© The Author(s), published by EDP Sciences, 2024

Nomenclature

ρ : Air density in (kg/m³)

ε : Emissivity of the panel

Γ: Gamma function

σ : Stefan Boltzmann constant (in W/m² K)

σ_std: Standard deviations

η_H: Hybrid Energy Efficiency

η_S: Solar Energy Efficiency

$\widehat{{\mathrm{fr}}_{\mathrm{j}}}$ : Estimated Frequency

${\dot{\mathrm{E}}}_{{\mathrm{x}}_{\mathrm{in},\mathrm{S}}}$ : Solar Input Exergy

${\dot{\mathrm{E}}}_{{\mathrm{x}}_{\mathrm{in},\mathrm{W}}}$ : Wind Input Exergy

${\dot{\mathrm{E}}}_{{\mathrm{x}}_{\mathrm{out},\mathrm{S}}}$ : Solar Output Exergy

${\dot{\mathrm{E}}}_{{\mathrm{x}}_{\mathrm{out},\mathrm{W}}}$ : Wind Output Exergy

$\dot{\mathrm{m}}$ : Mass flow rate

$A_s$ : Rotor Swept Area

A: Surface Area of the PV Array

AC: Alternating Current

AEDB: Alternative Energy Development Board

AEG: Annual Energy Generation

AEG_H: Hybrid Annual Energy Generation

AEG_S: Solar Annual Energy Generation

AEG_W: Wind Annual Energy Generation

BC: Black Carbon

c : Shape Factor

C_p: Power Coefficient

Cd–Te: Cadmium Telluride

CER_EN: Carbon Emission Reduction for Energy

CER_EX: Carbon Emission Reduction for Exergy

CF: Capacity Factor

CIGS: Copper Indium Gallium Selenide

CO₂: Carbon Dioxide

DFIG: Doubly fed Induction Generator

DHI: Direct Horizontal Irradiance

DMS: Degree Minute Second

DNI: Direct Normal Irradiance

E: Energy Delivered in kWh

e_Ki: Specific Input Kinetic Exergy

e_Ko: Specific Output Kinetic Exergy

E_t: Electricity Production of the year

EF_grid: Grid Emission Factor

EF_sys: System Emission Factor

EMJ: Empirical Method of Justus

EML: Empirical Method of Lysen

ENE: Energy

ENEN: Energo-Environmental

EPC: Engineering, Procurement, and Construction

Epf: Energy Pattern Factor

EPFM: Energy Pattern Factor Method

ESMAP: Energy Sector Management Assistance Program

ETB: Ethiopian birr

EXE: Exergy

EXEN: Exergo-Environmental

FF: Fill Factor

fr_j: Observed Frequency

FRM: Factor Rating Method

G: Irradiation on Tilted surface

GCER: Gross Carbon Emission Reduction

GCR: Ground Coverage Ratio

GEBA: Global Energy Balance Archive

GHG: Greenhouse Gas

GHI: Global Horizontal Irradiation

GIS: Geographic Information System

h_conv: Convective Heat Transfer Coefficient

h_rad: Radiative Heat Transfer Coefficient

i : Alternative Investments

I_sc: Short Circuit Current at STC

I_t: Investment & Expenditure for the year

IEC: International Electro-technical Commission

IPCC: Intergovernmental Panel on Climate Change

IPPs: Independent Power Producers

IRR: Internal Rate of Return

k : Scale Factor

LCOE: Levelized Cost of Electricity

M_t: Operational & Maintenance Expenditure for the year

MLM: Maximum Likelihood Method

mono-si: Monocrystalline Silicon

N : No. of Observations

n : Project Lifetime

NEPRA: National Electric Power Regulatory Authority

NOCT: Nominal Operating Cell Temperature

NPV: Net Present Value

NREL: National Renewable Energy Laboratory

O&M: Operation & Maintenance

P: Maximum Theoretical Power Output for Wind Turbine

P_e: Electrical Power Output of Wind Turbine a/c to Powell’s Model

P_nom: Nominal Power

P_p: Payback Period

P_S,r: Rated Solar Power

P_W,r: Rated Wind Power

poly-si: Polycrystalline Silicon

PR: Performance Ratio

PVS: Photovoltaic System

PVsyst: Photovoltaic System Software

PVWHS: Photovoltaic & Wind Hybrid System

Q : Heat Transfer

r : Discount rate

R ² : Coefficient of Determination

R_t: Net balance for the year

RMSE: Root Mean Square Error

ROI: Return on Investment

SAM: System Advisor Model

SAT: Single Axis Tracking

STC: Standard Test Conditions

t : Time period (in years)

T_a: Ambient Temperature

T_m: Module Temperature

T_S: Sun Temperature

T_sky: Effective Sky Temperature

U : Overall Heat Transfer Coefficient

v : Wind Speed (in m/s)

v_c: Cut-in wind speed (in m/s)

v_f: Furling Wind Speed (in m/s)

v_i: Inlet Wind Speed (in m/s)

v_o: Outlet Wind Speed (in m/s)

V_oc: Open Circuit Voltage at STC

v_r: Rated Wind Speed (in m/s)

WPD: Wind Power Density

WSA: Weighted Score Analysis

WTS: Wind Turbine System

1 Introduction

The energy sector of Pakistan is currently experiencing a series of systemic challenges that have far-reaching implications for both economic development and environmental sustainability [1]. In 2019–20, Pakistan’s annual power demand was recorded at 154,559 GWh/year, with projections indicating a substantial increase of 19.6% by 2024–25 [2, 3]. The nation’s existing power infrastructure is aging and unable to keep pace with the rising electricity demand, leading to frequent power outages and inefficiencies in energy distribution. Compounding these issues is the country’s overreliance on conventional energy sources, particularly imported fossil fuels, which are subject to price volatility and geopolitical risks. This reliance has resulted in high electricity generation costs and an increased financial burden on both the economy and consumers. Additionally, fossil fuel-based power generation contributes significantly to environmental degradation, with rising carbon emissions exacerbating climate change concerns in a country already vulnerable to extreme weather events and natural resource depletion. Therefore, urgent action is needed to address these challenges and transition to sustainable solutions [2, 3].

Renewable energy, especially solar and wind power, presents a compelling solution for Pakistan’s energy crisis. Unlike fossil fuels, renewable sources are abundant, sustainable, and have the potential to significantly lower greenhouse gas emissions. The geographic location of Pakistan is especially favorable for renewable energy development, with vast solar radiation across its southern regions [4] and considerable wind energy potential along the coastal areas of Sindh and Balochistan. The integration of these renewable sources not only aligns with global trends toward decarbonization but also provides an opportunity for Pakistan to reduce its dependence on fuel imports, enhance energy security, and stabilize long-term electricity costs [5].

