Performance Analysis and Numerical Modeling of Mechanical and Electrical Components in a Rooftop Vertical-Axis Wind Turbine

Abstract

This study explores the integration of wind power generation into urban infrastructure via a rooftop vertical-axis wind turbine. A rigorous experimental framework was established by installing a small-scale turbine on an urban building for performance assessment under controlled conditions. Simulated environmental conditions were created using a pedestal fan and blower to evaluate mechanical interactions between the components and electrical output efficiency of the turbine. Extensive numerical modeling was conducted to analyze turbine performance, by computational fluid dynamics using ANSYS FLUENT. The results reveal that the turbine operates efficiently even at low to moderate wind speeds (0.5–6 m/s), demonstrating its feasibility for urban deployment. Performance tests indicated that, as the shaft rotational speed increased from 55 rpm to 115 rpm, the output voltage, current and power varied nonlinearly in the ranges of 3–11.9 V, 20–130 mA and 0.05–2.7 W, respectively. Vibration measurement at specified turbine locations revealed nonlinear variation in displacement, velocity, acceleration and frequency without fixed patterns. Good agreement was observed between the experimental and numerical results. The numerical model yielded interesting profiles related to velocity and turbulence distributions, apart from torque, mechanical power and electrical voltage. Important conclusions were drawn from the entire work.

1. Introduction

Fossil fuels have historically dominated electricity generation, accounting for approximately two-thirds of global electricity production [1]. As a result of continuous large-scale consumption, the depletion of fossil fuel reserves has accelerated, necessitating a transition to sustainable alternatives. In view of these conditions, the production of energy from naturally available renewable energy sources (RESs) has become a global priority. Over the past decade, the use of RESs has expanded significantly, contributing 5–25% of global power generation [2]. This means that the world is gradually shifting towards a complete transition from conventional sources to RESs [3]. Having little to no pollution susceptibility, RESs are also called ‘Green Energy’ [4]. Among RESs, solar, wind, nuclear, hydroelectric and biomass have emerged as the most prominent options [5]. Their environmental benefits, inexhaustibility and cost effectiveness provide distinct advantages over fossil fuels [4]. However, RES technologies remain relatively under-utilized for large-scale power distribution, being primarily localized in their application [6].

Wind energy, in particular, is valued for its continuous availability, indigenous nature, adaptability, convenience, etc. Despite these benefits, some inherent limitations hinder its widespread adoption. These include insufficient wind in urban and suburban areas, high costs of energy delivery and storage, noise, and social acceptance challenges [7,8].

Wind turbines (WTs) are robust devices that convert wind energy into electrical energy [9]. Offshore and onshore regions are mostly chosen for their installation due to their higher wind speeds, continuous availability and higher capacity factors, combined with the possibility of less interference with the landscape, people and wildlife. However, these installations often result in elevated transmission costs due to the distance between generation sites and demand centers [10]. Robust foundation designs and advanced electromechanical and structural systems are essential for offshore installations [11,12].

The integration of WTs into urban environments faces several design and operational challenges. Urban areas are characterized by slow and unsteady wind conditions, compounded by turbulence caused by the obstructions from high-rise buildings and dense infrastructure [13]. Moreover, urban WTs face barriers such as noise generation [12], visibility concerns [8] and impacts on biodiversity [14]. These factors complicate conventional wind mapping and power extraction for the deployment of traditional WTs [13].

Among the available options, rooftop WTs emerge as the most promising technology for direct power generation to address urban energy needs. However, the design of rooftop WTs involves overcoming significant concerns. The main design challenge is the optimization of power generation with limited wind speed coupled with the minimization of noise and vibration generated by rotating components [15], which are also critical for ensuring social acceptance and functional integration [8].

Knowledge Gaps in Rooftop Wind Turbines

Current research reveals a significant knowledge gap in the optimization and implementation of rooftop WTs, especially to address urban-specific constraints and enhance their technological, architectural and operational feasibility [14]. These include the following:

• Technological optimization of the performance of rooftop WTs, particularly in improving the conversion efficiency of wind energy to electrical energy, coupled with strategies to reduce noise and vibrations caused by rotating components [15];
• Tailored design for small-scale rooftop WTs suitable for limited urban spaces, while addressing site-specific challenges such as architectural integration [8] and noise considerations [14];
• Efficiency of WTs in low wind and unsteady speeds typical of urban settings;
• Urban, landscape and architectural integration of WTs to facilitate their effective implementation in densely populated areas.

