Optimal Positioning of Small Wind Turbines Into a Building Using On-Site Measurements and Computational Fluid Dynamic Simulation

Controlling the escalating risks posed by climate change requires an immediate decarbonization of the global energy sector. Global average temperature for August 2022 was 0.3 °C higher than from 1991 to 2020 [1]. However, there exist other risks associated with an economy reliant on fossil fuels. The finite supply of fossil fuels, the concentration of these resources in a small number of countries, and the volatility of the energy market are crucial factors that weigh the pros and cons of the various sources. The United States, Saudi Arabia, and Russia led the top three oil producers, with 21%, 13%, and 10%, respectively, and the top ten accounts 74% globally [2].

Energy transition is imperative to tackle climate change. This shift needs to be addressed in the energy trilemma of energy security, energy equity, and environmental sustainability. The industry must navigate these complex issues and find solutions that effectively balance all three dimensions [3]. On the other hand, the rise in fuel prices (especially natural gas) in Europe is an obvious illustration of how the fuel market can shift in a relatively short period of time with grave implications for national economies due to external factors such as the pandemic, war, and geopolitical issues [4]. For example, the Dutch TTF natural gas increased from 79 to 210 EUR/MWh in roughly three months [5]. More than ever, governments, scientists, and politicians recognize the urgency of advancing alternative energies. To expedite the energy transition towards carbon-free services and to gradually deploy renewable energies, wind and solar photovoltaics (PV) are essential building blocks. The growth rates of wind and PV power were considerably high in 2021, increasing by 14% and 23%, respectively. Wind and solar power must continue to grow at high compound growth rates of 20% annually through 2030, the same rate as the previous ten years, in order to keep global warming on track at 1.5 °C [6].

Building consumption represents approximately 30% of primary energy consumption globally [7]. In the United States building consumption accounts 40% of total energy, making it crucial to achieve carbon neutrality through energy efficiency and the use of renewable energy sources [8]. Energy efficiency can result in savings of 20%–40%. According to Millward-Hopkins et al. [9], small wind turbine (SWTs) installations are anticipated to support the shift to low-carbon energy and improve urban energy efficiency. Urban areas have high population densities, drawing attention to the need for specific energy solutions, particularly close to the demand area, decentralized generation, such as PV and SWTs systems [10].

The utilization of urban wind energy could make a significant contribution to the expansion of wind power. The lack of available land in urban areas, however, is thought to be a significant barrier to the installation of large machines. Integrating wind energy systems into urban areas is an alternative. The ability to use produced energy on-site, thereby avoiding transportation losses and producing extra economic benefits, is a major benefit of this distributed renewable technology. There is enough wind to produce considerable microgeneration from this energy source when set up properly, especially in the vicinity of tall buildings. A systematic analysis is required to guarantee the feasibility of projects in urban wind energy harnessing [11].

Innovative energy solutions are necessary to produce clean, safe, and environmentally friendly energy to tackle the energy deficit, environmental pollution, and other issues pertaining to sustainable development. Wind energy, both large- and small-scale, is an alternative that is expanding quickly worldwide [12]. In extremely densely populated urban areas, there is great potential for distributed electricity generation using urban wind [13]. SWTs offer great potential for use in domestic installations, on building roofs or in surrounding areas, or even integrated into buildings from the design stage [14]. Reduction in the cost of high voltage transmission lines and control devices is another benefit of using wind energy in cities due to its close proximity to demand [15].

According to Rezaeiha et al. [16], wind energy systems can be classified into three groups, such as turbines installed in close proximity to buildings, retrofitted in existing building, or buildings that have turbines integrated into their architectural design. SWTs can be positioned in high-rise buildings, railroads, roadways, and other urban locations [17]. Since the cities are heterogeneous places and aerodynamically irregular, estimating wind energy becomes extremely difficult. Obstacles such as trees, buildings, and other objects produce turbulence in the wind flow [12]. The obstacles extract momentum from the wind flow and significantly reduce energy capture by the SWTs [18].