While numerous studies have investigated various aspects of renewable energy resources in Pakistan, particularly focusing on solar photovoltaic and wind energy systems, there is a notable gap in the literature regarding a comprehensive evaluation of hybrid renewable energy systems in Pakistan. Existing research tends to concentrate on individual components such as solar or wind energy without extensively exploring the synergies and potential of hybrid systems.

This paper aims to bridge this gap by conducting a comprehensive fourfold analysis (energy, exergy, economic, and environmental) of solar photovoltaic systems, wind turbine systems, and solar photovoltaic and wind turbine hybrid systems across selected sites in Pakistan. The primary goal is to provide insights into the potential of these renewable energy systems to address the country’s energy challenges, with a focus on their economic feasibility and environmental impact. Through this assessment, the study contributes valuable information for shaping sustainable energy strategies in Pakistan.

2 Literature review

The literature review is organized into three subsections: solar Photovoltaic Systems (PVS), Wind Turbine Systems (WTS), and solar Photovoltaic and Wind Turbine Hybrid Systems (PVWHS). Numerous studies have scrutinized the economic feasibility of each of these systems independently. By categorizing the literature in this manner, our goal is to present a systematic and comprehensive overview of the economic analyses carried out for each power system.

2.1 Solar Photovoltaic Systems (PVS)

A study conducted in Saudi Arabia investigated the performance of a Photovoltaic (PV) plant [6]. The findings indicate that the optimal system configuration, achieving zero unmet load and excess electricity, involves a PV size of 2200 MW and an inverter size of 2200 MW, maintaining a nominal power (P_Nom) ratio of 1.00 between the PV array and the inverter. The research suggests a potential reduction in the inverter size to 1500 MW (P_Nom ratio of 1.47) if the net present value is prioritized. However, this adjustment could result in surplus electricity and higher CO₂ emissions. The study emphasizes the importance of calculating the performance ratio using a standardized method such as the International Electro-Technical Commission’s (IEC) standard 61724, which defines the PV plant’s final yield and reference yield [7]. Uwho et al. [8] concluded that solar panels yield maximum solar irradiance (1657 kWh/m²) at a latitude angle of 5°, outperforming a 30° tilt with a solar irradiance of 1367 kWh/m². This suggests that, for fixed structures, tilting the panels with respect to latitude proves to be more optimal in maximizing solar energy capture. Further, it is observed that adjusting the azimuth angle by ±15 degrees from the south allows for up to 98% of maximum insolation in the northern hemisphere, regardless of the climatic region [9].

According to Kumar and Kumar [10], market analysis in 2015 revealed that monocrystalline (mono-Si) panels dominated with a market share of 93%, while Cadmium Telluride (Cd–Te) panels held a 7% share. The study concluded that mono-Si and polycrystalline (poly-Si) solar cells demonstrated efficiencies of 25.6% and 20.8%, respectively. At the laboratory scale, thin-film technology achieved the highest conversion efficiency of 21% for Cd-Te and 20.5% for Copper Indium Gallium Selenide (CIGS) solar cells.

In a study focusing on Malaysia, Sreenath et al. [11] simulated a 5 MW land-based solar PV farm across five different sites using RET Screen software. Site-2 emerged as the optimal location, displaying an 80.52% Performance Ratio (PR), a Levelized Cost of Electricity (LCOE) of 0.102 $/kWh, and a Gross Carbon Emission Reduction (GCER) of 4291 tCO₂/year, surpassing site-3 and site-1. Yaghoubirad et al. [12] analyzed the performance of a 235 W PV panel across six states in the USA – Chicago, Sacramento, Phoenix, Portland, Amarillo, and Denver. The findings revealed that Phoenix exhibited the highest energy generation at 538.66 kWh, while Chicago boasted the highest energy efficiency at 16.59%. Yousuf et al. [3] analyzed the PV potential at five airports. The results indicate that all airports exhibit favorable performance ratios. Notably, Quetta Airport stands out as the optimal location according to the 7E assessment. It demonstrates a reference yield of 2752 kWh/kW, a final yield of 2420.8 kWh/kW, a capacity utilization factor of 27.63%, a LCOE at $0.031/kWh, 5730 tons of CO₂ avoided annually, and $488,826 per year in greenhouse gas revenue. These outcomes are achieved using thin film-based technology with single-axis tracking.

Wassie and Adaramola [13] estimated that a solar-electrified household could achieve substantial savings, amounting to ETB 1278 or US$ 23.07 per year, attributed to reduced energy costs and the avoidance of mobile charging expenses. Additionally, the study suggests that such solar-electrified households could contribute to environmental conservation by saving 2.72 kg of Black Carbon (BC) emissions and 107 kg of CO₂ emissions annually compared to non-electrified households. Another study by Imam and Al-Turki [14] provides a comprehensive review of both large-scale and residential-scale solar PV projects implemented or planned in the country. Examples include the Solar Village, the KAUST rooftop system, the KAPSARC ground-mounted system, and the King Abdulaziz International Airport project. The study utilizes the System Advisor Model (SAM) software to simulate the energy production of a PV system, revealing that the system produced 23,589 kWh in the first year, averaging 1965.72 kWh/month. The estimated lifetime energy production of the system is projected to be 31,528,392.5 kWh, assuming a 0.5% annual degradation of the PV modules. Despite a continuous two-axis tracking system demonstrating a 35.2% increase in annual energy production compared to fixed systems, it is deemed economically unviable due to negative net present value and the fact that the LCOE exceeds existing electricity tariff rates.

2.2 Wind Turbine Systems (WTS)

Wind power has seen significant growth and expansion in recent decades [15]. Numerous studies have assessed the techno-economic performance of wind turbine systems, particularly in various wind corridors across Pakistan, with a significant concentration in Sindh. While evaluating new sites, Yousuf et al. [16] identified that Sujawal has the maximum average annual speed of 7.3 m/s with an annual power density of 376 W/m². Based on the energy, exergy, and cost assessment, Sujawal, Umerkot, and Sanghar are identified as the most suitable locations for new wind farms. For Sujawal, with a lifespan of 20 years, the payback period is estimated to be approximately 4.66 years with GHG credit and 7.08 years without GHG credit. Shami et al. [17] reported that the country’s total usable wind energy capacity is approximately 132,000 MW, with three provinces (Khyber Pakhtunkhwa, Sindh, and Balochistan) contributing to over 110,000 MW.