Effectively addressing these challenges requires a comprehensive approach that combines technological innovation, architectural adaptability and urban planning considerations. This study seeks to bridge these gaps by introducing innovative design solutions and strategic frameworks for the efficient deployment of rooftop WTs in densely populated urban environments.

The modern smart city era necessitates energy-efficient buildings and structures, wherein the installation of rooftop WTs is an acceptable alternative to conventional solar panels from the viewpoints of cost efficiency, maintenance and longevity [16]. The legislative framework in many countries demands appropriate utilization of available renewable energy in the built environment, especially in densely populated urban regions including architecturally sensitive historic towns and cities [17,18]. However, commercially available WT models hardly serve the purpose of sufficient power generation and lack technical soundness from the viewpoints of energy efficiency, stability and serviceability [19]. The current research is thus novel and significant in mitigating the energy crisis in densely populated metropolitan cities and suburban areas [20,21].

2. Materials and Methods

This study develops and evaluates a small-scale rooftop WT tailored for densely populated urban environments. The research focuses on balancing energy production with environmental considerations, such as noise and vibration reduction and minimizing impacts on biodiversity. Through the development, installation and performance analysis of the WT, this study provides actionable insights into the advantages and disadvantages, critical challenges, and opportunities for power generation from RESs in these areas. To achieve these objectives, this study is organized into the following parts:

• Prototype development and installation of a small-scale rooftop WT in a densely populated urban area;
• Performance analysis in terms of energy production, noise and vibration analysis;
• Numerical modeling to quantify various aerodynamic and operational parameters of the WT;
• Data analysis and interpretation to assess the turbine’s viability and performance from both experimental and numerical analyses.

The research goals described above have been addressed in the current study by developing and installing a small-scale rooftop vertical-axis WT which operated at low to medium wind speed in an urbanized locality, yielding moderate electrical output power. The production of vibrations and other salient mechanical and electrical parameters have been studied in depth as well. Numerical modeling has also been carried out by using computational fluid dynamics for comparative studies and visualizing critical flow parameters.

The first phase of the current work focused on the design and installation of the WT prototype. An extensive literature survey was conducted to contextualize the research and identify critical challenges. The design process prioritized compact dimensions to capture spatial constraints on urban rooftops, along with aerodynamic efficiency tailored for low-wind-speed conditions. Following the design phase, the WT was fabricated and installed on a rooftop in an Indian densely populated area. The second phase involved detailed evaluation of the WT operational performance via field measurements to quantify its ability to convert urban wind energy to electrical power. Data collection was conducted using precision sensors and measurement devices, ensuring high accuracy and reliability. The aerodynamic performance of the WT was further investigated through numerical simulations using computational fluid dynamics via ANSYS FLUENT software (version 16.2). The simulations were used to validate experimental data and conduct further analysis. Comparison with past studies and commercial and cost efficiency aspects were also covered briefly. The methodology followed a structured sequence, as depicted in Figure 1.

3. Prototype Development and Installation

A small-scale vertical-axis rooftop WT has been fabricated and installed at the main administrative block of Elitte College of Engineering in Sodepur, Kolkata, India. The site is in the eastern part of India, with global coordinates of 22.7° N and 88.4° E and a height of 6.4 m above sea level [22]. The Hooghly River is situated about 5 km from the site, which is a major reason for the significant wind availability at the location. The WT has been installed on the south-west corner of the roof, at a height of approximately 21 m from the ground. This location was chosen due to its maximum average wind speed, as determined by an anemometer. The WT is mounted on a 0.45 m thick parapet wall that is 0.9 m above the roof surface. A brick pillar with a plan area of 1.2 m × 0.6 m was constructed to support the WT assembly (Figure 2).

The WT assembly is mounted and fixed on the top of the brick pillar. The installed WT consists of mechanical, electrical and civil infrastructure components, which are sequentially discussed herein (Figure 3).

3.1. Mechanical Components

The WT assembly is securely fixed to a brick pillar surface using four 30 mm diameter stainless-steel bolts, each embedded up to 500 mm into the pillar. The bolts firmly hold a table measuring 865 mm × 250 mm × 10 mm (Figure 3).