Toja-Silva et al. [15] used computational fluid dynamics (CFD) simulations to present an analysis of multidirectional wind conditions. The authors examined a SWT located on a building's roof and its different behaviors. According to the flow fields, horizontal axis wind turbines (HAWTs) need to be positioned higher above the building's roof, whereas vertical axis wind turbines (VAWTs) perform better in turbulent impalements. To make the computational process easier for users with limited numerical method experience, Gagliano et al. [19] assessed the potential of wind energy in Italy using CFD simulation within a geographic information system (GIS). The best locations for the installation of SWTs in urban settings can be found using this methodology. According to Dhunny et al. [20], wind flow in complex terrain can be evaluated using CFD and validated using on-site measurements. The WindSim tool was utilized to create the wind map of Mauritius, and nine anemometric stations located across the island confirmed the accuracy of the map. The turbulence model, the discretization strategy, and the convergence criterion are emphasized in this study. Based on a thorough bibliometric review, Toparlar et al. [21] demonstrated the exponential trend in the use of this tool in publications pertaining to microclimate over the previous 20 years. Moreover, the tropical region lacked studies using CFD simulation. It is noteworthy that there are very few CFD studies conducted in Latin American countries, particularly in the Caribbean region. Significant conclusions about where to place SWTs on building roofs have been reached through CFD simulation. In 2018, Stathopoulos et al. [22] suggested placing HAWTs 1.4 m above building roofs and in the center of the roof. HAWT placement thresholds are 30% of the building height, while vertical axis wind turbine (VAWT) placement thresholds can be lower [23].

Arteaga-López et al. [24] provided a methodology for characterizing the urban wind potential in a Mexican university building, primarily considering the SWT's characteristics and the building's morphology. A 10 kW Bergey Excel SWT can generate enough energy for the campus' self-consumption or to light the public areas. The best configuration determined by the CFD simulations showed that the building under consideration can only accommodate five SWTs which are suitable to the building's annual energy demand. The CFD results were validated with the experimental results. Juan et al. [25] used CFD simulation to assess the effects of urban morphology on the utilization of urban wind energy for high-rise building arrangements. They adopted three-dimensional Reynolds-averaged Navier–Stokes (RANS) stable equations and verified their findings with wind tunnel measurements. The findings showed that turbulence and the beneficial effect of rounded corners increase with urban area density. To comprehend the relationship between the wind inside and outside the urban canopy layer (UCL). Ricci et al. [26] created a methodology for estimating urban wind within the UCL using on-site measurement and CFD simulation. They also pointed out the drawbacks of 3D Steady RANS simulations in two case studies. Table 1 lists the Scopus top publications which frameworks include CFD simulation in the resource assessment stage.

Computational fluid dynamics simulations can be effectively integrated into urban wind energy assessment techniques. In principle, these studies aim to characterize air flow through obstacles such as buildings and trees in a controlled volume. In an urban environment, the flow is considered turbulent. Earlier studies on building-mounted wind turbine installation in urban environments applying CFD analysis have shown results that indicate that SolidWorks Flow Simulation software is appropriate for this task [24].