In a study exploring wind resources in South Korea, Ali et al. [18] focused on three locations – Deokjeok-do, Baengnyeong-do, and Seo-San. The research concluded that the Doosan model WinDS 134/3000 wind turbine type is the most economically and technically feasible for Deokjeok-do and Baengnyeong-do. Specifically, Deokjeok-do could generate 6.37 MWh of power annually, while Baengnyeong-do could produce approximately 10 MWh. For Seo-San, the Hanjin HJWT 87/2000 wind turbine model demonstrated the potential to generate about 7 MWh of electricity annually.

Murthy and Rahi [19] highlighted the significance of various aspects of wind resource assessment. Their emphasis included the importance of preliminary analysis, providing a comprehensive overview of the local wind climate’s behavior on diurnal, monthly, seasonal, and annual scales based on available wind resources. They highlighted that wind data is crucial for wind resource assessment, and for estimating wind power potential, it must be available for at least one year. The study suggested that extensive data measures enhance accuracy, incorporating different models to determine wind speed distribution, directional distribution, frequency distribution, and annual energy yield. The outcomes of these resource assessments play a crucial role in determining the International Electrotechnical Commission (IEC) classes of the potential site and selecting the appropriate turbine model based on the IEC class to harness the site’s maximum potential.

Similarly, Shoaib et al. [20] conducted a study on the wind power generation capability for the site of Jhimpir using data from the Alternative Energy Development Board (AEDB). The findings indicated that the period from May through October exhibited higher windiness, resulting in elevated wind power density values during this timeframe. The Maximum Likelihood Method (MLM) approach recorded a peak wind power density of 745.06 W/m² between March and May, while a directional wind power density of 545.68 W/m² was noted from December to February. These insights contribute to understanding the temporal variations in wind power generation capability, aiding in effective planning and utilization of wind resources at the Jhimpir site.

2.3 Solar Photovoltaic and Wind Turbine Hybrid Systems (PVWHS)

Combining solar and wind energy into a hybrid renewable energy system can be achieved through various strategies that enhance energy production, efficiency, and reliability [21]. One simple method is co-located installations, where solar panels and wind turbines are installed at the same site. This setup feeds power into a shared electrical grid, offering a more consistent energy supply, especially in areas where solar and wind resources complement each other, such as sunny days with little wind and windy nights or cloudy days [22].

Another approach involves integrated controllers. These advanced systems optimize the performance of both solar and wind resources by balancing power output, diverting excess energy from one source to either charge batteries or meet immediate consumption needs [23]. In microgrid systems, typically used in remote or isolated areas, solar and wind energy can operate independently from the central grid. These setups often incorporate energy storage systems, allowing surplus energy from either source to be stored and used later [24]. Power electronics further improve integration by using devices that adjust voltage and frequency in real time, ensuring a steady power supply [25]. Additionally, optimization algorithms can analyze weather conditions, demand, and storage capacity to determine the best solar wind energy mix for a given location [26]. Also, some hybrid systems utilize demand response mechanisms, adjusting solar and wind energy contributions based on real-time consumption patterns to better match supply with demand [27].

Tazay et al. [28] conducted a modeling and evaluation of a grid-tied hybrid PV/wind power generation system in the Gabel El-Zeit region of Egypt. Their findings showed that the system generated 1509.85 GWh of electricity annually, with the PV station contributing 118.15 GWh (7.83%) and the wind farm producing 1391.7 GWh (92.17%). This highlights the substantial role that wind energy plays in the system’s overall output.

Taghavifar and Zomorodian [29] explored a micro-hybrid solar/wind energy system designed for buildings, with the ability to sell excess electricity back to the grid. Two scenarios were studied: one with an inflation rate of 10% and a wind speed of 6.8 m/s. The results showed that the system’s Net Present Cost (NPC) was $49,022, with a Renewable Fraction (RF) of 85.5% and a Cost Of Energy (COE) of $0.0024. When combined with the grid, the NPC increased to $224,430, the RF dropped to 63.6%, and the COE rose to $0.0272. The study concluded that while a simple system is suitable for low inflation rates, higher rates necessitate additional equipment, and increasing wind turbine capacity reduces both the NPC and COE as wind speeds rise.

Alharthi et al. [30] evaluated a grid-connected hybrid renewable energy system in four cities in Saudi Arabia, focusing on economic and technical performance. The analysis, conducted using HOMER, considered a hypothetical community load of 15 MWh/day. Simulation results indicated that Yanbu City achieved the lowest LCOE at 0.03655 $/kWh, followed by Hafar Albatin, Sharurah, and Riyadh. Yanbu City also exhibited the highest mean annual wind power density at 833.78 W/m². Sharurah showed the highest solar potential, with daily averages of Global Horizontal Irradiance (GHI) and Direct Normal Irradiance (DNI) at 6681.62 Wh/m² and 6206.91 Wh/m², respectively. These findings provide valuable insights into the economic viability and technical feasibility of hybrid renewable energy systems in different cities in Saudi Arabia.

Aydin et al. [31] analyzed western Turkey. GIS methodology and mathematical tools like Fuzzy Set Theory were employed to identify feasible sites for Hybrid Wind-Solar Power Systems. The study highlighted one of the challenges in renewable standalone sources, emphasizing the discontinuous power generation resulting from climatic variability. This recognition of the intermittency of power generation underscores the need for effective strategies to address and mitigate the impact of climatic variations on the reliability and stability of renewable energy systems.

3 Material and methods

This study employs a comprehensive 4E analysis encompassing solar Photovoltaic Systems (PVS), Wind Turbine Systems (WTS), and solar Photovoltaic and Wind Turbine Hybrid Systems (PVWHS) across four sites in Sindh, Pakistan, as illustrated in Figure 1. The analysis unfolds in multiple phases, initiating with an energy analysis to ascertain the optimal mix of solar and wind technologies for efficient energy generation. Subsequently, exergy analysis is conducted to assess exergy output and efficiencies. Finally, the system is analyzed based on economic and environmental parameters, evaluating the feasibility of Greenhouse Gas (GHG) mitigation across all three systems.

Figure 1 Sites locations for Sindh (https://globalwindatlas.info/en/area/Pakistan).

To identify the most promising sites, a Weighted Score Analysis (WSA) utilizing the Factor Rating Method (FRM) is employed. The selection of these sites is motivated by the presence of substantial wind resources and the availability of topographic data. The detailed geographical information for these sites is available in Table 1.