A flanged rotating tube (external diameter: 70 mm, flange diameter: 120 mm) is bolted to the table, supporting a 510 mm diameter rotating disc. This disc accommodates four WT blades, which are bolted at angles. The ball bearing arrangement between the flange and disc ensures frictionless rotational motion. To facilitate future performance improvements, the table includes a series of slots, allowing for potential enhancements such as a chain-sprocket or belt–pulley system to increase rotational speed.

A pair of stainless-steel channels is suspended from the base plate via four 16 mm diameter threaded studs. A detachable plate is bolted to these channels to hold the motor assembly. This plate is longitudinally adjustable to allow for modifications and realignment when necessary. The details of the turbine assembly are portrayed in Figure 4.

WT blades are subject to various forces, including aerodynamic, gravitational, centrifugal, gyroscopic and operational forces. To optimize torque production, the blades are designed to

• Maximize lift force while minimizing drag force;
• Ensure efficient momentum transfer from wind to the rotor;
• Maintain an optimal angle of attack for effective wind energy conversion [23,24].

This can be achieved by maximizing the lift force while minimizing the drag force. The current rooftop WT employs a straight-blade design with twisting at different sections using a trial-and-error technique. The maximum angular velocity of the rotor disc was determined by generating artificial wind in a laboratory setting using a blower. The assembly consists of four identical blades. Each blade is connected to the central tower shaft by a nut–bolt assembly and steel angles. The blades are tapered, with the horizontal chord length gradually decreasing from top to bottom. Figure 5 illustrates the blade assembly with the critical dimensions for future research, development and theoretical analysis.

3.2. Electrical Components

WT generators convert wind kinetic energy into electrical energy. The generated DC voltage is then routed into the inverter to produce AC voltage for general use. The inverter converts the direct current (DC) into an alternating current (AC), which is suitable for grid use. The rotational energy from WTs is utilized to produce a unidirectional DC electric current by the generator, following Faraday’s law of electromagnetic induction, according to Fleming’s right-hand rule [25]. In this case, the generator shaft is aligned and directly connected to the rotor shaft of the wind turbine to achieve optimal angular speed with minimal loss. The generator can produce moderate amounts of electricity for low and medium angular speeds (revolutions per minute, i.e., rpm), typically ranging from 50 to 200 rpm. The produced DC power is then allowed to pass through a three-phase switch and stored in a battery and can then be converted to AC voltage via the inverter (Figure 6).

The DC generator used in the WT of this study is a permanent magnet generator (PMG) designed for low- to medium-speed applications, typically in the range of 50–200 rpm. It directly converts wind energy into DC voltage, which is stored in the battery. A 12 V commercially manufactured battery produced from Exide Industries has been used to store the generated DC electricity. It features advanced electrodes for high discharge efficiency, robust non-woven gauntlets, a precision filling method and temperature-controlled recirculation cell generation for improved discharge characteristics and lifespan [26]. The DC voltage from the battery is fed to a UPS inverter, which includes features such as a resettable circuit breaker, auto-reset for overload conditions, intelligent microcontroller-based control, LED status indications and short circuit protection.

In the present work, the generator shaft is aligned in a manner to connect the WT shaft directly to capture the angular speed. The direct connection between the WT and the permanent magnet generator (PMG) minimizes mechanical losses to obtain maximum efficiency in power generation. The generator consists of individual components including a rotor, a stator, a yoke, armature windings, a pole shoe, brushes and a commutator. The generator is chosen in such a way that it can generate moderate electricity for a low to medium range of angular speed in the order of 50–200 rpm. A 12 V battery [26] is used to store the generated DC electricity. The used battery utilizes advanced electrodes, which results in higher discharge efficiency. The use of highly porous and robust non-woven Gauntlets, high-precision filling method and temperature-controlled two-shot recirculation cell generation in such a type of battery leads to significant improvements in discharge characteristics and design life. The DC voltage from the battery is first fed to the UPS inverter, which is equipped with a resettable circuit-breaker to ensure the safety of all connected equipment. It also consists of improved features like an auto-reset function for overload conditions, intelligent control design based on a microcontroller, LED status and fault indication arrangements, smart overload sensing and short circuit protection, a slide selection switch, etc. The current produced by the generator through the WT is first fed into the battery via a voltage/current digital display unit, while the battery is connected to the UPS inverter. The three-phase change-over switch was arranged in the system for DC, AC and neutral currents (Figure 7).