Urban wind energy assessment is a complicated process due to the high turbulence and complex orography of cities. Since the project's viability is heavily dependent on the estimate of energy produced, a thorough evaluation of the resource is essential. Numerous approaches have been proposed, such as site selection, wind speed estimation, small wind turbine selection, annual energy production estimation, economic assessment, and environmental analysis [36,37]. The framework presented by Rezaeiha et al. [16], is interesting since it makes use of software that is freely accessible for calculating the quantity of suitable building's rooftops in The Netherlands according to their height. A methodology that included financial profitability and avoidance of greenhouse gas emissions was recently presented [37]. This methodology complemented the previous one. Conversely, to supplement current methodologies, [38] has suggested including an analysis of energy resilience. The vulnerability that has widely dispersed SWTs in cities vulnerable to atmospheric events (hurricanes and tropical storms) in tropical zones, forms the basis of the authors' contribution. The availability and dependability of SWT systems integrated into buildings during the project's life (CAPEX and OPEX), which is typically 20 years, are also impacted by other low-probability and high-impact events, which must be considered. Moreover, Vallejo-Díaz et al. [39] have presented a hybrid Strengths, Weaknesses, Opportunities, and Threats–Analytic Hierarchy Process (SWOT–AHP) analysis that combines qualitative and quantitative approach. The authors also presented a complementary analysis to bring a wider perspective in decisions making. The goal of the authors' analysis was to pinpoint the major variables that have shaped technology development thus far. Thus, non-technical analyses that support the growth and application of urban wind energy can be conducted. Another tool used to identify stakeholders and their roles in projects or management is Responsible, Accountable, Consult, and Informed (RACI) [40]. The RACI matrix can be applied to the energy sector to identify the responsible actors [41]. It is important to complement the methodologies with organizational tools for decision making, since these shed light on hidden factors. In every technology deployment, decision-makers must converge with clear and precise roles and assignments [42]. Recently, Vallejo-Díaz et al. [43] presented a RACI analysis applied to the stakeholders at the national level that influence urban wind energy in the Dominican Republic. A total of 59 interviews were carried out, and 28 public and private institutions were evaluated. According to the experts' judgment, the prominent institutions were the Ministry of Energy and Mines (A), electricity distribution companies (R), energy associations and universities (C), and educational and justice institutions (I).

This work's primary contribution is the technique it suggests, which combines several different approaches to provide a comprehensive analysis of the research on urban wind energy. The technique considers a seven-step framework that is thorough. In the resource analysis step, the impact of location on energy production is studied using CFD modeling and in situ measurements. The purpose of this research is to help determine the best location for SWTs on building roofs. A case study in the Dominican Republic, which has a tropical climate, uses the methodology to determine the best positioning of a SWT to maximize the power produced by urban wind kinetic energy, using on-site measurements and several CFD simulations. The structure of this research is as follows: Sec. 2 describes the proposed methodology approach. Section 3 outlines the case study. In Sec. 4, results and discussion are presented. Finally, in Sec. 5, the conclusions are given.

The proposed methodology adopts several steps from some existing ones that have been published separately. This framework proposes to consolidate all the contributions of existing methodologies to achieve an integrated analysis of urban wind energy study. The proposed methodology is presented in three phases in Fig. 1. The first phase is about the evaluation of urban wind energy. This contains the seven steps such as (1) site selection, (2) wind resource analysis including using on-site measurements and CFD simulation, (3) small wind turbine selection, (4) annual energy production estimation, (5) environmental evaluation, (6) resilience and economic analysis, and (7) system installation. The second phase is concerned with identifying the key factors and prioritizing them. This phase is composed of two main steps, which are the determination of the key factors through a SWOT analysis and the prioritization of these factors through paired comparison with the AHP analysis [44]. Finally, the third phase proposes integrating the RACI matrix into the methodology to identify key stakeholders in the urban wind energy sector in three steps such as interviews with experts, role diagnoses, and stakeholder mapping in the RACI Matrix. This research focuses on the first phase, especially on the wind energy assessment. The second and third phases have been well described in previously cited investigations [39,43].

The choice of urban location for incorporating the SWTs is critical. Wind speeds in cities are typically low due to the obstacles in the urban morphology [36]. However, there is a lot of space available above buildings in cities that can be used for harnessing wind energy [45]. GIS tools can be used to identify these areas [46] or even in the data provided by the local urban planning authority. According to Vallejo et al. [45], Santo Domingo has great potential in tall buildings, especially those 90 m high, because the wind flows faster, and thus, more energy is produced, ensuring that the levelized cost of energy is comparable to the cost of the distribution network. Therefore, taller buildings are the most attractive to install SWTs since the wind flows faster at higher altitudes.