The Global Horizontal Irradiation (GHI) data is retrieved from Meteonorm while the wind data is gathered from the Energy Sector Management Assistance Program of the World Bank [32]. Meteonorm combines its sources from the Intergovernmental Panel on Climate Change (IPCC), Global Energy Balance Archive (GEBA), ERA5/T, and others to get accurate results for typical years and time series [33]. The data used in this study were obtained from the “Renewable Energy Resource Mapping and Geospatial Planning–Pakistan” project, supported by the Energy Sector Management Assistance Program (ESMAP) of the World Bank. The dataset includes measurements of wind speed at 80 m height, wind direction, turbulence intensity, temperature, pressure, and relative humidity.

3.1 System definition

The sizing of the PV array is conducted through the PVsyst software, utilizing its “System” feature to accommodate varying system capacities. The P_Nom ratio of 1.05 is chosen to minimize losses and optimize power output. The mounting structures are fixed with an optimal tilt angle of 25° across all sites. This angle was selected based on the latitude of the study sites, which range from approximately 24°N to 25.8°N. Therefore, the tilt angle closely approximates the latitude of the sites, which is a standard practice for optimizing solar energy capture.

To assess the performance of different PV panels, the initial phase involves simulating the first nine panels from various Tier-1 PV manufacturers, covering mono-Si, poly-Si, and thin-film (Cd–Te) technologies. To ensure consistency in the analysis, the study selects specific panels for mono-Si, poly-Si, and thin-film technologies, each with a capacity of 390 Wp. Detailed specifications for these chosen modules are provided in Table 2. This analysis serves as a benchmark for the subsequent shortlisting of panels in the study. For a 50 MW solar farm, 19 zones are allotted, as shown in Figure 2. For all panels, a TC500KH inverter is employed.

Figure 2 Area covered by 50 MW solar farm.

For the wind farm layout design, Google Earth Pro is employed to determine the geo-positioning of each wind turbine and extract their Degree Minute Second (DMS) coordinates, fundamental for the subsequent analysis. The hypothetical interspacing between wind turbines is set at 5D and 7D, as illustrated in Figure 3. The wind farm layout consists of seventeen wind turbines, simulating a hypothetical 51 MW wind farm.

Figure 3 Area covered by 51 MW wind farm.

For the wind farm simulation, the Acciona Nordex AW-125/3000 wind turbine is chosen. This turbine model is widely recognized in Pakistan, as indicated by various Wind Independent Power Producers (IPPs) NEPRA reports, including those from Shaheen Energy and Moro Power. Comprehensive details about this turbine model can be found in Table 3.

3.2 Energy assessment (1E) of solar Photovoltaic Systems (PVS)

The energy analysis of solar photovoltaic systems (PVS) is comprehensively carried out using the PVsyst software. The software is utilized to define the orientation and sizing of the system. Subsequently, detailed losses from the inverter, sourced from Javed Solar, are incorporated into the specific section for losses within PVsyst. To account for shading-related losses, a dedicated analysis is conducted in the near shading section. The flowchart of the analysis is illustrated in Figure 4.

Figure 4 Flowchart for the energy assessment of PVS.

3.2.1 Performance ratio

The Performance Ratio (PR) is ass measure of the actual energy output from a system compared to the potential output if the system operated at its Standard Test Condition (STC) efficiency all the time. Mathematically,$$\mathrm{PR}=\frac{E}{G\times P_{\mathrm{nom}}},$$(1)where E is energy delivered in kWh, G is Irradiation on the tilted surface in kWh/m², and P_nom is the nominal power at STC in kWp. The ratio accounts for various losses including optical losses, array losses, and system losses. Importantly, the PR is independent of the PV module’s efficiency. This means that both amorphous silicon modules and high-efficiency crystalline modules can achieve similar PR values.

3.2.2 Energy efficiency

Energy efficiency is the ratio of the electrical energy output to solar energy input. The energy efficiency is given as:$$\begin{array}{c}{\eta }_{\mathrm{S}}=\frac{{\mathrm{AEG}}_{\mathrm{S}}}{\mathrm{GHI}\times {\mathrm{A}}_{\mathrm{PV}}}\end{array},$$(2)where, GHI is the yearly global horizontal irradiance of the site and A_PV is the surface area of PV panel(s).

3.3 Energy assessment (1E) of Wind Turbine Systems (WTS)

In the Jupyter environment using Python, the energy analysis is conducted for the Wind Turbine System (WTS). The analysis employs the PyWake library, a Python package designed for wake effect analysis and the evaluation of Annual Energy Generation (AEG). This library implements various models to assess the impact on power generation from each wind turbine in the wind farm [34].

Specifically, the wake model utilized in this study is the Bastankhah wake deficit model [35], along with the weighted sum superposition model. Superposition models are crucial for computing the effective wind speed, taking into account local wind speed and deficits originating from multiple sources. As noted by [36], the Bastankhah wake deficit model performs well when coupled with the weighted sum, providing a robust estimate for wake effects.

To facilitate wind turbine placement, the Pyproj library is employed to convert Degree Minute Second (DMS) coordinates into the Cartesian system. This process ensures accurate positioning of wind turbines within the wind farm [37]. The sequential flowchart is visualized in Figure 5.

Figure 5 Flowchart for the energy assessment of WTS.

3.3.1 Mean wind power density

Wind Power Density (WPD) quantifies the amount of wind energy available at a specific location. It is commonly expressed in Watts per square meter (W/m²) and is subject to variation based on factors such as wind speed, air density, and altitude. The mean wind power density for sites can be determined using equation (3).$$\begin{array}{c}\mathrm{WPD}=\frac{P}{A_s}=\frac{1}{2}\rho c^3\mathrm{\Gamma }\left(1+\frac{3}{k}\right),\end{array}$$(3)where ρ is air density in (kg/m³), c is the shape factor that defines the width of the distribution, k is the scale factor that defines the range of data distribution, and Γ is the gamma function that can be expressed as,$$\begin{array}{c}\mathrm{\Gamma }\left(X\right)={\int }_0^{\mathrm{\infty }} t^{x-1}{e}^{-t}\mathrm{d}t\end{array}.$$(4)

3.3.2 Weibull probability density function

The Weibull Distribution (as seen in Eq. (3)) is a widely used probability distribution for assessing the best fit of available wind data, as indicated by [16]. This statistical model proves valuable in characterizing wind speed distributions and is commonly employed in the analysis of wind energy resources.$$\begin{array}{c}f\left(\nu \right)=\left(\frac{k}{c}\right){\left(\frac{\nu }{c}\right)}^{k-1}\exp \left[-{\left(\frac{\nu }{c}\right)}^k\right]\end{array}.$$(5)

In this study, an evaluation of five methods was conducted for calculating the values of k and c, namely, the Least Square Method, the Maximum Likelihood Method, the Empirical Method of Lysen, the Empirical Method of Justus, and the Energy Pattern Factor Method. The formulations for these methods are outlined in Table 4. The evaluation of the different methods was conducted based on two key metrics: Root Mean Square Error (RMSE) and the Coefficient of Determination (R ²).

