and all of the models were successfully iterated. The k-𝜀 turbulence model solves the flow based on the assumption that the rate of production and dissipation of turbulent flows are near-balance in energy transfer. The k-𝜀 model is one of the most common turbulence models [13]. Figure 1. Modelled roof shapes, from left to right: Building A, Building B, Building C D. P. Sari and K. P. Cho / Mechatronics, Electrical Power, and Vehicular Technology 05 (2014) 45-50 47 Figure 2 shows simulation results using CDF. The left hand side figure indicates vertical position (y), the right hand side figure demonstrates wind speed contour and horizontal position (x) from top view, and the middle figure plots velocity coefficient. In the middle figure, horizontal axis denotes horizontal position (x), vertical axis indicates velocity coefficient and each curve has certain vertical position (y) as its parameter. The wind velocity in Building A increases in the center between the rounded towers. Wind velocity coefficient increases to 1.4 (Fig. 2(a)) or 40% more than wind velocity approach. Figure 2(a) shows the best horizontal position for turbine, with height 2m up to 5m above the roof between the rounded towers. Building B and Building C are analyzed using the same method as Building A. As to Building B and Building C, wind velocity was calculated at vertical location from 1m up to 5m above the roof surface (Fig. 2(b)(c)). Figure 2(d) compares velocity coefficient among Building A, Building B and Building C. It can be seen that Building B has the highest wind velocity coefficient in the value of 1.49. IV. WIND TUNNEL EXPERIMENT CFD simulations result has shown that maximum wind speed could be reached in Building B and Building C. Hence, only Building B and Building C was tested using wind tunnel. Wind tunnel experiment is use to validate the CFD analysis result. Wind tunnel experiment was held in CKP Wind Solutions, Gimje, South Korea. Detail of the wind tunnel size and component is shown in Figure 3. Wind tunnel test is also possible to calculate turbulence intensity which cannot be detected in CFD. The results are more accurate than CFD analysis and include the information for structural engineer. Turbulence intensity is important to detect the turbulent flow around the wind turbine. This information is important to prevent the blade flicker and the user’s safety. Figure 2. Location of measurement point, wind speed coefficient result and wind speed contour: (a) Building A, (b) Building B, (c) Building C, (d) Wind velocity coefficient comparison between Building A, Building B and Building C. The highest wind velocity coefficient is in Building B D. P. Sari and K. P. Cho / Mechatronics, Electrical Power, and Vehicular Technology 05 (2014) 45-50 48 Wind Tunnel Experiment also uses roughness block for terrain B to create boundary layer as the same as in the CFD analysis (Fig. 3). Wind velocities in wind tunnel experiment are calculated using cobra probes and pitot tube. Figure 4 shows measurement set up and experimental result. Wind velocity increases in the center of rounded towers along horizontal direction (x axis). In Building B, the highest wind velocity is at point 3 in vertical direction between rounded towers. This point corresponds to the height of 3m above the roof surface. The maximum wind velocity coefficient is 1.33 or increase of 30% from approach wind velocity. In Building C the maximum velocity coefficient increases 43% from wind velocity approach after passing rounded tower. Wind velocity is higher and more stable compared to that in building B. The highest wind velocity is located in point 2 and 3. Maximum wind coefficient is 1.43. Turbulence intensity is shown in table 1. In Building B it increases at point 1 and 2, and Figure 3. (a) Wind tunnel at CKP wind solutions, South Korea; (b) Roughness block of terrain B in wind tunnel experiment Table 1. Turbulence Intensity Data from Cobra Probes in Building B and Building C Building B Building C Position I a overall (%) Iu b (%) Iv c (%) Iw d (%) Position I a overall (%) Iu a (%) Iv b (%) Iw c (%) V inlet 12,488 14,005 11,920 11,256 V inlet 12,773 14,574 12,394 11,161 Point 01 21,872 21,872 21,683 21,967 Point 01 25,759 23,010 30,593 22,631 Point 02 23,673 24,811 24,052 22,157 Point 02 19,218 23,389 17,702 15,616 Point 03 12,299 15,427 11,446 9,360 Point 03 9,266 11,730 9,019 6,251 Point 04 10,024 13,342 8,555 7,086 Point 04 9,550 12,488 8,640 6,754 Point 05 9,550 12,394 8,697 6,896 Point 05 9,218 11,730 8,555 6,650 a I is The value of turbulence intensity; b Iu is The value of turbulence intensity from the longitudinal (horizontal) direction; c Iv is The value of turbulence intensity from the vertical direction; d Iw is The value of turbulence intensity from the lateral directions. D. P. Sari and K. P. Cho / Mechatronics, Electrical Power, and Vehicular Technology 05 (2014) 45-50 49 being normal at point 3. In Building C it increases at point 1 and 2 and normal at point 