Several methods for prospecting and studying urban wind have been reported in the literature. On-site measurement, CFD simulation, numerical climate prediction (NWP), wind tunnels, and analytical methods are the most common [11,29,47]. These methods take different approaches from morphological and meteorological scales [48]. Typically, the NWP is used to examine the region in a mesoscale spatial resolution, on-site measurement is focused on local resolution or neighborhood scales, and finally the CFD simulation to microscale and consider the building details. The NWP may overestimate the potential because it does not account for terrain roughness and other aspects of thermal conductivity [38,49]. The main goal is to determine the wind speed, direction, turbulence, and energy density in the studied area.

On the other hand, the wind tunnel can be useful in determining a wide range of variables that cannot be determined using CFD simulation [22,50], but this is beyond the scope of this study. Analytical methods, which can be used to predict wind speed in urban areas [51], are less common in the reported literature [38]. This study utilizes on-site measurements and CFD simulation to identify the optimal location for SWT on a building's rooftop. Each method is described in more detail below.

However, the accuracy of CFD results, associated costs, and user expertise remain the primary barriers to its popularity. Consequently, verification and evaluation of CFD simulation results are crucial. Typically, validation for CFD studies in real urban settings is conducted using field measurements and wind tunnel experiments [47,56,57].

Mittal et al. [47] emphasize that the accuracy of CFD simulations heavily relies on the choice of turbulence model. Researchers have reported on the capabilities and precision of different turbulence models for various analysis resolutions. Notably, significant insights have been gained through this approach, such as the recommendation to install SWTs approximately 1.4 m above building roofs. Furthermore, for HAWTs, it is recommended to place them in the middle of the roof and as high as possible in the windward direction [22]. In urban areas, wind speeds are generally lower compared to rural open areas. Consequently, it is essential to leverage points within buildings where flow accelerates to achieve a substantial increase in power density, potentially reaching three–eight times higher than the average power density [60].

In order to assess the feasibility of the project, it is crucial to define the project location and its surrounding area, as mentioned in Sec. 2.1. Utilizing CFD simulations, a comprehensive understanding of wind flow patterns within the urban complex can be achieved. This aids in predicting the behavior of wind streams and evaluating the performance of SWTs installed on the roofs of specific buildings. After selecting the site, a standard wind resource assessment is conducted to estimate the power output of SWTs. This assessment utilizes statistical datasets. The next step involves a (CFD)-based assessment, which consists of four stages: site assessment modeling, building-structure modeling, SWT modeling, and structure-SWT modeling. These stages involve creating a CFD model of the site, including the building and its surroundings, as well as a model of the building or structure specifically designed for CFD analysis. This analysis aids in determining the most suitable type of turbine to be installed.

After the SWTs are selected, a performance study utilizing CFD analysis should be carried out. The results of this comprehensive analysis, using SolidWorks Flow Simulation and applying finite volume analysis [61], will determine the technical viability of the project.

SWTs are emerging urban power generation technologies that are gradually being integrated into buildings, and peripheral environments and are even considered in civil structure architectural design [19,22]. Building-integrated wind turbine systems are classified as large or small, with small ones being more convenient and cost-effective in terms of building structural modification [28]. SWTs are machines that have the ability to extract kinetic energy from the wind flow and convert it into electricity via the aerodynamic forces induced on the blades [62]. Electricity generation is proportional to the cube of wind speed, swept area of turbine blades, and air density [63]. Figure 2 shows the primary SWTs designs, which are primarily HAWTs and VAWTs. The VAWTs are classified as Savonius, Darrieus, and H-rotor type [22,64]. Darrieus and Savonius work by the action of the lift force and the drag force, respectively. VAWTs are the most used in urban areas because they are omnidirectional, operate at a lower speed, and account for better three-dimensional aesthetics that can make them more popular whereas HAWTs must always be oriented with the rotor perpendicular to the wind direction, they are ideal for mountainous places that have a constant gust of wind [65,66]. Darrieusis more suitable for low- to medium-wind speed areas; however, the cross-sectional profile of blades is a key parameter to keep the efficiency in the design stage [67]. HAWTs are less commonly encountered in the literature. These machines are installed in larger areas with significant heights above ground level, in accordance with local policies and regulations. The most appropriate type is determined by wind speed, installation and operation flexibility, height limitation, and aesthetic integration into the urban built environment [56].