Due to low wind speed in urban regions producing a lower magnitude of angular velocity in the WT shaft, the transformer chosen is a step-up-type transformer, so as to increase the voltage before sending the generated electrical power to the inverter. In the circuit, it is located between the battery/inverter system and the AC load to regulate the voltage levels. Its purpose is to ensure voltage stability and compatibility with appliances or grid integration, thereby enhancing efficiency by reducing the transmission losses. Figure 6 only illustrates the core process of energy conversion from the wind turbine to DC storage to keep the diagram simple and focused on generation principles. Hence, the power conditioning components (such as transformers or protective devices) have been included in Figure 7 and not in Figure 6.

The electric circuitry of the power generation unit includes a battery and inverter connection (Figure 8a). A change-over switch is implemented to isolate the AC and DC sections of the circuit (Figure 8b).

4. Performance Analysis

For the performance analysis of the rooftop WT, rotational energy was generated by blowing wind by means of a pedestal fan and high-speed blower, and the generated electricity was measured. Simultaneously, vibrations produced in different components of the WT are measured as well [27]. While recording the displayed parameters, namely, the angular speed, current and voltage, readings were taken at intervals of 5 s for a period of 1 min, and the average values were calculated therefrom. A similar procedure was followed in an earlier study, where more details are available [27].

4.1. Rotor Speed

The variation in angular velocity in the WT rotor resulting from the blowing wind was studied. The angular velocity was measured by a digital tachometer with a precision of 0.1 rpm, while the wind speed was measured with the help of a digital anemometer having a precision of 0.01 m/s. Due to the monsoon season, the speed of natural wind was significantly high, which eliminated the requirement of generating artificial wind by blowers. The wind speed was measured in the X and Y directions (Figure 9a). These measurements were conducted using an anemometer and tachometer (Figure 9b). The plot of Average angular velocity N_av was plotted against the wind velocity components v_x and v_y (Figure 9c). As observed, the parameter N_av increased from 75 to 100 rpm as the magnitude wind velocity (v) increased in ranges of 0.4 m/s < v_x < 3.8 m/s and 0.5 m/s < v_y < 3.8 m/s. With ascending magnitudes of v components, the rotor angular speed increased, although no definite pattern of variation could be recognized. The plots of the resultant velocity v $\left[=\sqrt{{v_x}^2+{v_y}^2}\right]$ versus the produced average angular speed N_av show that the angular speed increased with wind speed, but no definite pattern of variation could be concluded (Figure 9d,e).

4.2. Power Generation

The power generated from the WT was analyzed. The tower shaft is allowed to rotate by directing blowing wind towards the blades using a pedestal fan and/or high-speed blower. Different angular velocities of the tower were achieved by varying the speeds of the pedestal fan and the blower. The angular speeds were varied in the range of 55 rpm < N_av < 115 rpm, and the DC voltage and current of the generated electrical energy were found to vary in the respective ranges of 3 volts < V_av < 11.9 volts and 20 mA < I_av < 130 mA (Figure 10). With ascending angular speed, the DC voltage was found to vary, following a nonlinear pattern (Figure 10a). The best-fit curve was found to be a cubic parabola having a regression coefficient of R² = 0.974. The maximum value was observed to be 1.8 volts, occurring at an angular speed of 105.5 rpm. The DC current, on the other hand, was found to increase with angular speed in a curvilinear manner (Figure 10b). The best-fit curve was observed to be exponential with an ascending slope (R² = 0.978). The above observations may be justified by the fact that the voltage and current are controlled by the generator’s electromagnetic field. Although theoretically proportional, the associated losses in the circuit have resulted in curvilinear variation [28]. The variation in average power (P_av) with angular speed (N_av), using the standard correlation (P_av = V_av × I_av), shows that the best-fit curve represents a power curve where R² = 0.96 (Figure 10c).

4.3. Vibration Analysis

Due to the rotation of the blades and the supporting disc, vibrations of various magnitudes are produced in different components of the WT. Such vibrations are likely to produce structural distress if not controlled. This necessitates appropriate analysis and quantification of the vibration pattern. In the current study, the vibrations are measured at 16 critical points (Figure 11a). By means of a digital vibration meter (Figure 11b), the peak values of displacement (δ), v_v, and acceleration (dv_v/dt) produced are measured along the X, Y (Figure 11a) and Z directions (vertically upward). The shaft angular speed is varied in the range of 45–180 rpm, and the critical parameters above are measured.