The Root Mean Square Error (RMSE) serves as a measure of the average discrepancy between predicted and observed values, quantifying the dispersion of these discrepancies or residuals. A smaller RMSE value signifies a model with more accurate predictions, while a larger RMSE indicates a higher frequency of prediction errors. The expression of RMSE is given in equation (6) as follows,$$\begin{array}{c}\mathrm{RMSE}=\sqrt{\left(\frac{1}{N}\right)\sum _{j=1}^N{\left(f{r}_j-\widehat{f{r}_j}\right)}^2,}\end{array}$$(6)where $f{r}_j$ represents the observed frequency, while $\widehat{f{r}_j}$ stands for the estimated frequency.

The coefficient of determination (R ²) measures the extent to which changes in the dependent variable can be predicted from changes in the independent variable(s). It is a ratio that interprets the proportion of variability in the outcome that can be explained by the model. In simpler terms, R ² indicates how well the model aligns with the observed outcomes, providing insight into the goodness of fit of the model. The formula for R ² is expressed in equation (7) as follows,$$\begin{array}{c}{R}^2=1-\frac{\sum _{j=1}^N{\left(f{r}_j-\widehat{f{r}_j}\right)}^2}{\sum _{j=1}^N{\left(f{r}_j-\overline{\mathrm{fr}}\right)}^2}\end{array}.$$(7)

3.3.3 Power curve

Powell’s model provides a set of equations that represent the electrical power output of a wind turbine and is expressed as [38],$$\begin{array}{c}{P}_e=\{\begin{array}{c}0\hspace{1em}\left(v<v_c\right)\\ a+b{u}^k\hspace{1em}\left(v_c\le v\le v_r\right)\\ \begin{array}{c}{P}_{W,r}\hspace{1em}\left(v_r\le v\le v_f\right)\\ 0\hspace{1em}\left(v>v_f\right)\end{array}\end{array}\end{array},$$(8)where $a=\frac{P_{W,r} v_e^k}{v_c^k-v_r^k}$ and $b=\frac{P_{W,r} }{v_r^k-v_c^k}$. Equation (8), P_W,r denotes the rated power in kilowatts (kW), while v_c, v_r, and v_f represent the cut-in, rated, and furling speeds in meters per second (m/s), respectively.

3.3.4 Power coefficient and capacity factor

The power coefficient, commonly denoted as C_P, serves as a measure of how efficiently a turbine converts the kinetic energy in the wind to mechanical energy. It can be calculated using equation (9) as,$$\begin{array}{c}{C}_P=\frac{{\mathrm{AEG}}_{\mathrm{W}}}{\frac{1}{2}\rho A_s{\nu }^3}\end{array},$$(9)where ρ is the density of the air in kg/m³, A is the rotor swept area of the wind turbine in m², v is the speed of the wind in m/s, and Annual Energy Generation (AEG).

The capacity factor assesses the actual energy output relative to the theoretical maximum output if the system consistently operates at full capacity. It is expressed as a percentage and can be calculated as in equation (10) $$\begin{array}{c}\mathrm{CF}=\frac{\mathrm{AEG}/8760}{P_{W,r}}\end{array}.$$(10)

3.4 Energy assessment (1E) of solar photovoltaic and wind turbine hybrid systems (PVWHS)

While the specific formulations for PVS and WTS have been presented individually, equation (11) can be utilized to calculate the capacity factor for PVWHS.$$\begin{array}{c}{\mathrm{CF}}_H=\frac{{\mathrm{AEG}}_{\mathrm{W}}+{\mathrm{AEG}}_{\mathrm{S}}}{P_{\mathrm{W},\mathrm{r}}+P_{\mathrm{S},\mathrm{r}}}\end{array},$$(11)where AEG_S and AEG_W represents the annual energy generation for the standalone Solar Powered System (PVS) and Wind Turbine Systems (WTS) respectively. The summation of both denotes the annual energy generation for the solar Photovoltaic and Wind turbine Hybrid Systems (PVWHS), which is provided as:$$\begin{array}{c}{\mathrm{AEG}}_{\mathrm{H}}={\mathrm{AEG}}_{\mathrm{W}}+{\mathrm{AEG}}_{\mathrm{S}}\end{array}.$$(12)

The energy efficiency of PVWHS is given as,$$\begin{array}{c}{\eta }_H=\frac{{\mathrm{AEG}}_{\mathrm{W}}+{\mathrm{AEG}}_{\mathrm{S}}}{\frac{1}{2}\rho A_s{\mathrm{\nu }}^3+\mathrm{GHI}\times A_{ }}\end{array}.$$(13)

3.5 Exergy assessment (2E) of Solar Photovoltaic Systems (PVS)

Exergy is commonly referred to as “useful work potential”, serving as a fundamental concept in thermodynamics and aiding in assessing the viability of the system. Exergy primarily depends on factors such as the amount of global horizontal irradiation and the ambient temperature, which directly influences the module temperature.

The input exergy necessitates several factors, including the surface area of the PV array (A), solar irradiation intensity, Sun Temperature (T_s), and ambient temperature (T_a).$$\begin{array}{c}{\dot{E}}_{x_{\mathrm{in},S}}=\mathrm{AG}\left[1-\frac{4}{3}\left(\frac{T_a}{T_s}\right)+\frac{1}{3}{\left(\frac{T_a}{T_s}\right)}^4\right]\end{array}.$$(14)

The two primary components of output exergy ${\dot{E}}_{x_{\mathrm{out},S}}$ are thermal exergy and electrical exergy and are evaluated as:$$\begin{array}{c}{E}_{x_{\mathrm{thermal}}}=Q\left[1-\frac{T_a}{T_m}\right]\end{array},$$(15) $$\begin{array}{c}{E}_{x_{\mathrm{electrical}}}=V_{\mathrm{oc}}\times I_{\mathrm{sc}}\times \mathrm{FF}\end{array}.$$(16)

Initially, the module temperature can be determined using values such as NOCT, ambient temperature (T_a), and solar irradiance intensity (G).$$\begin{array}{c}{T}_{\mathrm{m}}=T_{\mathrm{a}}+\left(\mathrm{NOCT}-20\right)\cdot \frac{G}{800}\end{array}.$$(17)

The heat transfer to the surroundings is calculated as,$$\begin{array}{c}Q=\mathrm{UA}\left(T_{\mathrm{m}}-T_{\mathrm{a}}\right)\end{array},$$(18)where U is the overall heat transfer coefficient and is evaluated as,$$\begin{array}{c}U=h_{\mathrm{conv}}+h_{\mathrm{rad}}\end{array}.$$(19)