According to the International Electrotechnical Commission (IEC) standard [68], SWTs must have a capacity of 50 kW and a rotor-swept area up to 200 m². China, Germany, UK and United States have different classifications based on power, hub height, and swept area [16]. On the other hand, SWTs are classified into four categories based on rated power: micro (less than 10 kW), mini (10–100 kW), median (100–1000 kW) and large (greater than 1000 kW) [69].

Because VAWTs are less sensitive to changes in wind direction, they have a lower manufacturing cost because they lack a yaw mechanism, and as a result, fewer failures occur [70]. The HAWT and Darrieus are lift SWTs, and the Savonius is a drag SWT [71]. Tan et al. [72], presented a review of Savonius-Darrieus hybrid radial, counter-rotating and rotor wind turbines, particularly for low wind speed operation and noise reduction. They demonstrated that redesigning the chimney's height and the volute-shaped structure triples the power coefficient. They also discussed the benefits of piezoelectric generators and triboelectric nanogenerators for SWTs. Lipian et al. [73], on the other hand, presented an analysis using CFD simulation for the optimization of multi-rotor revolver-type SWTs, demonstrating that the relationship between the angle of attack and the velocity fields in the plane of rotation is relatively robust. Fatigue load characteristics, dynamics loads, and Taguchi Technique have been evaluated to the adequate laminate stacking sequence and lamina angles for the composite blade of Savonius VAWT [74].

Turbine positioning is critical because turbine performance can be hampered by low wind speed, high turbulence, or high noise levels produced by the turbine [71]. Several analyses using CFD simulation have been performed to study the optimal positioning of SWTs on building roofs of various shapes. Turbulence in the flow has been measured to extend up to 1.3 times the height of the building [75]. The authors believe that determining the optimal positioning of the SWTs in tropical environments while considering the morphology of the countries involved would lead to very interesting future research. To demonstrate that the turbines meet the levels of performance, durability, and other quality aspects and metrics used for comparison, SWTs must be certified by a competent standard. Some critical aspects include safety, acoustic level requirements, structural resistance, and so on. A list of certified SWTs and the best practices are presented by the American Wind Energy Association (AWEA) [76].

Naturally occurring catastrophes and other high-impact low-probability (HILP) occurrences put power systems at risk of both operational and infrastructural harm. To adopt adaptive measures in the face of such occurrences, resilience metrics are therefore required to measure the effect and vulnerability [83]. Reliability is the lens through which events with low-impact high-probability (LIHP) are examined. Relational dependability to HILP and resistance to LIHP are the most widely held presumptions [84]. According to Selga et al. [84], distributed renewable energies, such as SWTs integrated into buildings, can be a solution to prevent and mitigate the vulnerability and cascading failure of distribution networks when they are affected by climatological threats. The authors classified the threats into three groups, such as, catastrophic weather events, cyber-attacks, and sudden changes in the use of network infrastructure (the effect of the COVID-19 pandemic). As meteorological threats due to climate change increase, the intensity of these events significantly reduces the resilience indices of distributed energy systems such as urban wind energy. Engineers and academics must formulate proposals that improve the resilience of the system proportionally based on the potential threat to each of the generation technologies to achieve independence of energy systems and achieve their alternation [85].