The variation in the peak values of δ, v_v and dv_v/dt with the average shaft angular speed N_av are presented below (Figure 12, Figure 13 and Figure 14).

The patterns of variation are nonlinear, although a general ascending trend is observed in most cases. For the given range of N_av, the parameters δ, v_v and dv_v/dt varied in the ranges of 0.8–46 mm, 4.5–48 mm, 1.4–45 mm, 34–395 mm/s, 10–390 mm/s, 13–399 mm/s, 2–29 m/s², 0.8–29 m/s² and 1.1–29 m/s², along the X, Y and Z directions, respectively. The manufacturer of the vibration meter provides a calibration chart which contains values of peak frequency (f) for any given peak magnitudes of displacement, v_r and dv_v/dt. Utilizing this calibration chart, the peak frequency relevant to the vibrations produced by the rotation of the WT was derived. The variation with N_av is presented below (Figure 15).

As observed, the pattern of variation is nonlinear and highly random. For the given range of angular speed (45–180 rpm), the magnitudes of peak frequencies (Hz) vary in the ranges of 17–46, 10–40 and 10–31, along the X, Y and Z directions, respectively. It is evidenced from the above observations that a general tendency of the measured parameters, namely, displacement, velocity and acceleration, exhibited an increasing trend in their relevant peak values with ascending magnitudes of the average angular speed, although no definite patterns could be concluded.

In the case of peak frequency, on the other hand, the pattern of variation was entirely random without any specific trends. Such vibrational features are possibly due to the material and geometric properties and complex interactions between connected components of the WT, together with mechanical, aerodynamic and electromagnetic forces induced. Similar observations were found in previous studies also [29,30].

5. Numerical Modeling

Since the prototype model is a drag-type vertical-axis WT, a numerical simulation was performed by means of commercially available ANSYS FLUENT version 16.2 software [31,32]. The details of the simulation technique adopted, and the output results are presented here.

5.1. Geometry and Meshing

In the simulation, the blade geometry is considered for different inlet wind velocities (v) and the angular speed (N_av) produced. The blade dimension has been simplified as a uniform value corresponding to its mid-section, to avoid the complexity of its varying dimensions (Figure 16).

The rotating zone is circular, with a diameter of 510 mm, while the stationary zone measures 7000 mm in length and 5000 mm in width (Figure 17).

Element sizes of 2 mm and 20 mm were chosen for the rotating and stationary zones, respectively. It is evident that the WT has an axisymmetric design; thus, the distribution of v on the four blades is symmetric as well (Figure 18).

5.2. Simulation Methodology

In the simulation, required zones were created corresponding to the turbine blades, rotating zone interface, stationary zone interface, and top and bottom walls. The inlet and outlet zones were also considered for the entry and exit of air flow. A transient double-precision parallel solver process was adopted for the simulation, whereby a K-epsilon model was considered for turbulence modeling [33]. Analysis was performed for the ranges of inlet v and resultant rotor angular speed of 2.5–4.5 m/s and 70–100 rpm, which is consistent with the available free stream air velocity on site. A hybrid initialization was conducted for the simulation, having time steps of 1000, with the size of a typical time step size being 0.01 s [34].

5.3. Numerical Results

A typical velocity distribution obtained from the software corresponds to the values of the inlet v and rotor speeds (N_av) of 4.5 m/s and 100 rpm (Figure 19).

For different values of v and N_av, the simulated distributions of velocity and turbulence are presented (Figure 20 and Figure 21).

The simulation also yielded computed output data pertaining to the generated force, torque, mechanical power and voltage for different values of v and N_av, as summarized in

Table 1. Computed output parameters (source: Authors’ elaboration).

Input Parameter			Computed Parameter			Test Parameter
Inlet Wind Velocity, v (m/s)	Rotor Speed, Nav (rpm)	Rotor Angular Velocity, ωav (rad/s)	Torque, T (Nm)	Mechanical Power, P (Nm/s)	DC Voltage, Vav (Volts)	DC Voltage, Vav (Volts)
						Measured Data	Best-Fit Cubic Polynomial
2.5	70	7.33	3.61	26.4613	4.43	6.5	7.2
3.0	76	7.99	5.54	44.2646	6.12	8.9	8.6
3.5	86	8.9	5.73	50.997	8.23	11.1	10.4
4.0	96	10.05	9.63	96.7815	10.01	11.2	11.6
4.5	100	10.47	13.55	141.8685	10.65	11.5	11.8

Notes: (1) Wind speed assumed from average values for points 1, 2, 3 and 4 in Figure 9. (2) Measured voltage derived from the relevant best-fit curve in Figure 10. (3) Computed voltage obtained from the assumed value of low-speed shaft stiffness, 0.97 [35].