3. V. VALIDATION Figure 5 shows validation result between CFD analysis result and wind tunnel testing result. Figure 5 (a) shows the result of Building B while figure 5 (b) shows the result of Building C. Wind velocity increases in the center of rounded towers. As to Building B, wind velocity increases 30% in wind tunnel experiment and 50% in CFD analysis. One reason of wind velocity decrease in wind tunnel test is because of the roughness surface of the rounded tower in the model. Even though wind tunnel result has been proven to be representative of real world situation, it requires a correct model which accounts for the features of the atmosphere and exact scaling as well as model shape. Building C is a modified twin tower from Building B. Wind velocity increases at the center of rounded towers. The highest wind velocity is Figure 5. Validation result from CFD simulation and wind tunnel test: (a) Building B and (b) Building C Figure 4. Cobra probes test in wind tunnel windward surfaces in the middle of rounded towers and wind velocity coefficient results: (a) Building B, (b) Building C D. P. Sari and K. P. Cho / Mechatronics, Electrical Power, and Vehicular Technology 05 (2014) 45-50 50 in point 3 between rounded towers. In wind tunnel experiment, wind velocity increase 32 to 42% from wind velocity approach. Wind velocity increase 42% in CFD. Blockage effect in wind tunnel test because of swept area (A) is covered by cobra probes (blockage effect in Building C is 7.49%). Normally, blockage effect is not more than 5% [12]. This model shape is less performance than Building B. On the other hand, Building C’s turbulence intensity is more stable than Building B, which is safer and less vibration. VI. CONCLUSIONS Based on CFD simulation and wind tunnel test result on three roof shapes, this study evaluated wind speed, wind flow and turbulence intensity on the roof top of tall building. The most important results of this study can be summarized as follows: 1. Wind speed plays an important role to maximize performance of mounted-wind turbine in a tall building. Modifying the building roof top could increase wind speed ranging from 27 to 55%. 2. CFD simulation and wind tunnel test on roof top designs show that wind speed increases at the center of rounded towers. 3. The CFD simulation results show that the minimum performance is in Building A and the maximum performance is in Building B. 4. The wind tunnel test results in Building B and Building C indicate a good agreement with the CFD simulation results. 5. The wind tunnel test data also shows that Building C is the safer and less vibration building design, based on turbulence intensity result. ACKNOWLEDGEMENT We would like to thank The Indonesian Institute of Sciences (LIPI), Wonkwang University and CKP Wind Solutions for allowing this study to be undertaken and to the staff at each institute and university for taking the time to participate in this study. REFERENCES [1] Presidential Regulation (Peraturan Presiden) No.6/2011, “Rencana Aksi Nasional Penurunan Emisi Gas Rumah Kaca,” in National Standarization Agency of Indonesia, 2011. [2] Directorate General of New Renewable Energy and Energy Conservation, “Renewable Energy Development,“ in Ministry of Energy and Mineral Resources, 2012. [3] S. Deneen and B. Howard, “Buildings that Breathe: Green Construction is Coming of Age,” E Magazine January/February, 2007. [4] J. Ola, “The Spatial Diffusion of Green Building Technologies: The Case of Leadership in Energy and Environmental Design (LEED) in the United States,“ International Journal of Technologies Management and Sustainable Development, Vol.10, No.3, 2011, pp.251-266. [5] L. Ledo, et al., “Roof Mounting Site Analysis for Micro-Wind Turbines,“ Renewable Energy, Vol.36, 2011, pp.1379- 1391. [6] I. Abohela, et al., “Effect of Roof Shape, Wind Direction, Building Height and Urban Configuration on the Energy Yield and Positioning of Roof Mounted Wind Turbine,“ Renewable Energy, Vol.50, 2013, pp.1106-1118. [7] S. Sinisa, et al., “Urban Wind Energy,“ Earthscan, UK and USA, 2009. [8] Q.S. Li, et al., “Implementing Wind Turbines in a Tall Building for Power Generation: A Study of Wind Loads and Wind Speed Amplifications,“ Journal of Wind Engineering and Industrial Aerodynamics, Vol.116, 2013, 70-82. [9] W. Sara Louise, “Building Mounted Wind Turbines and Their Sustainability for the Urban Scale-a Review of Methods of Estimating Urban Wind Resource,“ Energy and Buildings, Vol.43, 2011, pp.1852-1862. [10] Pudji Irasari, “Experiment and Analysis of Car Alternator for Wind Turbine Application,” Journal of Mechatronics, Electrical Power and Vehicular Technology, Vol.02, pp.1-10, 2011. DOI: 10.14203/j.mev.2011.v2.1-10. [11] Korean Building Code-Stuctural, 2009. [12] N.M. Guirguis, et al., “Study of Wind Effects on Different Buildings of Pitched Roofs“, Desalination, Vol.209, 2007, pp.190-198. [13] L. Henry, “Wind Engineering: A Handbook for Structural Engineers,” Prentice Hall, eBook, 2008.