Vallejo-Díaz et al. [38] proposed a preliminary approach to incorporate the economic implications of atmospheric events into the analysis of SWTs projects. This is crucial considering the Dominican Republic's susceptibility to such events, with 17 atmospheric occurrences affecting the country in the past two decades. Three scenarios were determined to evaluate the resilience of urban wind energy. In the first scenario, the SWTs system resists the predicted atmospheric event due to its wind speeds are below the operational threshold. This scenario applies to tropical storms or Category 1 hurricanes. In the second scenario, the SWTs system is disassembled in order to anticipate the disaster of the system due to the forecasted atmospheric events surpassing the wind turbines' functional limit. Consequently, there are costs associated with disassembly and reassembly. Finally, in the third scenario, the SWT system's functioning is harmed since it was not disassembled in the case of an atmospheric event. In this instance, restoring the SWTs system will cost more money. Over the last 50 years, tropical cyclones have caused approximately 2000 disasters, resulting in human deaths and economic losses, with an average of 43 deaths and US$78 million in damages per day [86].

The cost of wind power electricity is an important factor to consider, especially in small-scale systems that require proper evaluation for financing [87]. The total costs associated with the concentrated energy market include generation costs (power, capacity, and ancillary services), transmission and distribution costs, and marketing costs, as well as government subsidies in cases where end users do not receive price signals [88]. The authors presented a detailed comparison of different approaches to calculating the LCOE, outlining each method's strengths and weaknesses. The one that discounts energy and cash flow amounts over the course of the project is the most widely used. In addition, each country must consider other fiscal and economic incentives.

This final stage has to do with the implementation of the project. It is an engineering job. In this case, it is recommended to follow NREL's best practices, well presented by Fields et al. [18].

A case study was carried out at the Santo Domingo Institute of Technology (INTEC), to determine the potential of urban wind energy and the best SWT positioning on the building rooftop. INTEC is located at 18 deg29′16.81″N, 69 deg57′45.18″W. During 2017, two anemometers (Anem1-INTEC and Anem2-INTEC) were collecting data. The Anem1-INTEC recorded data from May to December in 2017, and Anem2-INTEC from June to December in the same year. Both anemometers were installed in a building 22.5 m height above the ground. A total of 208,881 and 152,816 records were taken from Anem1-INTEC and Anem2-INTEC, with an availability of 71% and 77%, respectively. Figure 3(a) shows the physical model of the INTEC campus and Fig. 3(b) the Eduardo Latorre (EL) building and sites of interest (EL building).

This study evaluates the INTEC campus with numerical simulation to determine the best positioning of the turbines in the geometry of the building's roof. The main idea is to determine the optimal positioning from the data records obtained in a field measurement campaign, which contains the wind flow pattern at the site of interest.

The findings from the case study conducted on the INTEC campus are presented in this section, with a focus on the assessment of urban wind energy for the production of electricity. Figure 4(a) shows the monthly average wind speed of Anem2-INTEC. The average wind speed at a height of 22.5 m was approximately 2.49 m/s during the periods under consideration. However, at a height of 34 m, the estimated average wind speed, using the Hellmann exponential law, wind profile law, was found to be 3.05 m/s. This approach allows for the extrapolation of wind data to higher elevations, providing valuable insights into wind patterns and velocities at different heights. A wind rose diagram is shown in Fig. 4(b), where the prevailing direction is 15 deg to the north, indicating that the wind is most frequently coming from the north-east.

Figure 5(a) depicts the intensity of wind speed over the course of a 24-h period. Records were used to construct the Weibull probability density function in accordance with the IEC standard; the probability density function is shown in Fig. 5(b). The Weibull distribution parameters for a wind speed of 3.05 m/s were determined to be 3.44 m/s and 2.00 for the shape factor and scale factor, respectively. Additionally, the turbulence intensity was measured to be 63%, 29%, and 8%, indicating moderate, high, or low levels, respectively. These findings can be attributed to the terrain orography as the measurements were taken in an urban environment. The presence of buildings and other structures in urban areas can cause variations in wind patterns, resulting in different turbulence intensities. Understanding these factors is crucial for accurately assessing wind conditions and optimizing the urban wind energy generation.