. A comparison of the computed values with the relevant test data indicated fair agreement, as presented in Figure 22.

Regarding the plot of inlet v versus rotor speed (Figure 22a), both the curves are observed to be nonlinear, with an average deviation below 10%. In the case of the plots of rotor speed versus DC voltage produced (Figure 22b), the nature of both the curves is found to have similarity, wherein a stabilizing trend is observed as the inlet v is increased. The deviation of the simulated and measured voltages is below 20%, the computed values being on the lower side. Furthermore, comparison of the computed and measured power output has been included in Figure 22c, wherein the computed power has been estimated by using values of the electrical resistance of the WT circuit obtained from the measured DC voltage and current. As observed, the agreement between the computed and measured power output is acceptable, with an average deviation below 20%. A possible reason for such variation is the assumed simplification of the blade geometry. Also, the assumption of 2D conditions in the simulation against the 3D nature of the WT influenced the computed results [36].

5.4. Mesh Sensitivity Analysis

Prior to conducting the analysis illustrated above, a mesh sensitivity analysis was conducted as well to ensure the accuracy and reliability of numerical simulations by evaluating how different mesh resolutions affect the solution convergence and accuracy. To serve this purpose, different mesh resolutions were selected, namely, coarse mesh, fine mesh and extra-fine mesh, for convergence to the expected results for a given set of input parameters with an input velocity of 3.5 m/s, keeping all other values constant. The velocity distribution was compared with the help of the values of rotor speed, rotor angular velocity and torque achieved from the results of the simulations conducted with the three types of mesh resolution. The comparative study is illustrated in

Table 2. Mesh sensitivity analysis for an air input velocity of 3.5 m/s (source: Authors’ elaboration).

Type of Mesh Resolution	Mesh Count		Rotor Speed, Nav (rpm)	Rotor Angular Velocity, ωav (Rad/s)	Torque, T (Nm)
	Rotating Zone	Stationery Zone
Coarse Mesh *	22,690	56,000	85.67	8.87	9.58
Fine Mesh **	51,045	87,500	86.00	8.90	9.63
Extra-Fine Mesh #	90,746	155,560	86.05	8.90	9.64

Note: Respective element sizes for rotating and stationary zones: * 4 mm and 25 mm; ** 2 mm and 20 mm; ^# 1.5 mm and 15 mm.

.

The comparative study from Table 2 above reveals that the variations in the output results between the fine and extra-fine meshes are insignificant, compared to the computational time and efforts required. Hence, fine mesh was considered for this study.

Like the mesh independence test, various parameters of iteration were tested for the final set of simulations. Coarse-mesh mode uses a low-resolution mesh for faster initialization, achieving maximum computational speed. Large-time-step mode runs the initialization with a higher Courant–Friedrichs–Lewy (CFL) number, giving a moderate speed that is less than that from the large-time-step mode. Hybrid mode combines both but is less effective [34]. Reducing the number of iterations through better initialization speeds up CFD simulations, but when coarse-mesh initialization or large time steps are used, solver performance could be optimized without sacrificing accuracy. From this viewpoint, a hybrid initialization was conducted for the current simulation, having time steps of 1000, with the size of a typical time step being 0.01 s.

6. Comparative Studies

A few recent studies were conducted on rooftop wind turbines, among which Huang et al. [37] conducted analytical modeling on wind turbine location and energy evaluation based on wind data from various ports in China. A study from Mansour et al. [38] was based on the application of a data-based machine learning technique to analyze and design a small-scale WT. Various NACA profiles were chosen, and different output parameters including tip-ratio, blade forces, rotor torque, power coefficient, etc., were analyzed. The work of Wang et al. [39] focused on the dynamic analysis of vertical-axis rooftop WTs based on stochastic analysis and obtained important observations on induced vibrations in tall buildings. Jaszczur et al. [40] performed numerical modeling using computation fluid dynamics to study the performance of a rooftop wind turbine.