The radiative heat transfer coefficient (h_rad) is the function of the module temperature (T_m) and the effective sky temperature (T_sky),$$\begin{array}{c}{h}_{\mathrm{rad}}=\epsilon \sigma \left(T_{\mathrm{sky}}+T_{\mathrm{m}}\right)\left(T_{\mathrm{sky}}^2+T_{\mathrm{m}}^2\right)\end{array}.$$(20)

The effective sky temperature is dependent only on the ambient temperature (T_a) the relation can be given as$$\begin{array}{c}{T}_{\mathrm{sky}}=T_{\mathrm{a}}-6\end{array}.$$(21)

The convective heat transfer coefficient (h_conv) for solar systems is dependent on the wind speed (V_w)$$\begin{array}{c}{h}_{\mathrm{conv}}=2.8+3V_w\end{array}.$$(22)

3.6 Exergy assessment (2E) of Wind Turbine Systems (WTS)

The net exergy input rate to the system is the total amount of input exergy to the system can be given by,$$\begin{array}{c}{\dot{E}}_{x_{\mathrm{in},W}}=\dot{m}\cdot \left[\left(e_{\mathrm{Ki}}-e_{\mathrm{Ko}}\right)\right]\end{array},$$(23)where $\dot{m}$ is the mass flow rate which can be given as:$$\begin{array}{c}\dot{m}=\frac{2}{3}\rho A_s{\nu }_i\end{array}.$$(24)

The output exergy of the Wind Turbine Systems (WTS) is determined by the summation of kinetic, potential, physical, and chemical exergy. However, the following assumptions have been considered in this study:

• Elevation changes do not affect properties other than wind speed.

• Any phase changes are considered insignificant.

• Air density remains constant based on site conditions.

• Only kinetic exergy is taken into account.

• The manufacturer’s thrust coefficient curve is utilized for all sites.

The specific kinetic exergy has two parameters, input, and output exergies,$$\begin{array}{c}{e}_{\mathrm{ki}}=\frac{1}{2}{v_i}^2\end{array},$$(25) $$\begin{array}{c}{e}_{\mathrm{ko}}=\frac{1}{18}{v_o}^2\end{array},$$(26)where v_i is the inlet wind speed (m/s).

3.7 Exergy assessment (1E) of solar Photovoltaic and Wind turbine Hybrid Systems (PVWHS)

While the specific formulations for PVS and WTS have been presented individually, a combination can be utilized to calculate the exergy for PVWHS.

3.8 Economic assessment (3E)

The economic feasibility of the project is assessed by analyzing key parameters including Net Present Value (NPV), Payback Period, LCOE, and Return on Investment (ROI).

NPV can be determined by having an expected cash flow for the tth year (Rt), the discount rate that could be earned in alternative investments (r), electricity production of the year (Et), and project lifetime (n) [39]. Given the initial investment (I_o),$$\begin{array}{c}\mathrm{NPV}=\sum _{t=1}^n \frac{{\mathrm{CF}}_t}{(1+r{)}^t}-I_o\end{array}.$$(27)

LCOE is the cost of energy generation per unit of power generation, which can be found out if investment and expenditures for the year (It), operational and maintenance expenditures for the year (Mt), electricity production of the year (Et), discount rate (r) and project lifetime (n) are known.$$\begin{array}{c}\mathrm{LCOE}=\frac{\sum _{t=0}^n \frac{I_t+M_t}{(1+r{)}^t}}{\sum _{t=0}^n \frac{E_t}{(1+r{)}^t}}\end{array}.$$(28)

Return on Investment (ROI) is also a useful indicator because it enables investors to evaluate the effectiveness of their investment in a project. It is vital to note that ROI does not account for the duration of the investment, or the potential dangers involved. As a result, while making investment decisions, it is critical to use ROI along with other financial measures and to evaluate the broader context.$$\begin{array}{c}\mathrm{ROI}=\frac{\mathrm{Net} \mathrm{benefit} \mathrm{at} \mathrm{the} \mathrm{end} \mathrm{of} \mathrm{lifetime}}{\mathrm{Total} \mathrm{Investment}}\end{array}.$$(29)

Another economic parameter is the internal rate of return. The Internal Rate of Return (IRR) is the discount rate ‘r’ that balances out the costs and benefits of an investment. In other words, the rate would make the Net Present Value (NPV) equal to zero.$$\begin{array}{c}\sum \left[\frac{\mathrm{Cost}}{{\left(1+r\right)}^n}\right]=\sum \left[\frac{\mathrm{Benefit}}{{\left(1+r\right)}^n}\right]\end{array}.$$(30)

The discounted payback period is determined by discounting the project’s cash flows with an appropriate discount rate and then comparing the cumulative discounted cash flows to the initial investment. The calculation of the discounted payback period is performed using the formula provided in equation (31).$$\sum _{n=1}^{\mathrm{DPP}}\frac{{\mathrm{CIF}}_n}{{\left(1+r\right)}^n}-C_{\mathrm{in}}=0.$$(31)where CIF is the cash inflow and is assumed to be the same throughout the life cycle of the system, n is the system life considered as 20 years and r is the discount rate taken as 22%.

The cost breakdown structure in capital investment, Operation Maintenance (O&M) cost, and electricity tariffs, are sourced from NEPRA IPPs Reports.

For PVS, economic data is sourced from Siachen Energy, referencing a 100 MW Solar Power project initiated by Siachen Energy Ltd. in Taluka Mirpur Sakro, District Thatta, Sindh. The cost breakdown structure is detailed in Table 5. Although the analysis was conducted on fixed-tilt structures with monofacial modules, it’s noted that the referenced project utilizes Single-Axis Tracking (SAT) structures and bifacial modules, requiring adjustments for our analysis. A comparison between monofacial and bifacial modules is sourced from Kumbaroğlu et al. [40], and a comparison between fixed and single-axis tracking structures is referenced from García et al. [41], as illustrated in Table 6. For WTS, economic parameters are extracted from Shaheen Energy as seen in Table 7.

3.9 Environmental assessment (4E)

As concerns about climate change and the sustainability of conventional energy sources intensify, solar PV and wind technologies emerge as promising and environmentally friendly alternatives to traditional fossil fuel-based energy generation. Table 8 compares the Gross Carbon Emission Reduction (GCER) of all energy sources provided by the Intergovernmental Panel on Climate Change (IPCC) [42]. Various phases can affect these emissions, from system initiation and construction to transportation, operation, maintenance, and recycling. Additionally, according to a technical paper by Brander et al. [43], the grid emission factor for Pakistan is around 451 gCO₂/kWh. The environmental analysis in this study is subdivided into Energo-Environmental (ENEN) and Exergo-Environmental (EXEN) assessments.