Table 2 provides a detailed description of the initial conditions chosen for the CFD simulations conducted in this study. The wall condition of 1,000,000 µm was determined based on the “Guide to Meteorological Instruments and Methods of Observation,” which classifies terrain roughness based on terrain class [89]. The INTEC campus is surrounded by class 7 terrain, characterized by large regular obstacles commonly found in suburban areas, resulting in a roughness height of 1 m.

The configuration of the grid and the refinement of the mesh play a crucial role in ensuring the precision of the CFD simulations. In this study, global mesh structures were selected, with a manual refinement definition of 249 cells per X, 100 cells per Y, and 100 cells per Z. Global mesh refinement surpasses the number of cells assigned in all axes through the highest level of auto refinement, aiming to guarantee precision in the areas of interest. This results in a total of 2,643,484 cells in the mesh, with 78,416 cells in contact with solids. In addition to mesh refinement, a control plane tool was utilized to increase the cell density in the areas of interest, further enhancing the accuracy of the simulations. This comprehensive approach to meshing and refinement ensures that the CFD simulations capture the intricacies of the wind flow and provide reliable results for analysis. A mesh independent test was conducted to ensure accurate and reliable results. This test involved varying the mesh size and analyzing the impact on the solution. By comparing the results obtained with different mesh sizes, the optimal mesh size that provides consistent results was determined. This process eliminates any potential errors or inconsistencies caused by an inadequate mesh, enhancing the overall accuracy of the simulation. A summary of the mesh test for the case is shown in Table 3, where it can be observed the relative error stabilization of about 9.8% with the mesh refinement.

For validation purposes, a series of modeling exercises were conducted. These exercises involved defining the parameters of the freestream and configuring the software settings. Subsequently, the software was run, and the results obtained from the modeling were compared with the measured parameters at a specific point. In this case, the freestream velocity and direction were determined based on the data reported in the NASA database at a height of 50 m for the INTEC location. These datasets were collected for the date specified in Table 4. The wind parameters measured during the same period were also recorded and are presented in Table 4. After conducting the modeling exercises, the measured datasets were compared with the modeling results, and the relative error was calculated. Table 4 provides a summary of the validation results. The average error is 3.5%, below 5% is considered acceptable.

Figure 6 depicts a top view of the velocity contour at 3 m/s and 15 deg, obtained through CFD simulations. The results illustrate the distribution of wind across the campus and the presence of high-speed zones around buildings. Additionally, are highlighted potential locations for the installation of SWTs on the EL building. It is important to note the significant variations in wind speed observed across different sites of interest. This observation emphasizes how the energy generation of a SWT can vary considerably depending on the specific installation site, even when located on the same roof.

Table 5 provides a summary of the analysis conducted on wind speed behavior, comparing measurements obtained from anemometers near the studied building with the wind behavior analyzed through CFD analysis. Figure 7 illustrates the differences in wind speed observed for the different positions on the roof, it should be noted that even when the horizontal positions are not significantly different (below 10 m) at higher freestream velocity the differences in wind speed at positions increase.

Table 6 provides an overview of the estimated annual power generation that can be achieved by a VAWT installed at each sensor position within the specified wind speed.

Figure 8 visually demonstrates the significant variations in energy generation that can occur depending on the specific installation site of a wind turbine, even when located on the same roof. This emphasizes the importance of careful site selection to optimize energy generation. Furthermore, Fig. 8 presents the estimated annual power and energy generation by a VAWT installed at each modeled positions on the roof of interest within the given speed ranges. Notably, the Anem2-INTEC position has the most potential, capable of generating 1030 kWh/year. It should be noted that this energy per turbine is small compared to what the building demands. According to the nine residential consumption profiles presented by Garabitos Lara et al. [90], residential consumption in one ranges between 242 and 1594 kWh/month. These consumptions depend on multiple behavioral factors of residents, such as socioeconomic level, environmental awareness, and other cultural aspects. However, to the extent that the roofs are higher above the ground, higher wind speeds are expected and therefore greater annual energy produced, which may have a better utilization factor from 80 m in height, as well as a more competitive LCOE [45].