The power output obtained from the present study via experimental and numerical works was compared with those obtained from Jaszczur et al. [40] for various WTs at a rotor speed of 100 rpm. A comparative bar chart is depicted in Figure 23.

As observed, the power output for the present WT is significantly more than that for the simulated case studies by Jaszczur et al. [40], with the deviations varying from about 109% to 404%. Although such deviations were the result of different configurations of various WTs, apparently such results ensured enhanced power generation in the current rooftop WT.

The current study captures an in-depth experimental and numerical investigation on a real-world rooftop WT, including electrical and mechanical performance for practical applications, not conducted by earlier similar studies.

7. Commercial Feasibility and Cost–Benefit Analysis

Compared to the large-scale offshore and onshore WTs, the rooftop WT installed in the current study was found to be significantly cheaper considering the benefits obtained. The current WT was installed during the year of 2023 in Kolkata, India, and the approximate cost was INR 75k, including the material costs of the mechanical and electrical components and installation charges. Based on the experimental data, the system produces 3 V–11.9 V at 20 mA–130 mA, indicating low power output, thereby supplementing energy needs rather than replacing home power supply, by reducing dependence on grid electricity, leading to lower electricity bills. Other benefits include low maintenance due to fewer moving parts compared to horizontal-axis WTs and eco-friendliness due to zero carbon emissions, supporting urban sustainability goals [41]. The estimated payback period is 7–15 years, for low- to medium-wind-speed conditions [42,43].

8. Discussion

Although the measured output data from the developed small-scale WT reveal that it is not capable of generating electrical power at a large scale compared to a large-scale offshore WT, proper optimization of the electrical and mechanical components, together with installing such rooftop WTs at scale, is expected to serve the purpose of providing sufficient power generation for buildings [44,45]. Moreover, past investigations have revealed that a greater wind speed does not necessarily enhance the power efficiency of WTs due to inherent losses [27,46]. The application of artificial intelligence would improve the design optimization and performance of WTs [47]. The measurements in the current work were taken with precision instruments that were properly calibrated and cross-checked for accuracy. Thus, the measured parameters and their patterns of variation seem to be reliable. However, uncertainty in the annual variation in wind speed in urban regions would likely result in discrepancies in power output, as well as deviations in the optimal performance of mechanical, electrical, mechanical and structural components. This necessitates the incorporation of appropriate probability models, Monte Carlo-based numerical simulations, MCP (Measure–Correlate–Predict) approaches, etc. [48].

Recent studies have employed the finite element method (FEM) to analyze fluid–solid interactions in various engineering problems [49,50]. Similar analyses for wind turbine aerodynamics have also been conducted in a few recent contributions [51,52]. However, such analyses are more relevant to large-scale wind turbines.

Creating environmentally friendly and resilient cities, which is essential in the modern age, and sustainable urban power generation are becoming an integral part of urban development. With the rapid growth of urban areas, the demand for energy is increasing, making it imperative to move towards sustainable and renewable sources. Innovative strategies for sustainable urban power generation are being explored, of which wind power generation using building integrated wind turbine (BIWT) is an important one. A BIWT can be any type of wind turbine that needs to be selected based on the nature and availability of wind speed [53]. Mostly, it is seen that a vertical-axis wind turbine (VAWT) is a better choice in the urban landscape [54].

VAWTs can generate power from low to moderate wind speeds and also address potential noise and vibration issues. The visual impact of rooftop turbines ensures compliance with local regulations, and architects and urban planners can easily integrate turbines seamlessly into the urban landscape. VAWTs can also be integrated into urban buildings without compromising the structural capacity to support the additional weight and dynamic forces imposed by the wind turbines. VAWTs are designed to capture wind from all directions, making them well suited to turbulent urban wind conditions. The small footprint and the ability to be part of a hybrid power generation system with solar power make VAWTs a popular choice in urban areas.

Briefly, the utilities of the proposed VAWT over traditional technology are as follows:

• Smaller dimensions;
• Integration with vernacular architecture;
• Consideration for biodiversity.

The developed VAWT can contribute to conventional technology in several ways. By producing green power, it can contribute to reduction in the dependence on fossil fuels, which is associated with air pollution, greenhouse gas emissions and climate change. It can generate decentralized power in urban and remote areas, eliminating the need for large, centralized power plants and long-distance energy transmission, minimizing transmission losses and increasing energy resilience. The overall benefit of VAWTs in urban environments supports biodiversity. Proper site selection of a rooftop VAWT includes various factors, including considerations of bird migration routes, nesting areas, important breeding grounds, risk assessments, etc. A comprehensive environment impact assessment before rooftop VAWT installations to assess potential impacts on biodiversity is a necessity [55].