In the energy-environmental analysis, the assessment includes both energy production and carbon dioxide emissions during operation. As outlined by Sreenath et al. [11], the gross carbon emission reduction can be calculated using parameters such as annual energy generation (AEG), system emission factor (EF_sys), and grid emission factor (EF_grid).$$\begin{array}{c}{\mathrm{CER}}_{\mathrm{EN}}=\mathrm{AEG}\times \left(\mathrm{E}{\mathrm{F}}_{\mathrm{grid}}-\mathrm{E}{\mathrm{F}}_{\mathrm{sys}}\right)\end{array}.$$(32)

The emission factor ranges for PVS and WPS spanning from approximately 18 to 180 gCO₂/kWh and 7 to 56 gCO₂/kWh, respectively, are reported by Schlömer et al. [42] and Yaghoubirad et al. [12].

The calculation methodology of exergo-environmental identical to that used in the ENEN portion focuses on exergy output instead of energy production for evaluation. It is a valuable tool for assessing the environmental implications of various processes throughout the entire generation chain up to the point of consumption.

The gross carbon emission reduction can be calculated as:$$\begin{array}{c}{\mathrm{CER}}_{\mathrm{EX}}=\mathrm{Output} \mathrm{Exergy}\times \left(\mathrm{E}{\mathrm{F}}_{\mathrm{grid}}-\mathrm{E}{\mathrm{F}}_{\mathrm{sys}}\right)\end{array}.$$(33)

3.10 Factor Rating Method

Factor Rating Method (FRM) was employed to select the optimal PVWHS system. The FRM involved initially defining the criteria and assigning weights to each factor based on its perceived importance. Subsequently, the performance of the configuration was evaluated against these criteria, taking into account the assigned weights. This evaluation process served as the basis for comparison and decision-making. By systematically varying the weights and analyzing the outcomes, the sensitivity of the results to changes in priority emphasis was assessed.

The identified key factors included annual energy generation, the area covered by each PVWHS, and the LCOE. These factors were carefully chosen to comprehensively capture the various dimensions of system performance, economic feasibility, and environmental impact.

4 Results and discussion

4.1 Energy assessment (1E)

Global horizontal irradiance data for the selected four sites is obtained using Meteonorm, ranging from 1825.5 to 1848.7 kWh/m², with Sanghar and Tando-Ghulam Ali exhibiting greater solar potential, as shown in Table 9.

The performance of three types of PV technologies, namely mono-Si, poly-Si, and Cd–Te, is simulated in PVsyst software. Nine PV modules are selected from different companies, including First Solar, Canadian Solar, BYD Solar, JA Solar, Longi Solar, and Trina Solar. Among these, four panels are mono-Si, four are poly-Si, and one is Cd–Te. After benchmarking the Performance Ratio (PR), it was determined that Longi Solar mono-Si panels have the highest PR of about 80.68%, followed by CSI Solar poly-Si panels with a PR of 80.17%, and First Solar Cd–Te panels with a PR of 79.63%, as depicted in Figure 6. Although there is minimal difference in PR among the selected modules, a PR exceeding 80% is generally considered favorable for performance (Sreenath et al. [11]). Given the negligible variance in PR, PV modules rated at 390 W for all three types are chosen for a justified analysis, as shown in Table 2.

Figure 6 Performance ratio comparison of 9 PV modules.

Tando-Ghulam Ali and Sanghar exhibit higher energy generation potential for a 50 MW PVS, ranging from 79,909 to 81,210 MWh/year and 79,656 to 80,995 MWh/year, respectively. Following closely are Umerkot and Sujawal, with energy generation estimates ranging from 79,471 to 80,786 MWh/year and 79,454 to 80,733 MWh/year, respectively.

Cd–Te panels, despite exhibiting the lowest efficiency levels ranging from 13.58% to 13.94%, demonstrate a higher energy production capability compared to other panel types. Monocrystalline and polycrystalline panels display higher efficiencies, ranging from 16.61% to 17.13% and from 16.6% to 17.11%, respectively. The lower efficiency of Cd–Te panels can be attributed to their larger surface area requirement, necessitating more land area. However, Cd–Te panels achieve a higher capacity factor due to their ability to generate more energy.

The site with the highest potential for a 51 MW WTS is Tando Ghulam Ali, displaying an annual energy generation of 138,902 MWh/yr, a CF of 31.09%, and an energy efficiency of 35.57%. Following closely is Sanghar, with an annual energy generation of 131,181 MWh/yr, Sujawal with 129,242 MWh/yr, and Umerkot with 119,246 MWh/yr, considering wake losses.

The wake losses for the 51 MW WTS in Tando Ghulam Ali are 7.07%, while Sanghar, Sujawal, and Umerkot experience wake losses of 8.90%, 9.02%, and 6.71%, respectively. Throughout the year, wind direction predominantly remains southwest for all sites. Figure 7 shows a wake map from the Bastankhah Wake Deficit Model with a weighted sum.

Figure 7 Wake map from Bastankhah wake deficit model with weighted sum.

Regarding the Weibull Distribution analysis, the least square method emerges as the most accurate among the five methods for all sites, boasting an R-squared ranging from 0.999836 to 0.99969 and RMSE ranging from 0.0051 to 0.0117. However, the other methods also demonstrate proficiency in estimating the shape and scale factors, as depicted in Figure 8.

Figure 8 Wind rose and weibull distribution curves for the selected sites.

The PVWHS comprises six distinct combinations of WTS and PVS systems, each evaluated through a weighted score analysis. Tando Ghulam Ali emerges with the highest annual energy generation of 90,863 MWh/yr, a capacity factor of 21.17%, and an energy efficiency of 16.90%, followed by Sanghar, Sujawal, and Umerkot. Further details for all systems are illustrated in Figure 9.

Figure 9 Annual energy generation, capacity factor, and energy efficiency.

4.2 Exergy assessment (2E)

The total exergy output and exergy efficiencies across various panel technologies and potential sites were analyzed. Tando Ghulam Ali exhibits the highest exergy output, ranging from 125,062 to 137,753 MWh/yr, followed by Sanghar with 124,891 to 137,641 MWh/yr, Umerkot with 124,026 to 136,580 MWh/yr, and Sujawal with 123,573 to 135,981 MWh/yr. However, it is noteworthy that, similar to the energy analysis, thin-film panels demonstrate lower efficiencies compared to mono-Si and poly-Si panels. Additionally, exergy efficiencies remain consistent across different sites, approximately 7.5% for mono-Si, 7.48% for poly-Si, and 6.58% for Cd–Te.