Taking into account the estimated AEP and emission factor of 0.62 tCO₂e/MWh, it is projected that placing a SWT in the best location could potentially help avoid emitting 0.640 tCO_2e into the atmosphere [81]. This highlights the environmental benefits of selecting the optimal installation site for SWTs.

The study focuses on the optimal positioning of SWTs in urban buildings, intending to promote energy sufficiency in urban areas. A robust framework consisting of seven steps was developed: site selection, evaluating urban wind energy using computational fluid dynamics (CFD) simulation and on-site measurements, selecting an appropriate SWT, estimating the annual energy production (AEP), evaluating the environmental impact, resilience, and economic analysis, and finally, installing the system. The paper highlighted the use of CFD simulations in assessing wind energy potential and optimizing the positioning of SWTs on building roofs. Several studies demonstrated the effectiveness of CFD simulations in analyzing wind flow patterns and determining optimal SWTs placement thresholds.

By following this comprehensive framework, it is possible to assess the viability of wind energy utilization in urban areas. A case study was carried out at INTEC university, where on-site measurement campaigns and CFD simulation analysis were used. The average wind speed measured was about 2.49 m/s at 22.5 m height, which is considered low. At 34 m, the average wind speed was estimated to be 3.05 m/s. The predominant wind direction with the greatest intensity was registered at 15 deg respect to the north. The findings of the study showed that the most suitable place for installation had an estimated AEP of around 1030 kWh, which means that only one SWT could lead to a reduction of 640 kg of CO₂ per year.

The study also highlighted the significant impact of the building's geometry on wind flow and energy generation. It was observed how a substantial decrease in wind speed for different positions in the same roof occurs, and its impact on power generation. For instance, for a freestream speed of 4 m/s, a decrease in wind speed from 4.7 m/s to 2.5 m/s was determined in the southwest sensor position (X₄,Y₄) regarding the northeast (X₂,Y₂). With less kinetic energy available, SWTs generate less power, which represents a significant energy drop. This emphasizes the importance of considering wind patterns and selecting optimal sites to maximize energy generation in urban wind energy projects. The findings emphasize the potential of rooftop-mounted turbines in promoting energy sufficiency, reducing emissions, and addressing climate change.

The study also emphasized the significance of energy efficiency in buildings, as they account for a significant share of global energy consumption. SWTs were identified as a potential solution to support the shift towards low-carbon energy and improve urban energy efficiency. The proximity of urban areas to energy demand makes distributed generation technologies like SWTs and solar PV systems particularly suitable.

The first author expresses sincere gratitude for the financial support provided by FONDO NACIONAL DE INNOVACIÓN Y DESARROLLO CIENTÍFICO Y TECNOLÓGICO, Grant No. 2022-3C1-141, through the Ministry of Higher Education, Science and Technology (MESCYT) from the Dominican Republic. Additionally, special thanks are extended to Johemil Ramírez for the valuable assistance in proofreading and editing the manuscript.

There are no conflicts of interest.

The datasets generated and supporting the findings of this article are obtainable from the corresponding author upon reasonable request.

$c$ =

scale factor (m/s)

$k$ =

shape factor

$A$ =

swept area (m²)

$Cp$ =

power coefficient

$Vavg$ =

average wind speed (m/s)

$Vi$ =

desired velocity at height Z_i (m/s)

$V0$ =

known velocity at height Z₀ (m/s)

$Zi$ =

height 2 (m)

$Z0$ =

height 1 (m)

$p(Vavg)$ =

Weibull probability density function (%)

$α$ =

Hellmann exponent

$ρ$ =

air density (kg/m³)