9. Limitations

The present work has several inherent limitations, which opens the scope of future research, as described below:

• Stainless-steel bearings, wheels and metal blades increased the dead weight of the WT, leading to a possible reduction in shaft speed to some extent. Hence, using lightweight material without adversely affecting structural stability would improve WT performance significantly.
• The inclusion of step-up gear, chain-sprocket or belt–pulley systems would enhance the rotational speed, although appropriate design techniques need to be applied to reduce the induced frictional resistance.
• A system comprising an array of rooftop WTs would be interesting from the viewpoint of enhancing power generation.

Thus, reducing the self-weight of the WT, along with decreasing the frictional resistance between various mechanical components, is expected to enhance the WT performance significantly.

10. Conclusions

In contrast to offshore and onshore regions, wind turbines are rarely installed in urban and suburban areas, mainly due to the lack of wind, both in terms of quality and quantity. From this viewpoint, the current work was conducted to develop and install a small-scale rooftop wind turbine in a densely populated location in Kolkata, India, and analyze its performance critically. The blade was designed by trial and error for optimum speed, while the electrical circuit consisted of a DC generator, battery, UPS inverter and other related equipment. The performance analysis examined the conversion of wind speed into rotational energy, power output and vibrations. In parallel, a numerical model based on computational fluid dynamics using ANSYS FLUENT was developed.

This study reveals that the rotational speed of the tower shaft increases with an increase in the magnitude of wind, without any definite pattern of variation. As the magnitude of wind velocity increased in the range of 0.64 < v < 5.4 m/s, the rotational speed of the shaft increased from 75 rpm to 100 rpm.

With ascending shaft speed in the range of 55–115 rpm, the voltage and current of the generated electrical power increased nonlinearly in the ranges of 3–11.9 volts and 20–130 mA, respectively. The patterns of variation in voltage and current were observed to exhibit cubic parabola and exponential trends, with descending and ascending slopes, respectively. The power output, on the other hand, increased from 0.05 W to 2.7 W following a power curve with an ascending slope.

From the vibration measurements, the magnitudes of peak vibrations in terms of displacement, velocity and acceleration were found to vary nonlinearly with shaft speed, having a general ascending trend without any definite patterns. The peak frequency, as derived from the manufacturer’s chart, exhibited a nonlinear and highly random variation with shaft speed.

The numerical simulation conducted via commercial software ANSYS version 16.2 yielded interesting profiles relevant to velocity and turbulence of wind. The generated torque, power and DC voltage were computed by the numerical simulation. Comparison of the measured and computed DC voltage indicated reasonable agreement.

This entire work documents a remarkable achievement in terms of developing efficient power generation technology in an urban landscape through an innovative rooftop WT, which is beneficial from the techno-economic viewpoint in terms of possible future commercialization and mass-scale application [21]. In the case of offshore and onshore WTs, the increased cost of supply and transmission to distant locations necessitates the adoption of other alternatives. Rooftop WTs would be a convenient substitute, provided the minimum wind speed is readily available. The present study area is in a densely populated urban area in the city of Kolkata, India, where the wind speed is low to moderate, typically in the range of 2.6–10 m/s [20]. The rooftop WT installed in the study area has been found to perform satisfactorily in terms of electricity generation, even at low wind speeds, without much impact on the architecture of the built environment [8].

11. Scope of Future Research

Recent research on domestic power generation includes rooftop WTs, which have been observed to produce a reasonable amount of electricity, provided that an optimal design technique is adopted [14,56]. The primary objective of this work was to conduct a detailed performance analysis of the developed WT via artificially simulated wind using a pedestal fan and blower, to vary the induced speed from low to high. The authors are currently involved in exploring the possibility of developing hybrid rooftop turbines comprising combined wind and solar energy sources. Consistent performance of such hybrid turbines is expected, for which stormy weather will generate sufficient wind energy. Even in the absence of such weather, bright sunlight will also result in the adequate functioning of turbines [57,58]; channeling the wind at roof level via a suitably designed wind tunnel is also expected to result in this [59,60].