Regarding WTS, the output exergies align with their annual energy generation, but the exergy efficiencies vary across. For the optimal PVWHS configuration, the output exergies are 136,097 MWh/yr and 135,009 MWh/yr for Tando Ghulam Ali and Sanghar, respectively, and 133,462 MWh/yr and 131,367 MWh/yr for Sujawal and Umerkot, respectively. Additional insights from the exergy analysis are depicted in Figure 10.

Figure 10 Annual output exergies.

4.3 Economics outcome (3E)

The thin-film PVS systems exhibit the lowest LCOE among all configurations, ranging from 0.1144 to 0.1150 $/kWh, with the lowest recorded for Tando Ghulam Ali. Correspondingly, their NPV varies from 89.64 to 90.39 million $. Conversely, WTS systems demonstrate the highest NPV values, ranging from 87.57 to 122.36 million $, albeit with higher LCOE values. For the PVWHS setups, LCOE values range from 0.1235 to 0.1294 $/kWh, while NPV values range from 89.04 to 96.35 million $. These LCOE computations consider a 22% discount rate. Further insights can be seen in Figure 11.

Figure 11 Economic evaluation of PVS, WTS, and PVWHS.

Figure 12 illustrates the cumulative impact of discount rates and initial investments on LCOE. Higher discount rates, coupled with larger initial investments, significantly affect the LCOE, often rendering projects unprofitable over their lifetime. This occurs because the increased discount rate accelerates the devaluation of future cash flows, intensifying the pressure to recover the larger initial cost. Consequently, the LCOE may increase substantially. Conversely, when faced with a higher discount rate and a lower initial investment, the effect on the LCOE is more subdued. While the heightened discount rate still accelerates the devaluation of future cash flows, the lower initial investment mitigates the need to recover capital. As a result, the increase in the LCOE tends to be more moderate. Thus, the initial investment plays a crucial role in determining financial performance despite higher discount rates.

Figure 12 Effect of interest rate on LCOE.

4.4 Environmental outcome (4E)

Figure 13 demonstrates the environmental outcomes for PVS, WTS, and PVWHS. For PVS, the median value of carbon emissions is 42 gCO₂/kWh, while for WTS, it is 11 gCO₂/kWh. The gross carbon reduction based on energy metrics reveals Tando Ghulam Ali as the best site, with emission reductions ranging from 36,027 to 36,613 tCO₂/yr. Following suit are Sanghar, Umerkot, and Sujawal, aligning with their annual energy generation trends. Similarly, for WTS, Tando Ghulam Ali boasts a GCER of 67,370 tCO₂/yr, while for PVWHS, it stands at 41,854 tCO₂/yr.

Figure 13 Environmental outcomes for PVS, WTS, and PVWHS.

Given that exergy outputs exceed annual energy generation, the EXE gross carbon reductions surpass ENE. However, this does not alter the trend seen in total exergy output. Tando Ghulam Ali maintains its lead in gross carbon reduction, ranging from 56,384 to 62,105 tCO₂/Year. The WTS segment mirrors the energo-environmental findings, while for PVWHS, it is 62,248 tCO₂/yr.

To determine the optimal mix for the PVWHS system, a weighted score analysis is conducted using the Factor Rating Method (FRM). Three key factors serve as the basis for this analysis: annual energy generation, the area covered by each PVWHS, and LCOE of each system. The analysis involves adjusting the weights assigned to these factors across four cases. Upon analysis, it is concluded that the combination of 40 MW PVS and 9 MW WTS emerges as the best mix, with weight varying from 18% to 25% across the different cases as seen in Figure 14.

Figure 14 Weighted score analysis using the Factor Rating Method (FRM).

5 Conclusion

This study provides a comprehensive evaluation of renewable energy technologies in the southern region of Pakistan through a comprehensive 4E analysis (energy, exergy, economic, and environmental perspectives). The analysis reveals that all four sites of the southern region (Tando-Ghulam Ali, Sanghar, Sujawal, and Umerkot) demonstrate promising energy generation capacities for PVS, WTS, and hybrid systems. Among these, Tando-Ghulam Ali is identified as the most optimal site, with thin-film PVS systems showing the lowest Levelized Cost of Energy (LCOE) and significant Net Present Value (NPV), highlighting their economic viability.

Annual energy outputs for a 50 MW solar photovoltaic system (PVS) were estimated to be between 79,909 and 81,210 MWh/year at Tando-Ghulam Ali and between 79,656 and 80,995 MWh/year at Sanghar. For wind energy, Tando-Ghulam Ali is achieving an annual energy generation of 138,902 MWh/year, a capacity factor of 31.09%, and an energy efficiency of 35.57%. Furthermore, the weighted score analysis, using the Factor Rating Method (FRM), identifies a 40 MW PVS combined with a 9 MW WTS as the most feasible and effective configuration.

In terms of photovoltaic panel efficiency, monocrystalline and polycrystalline panels demonstrate higher efficiencies, ranging from 16.61% to 17.13% and 16.6% to 17.11%, respectively. Cadmium telluride (Cd–Te) panels, while exhibiting lower efficiencies (13.58% to 13.94%), are noted for their superior energy production capabilities compared to other panel types. Economically, thin-film PVS systems present the lowest Levelized Cost of Energy (LCOE), ranging from $0.1144 to $0.1150 per kWh, with the most favorable LCOE observed at Tando-Ghulam Ali. Conversely, WTS systems show the highest Net Present Value (NPV) although with higher LCOE values. The study also finds that Tando-Ghulam Ali achieves substantial carbon reductions, ranging from 36,027 to 36,613 tons of CO₂ annually, followed by Sanghar, Umerkot, and Sujawal.

While this study offers valuable insights for policymakers, investors, and stakeholders, guiding informed decision-making towards a sustainable energy future, it does have limitations. The economic parameters are based on current discount rates and currency conversion factors, which may fluctuate during project implementation. Additionally, the analysis is limited by the availability of weather stations, focusing on selected locations. Future research should extend to a broader range of sites, including both southern and northern regions of Pakistan, utilizing satellite data for enhanced accuracy. Future work should also explore the integration of advanced technologies, such as energy storage systems and smart grid solutions, to further optimize the efficiency and reliability of hybrid renewable energy systems.

Acknowledgments

The authors express their gratitude to ESMAP for providing access to publicly available data. Additionally, the authors extend their thanks to Muhammad Rafay Naeem, Osama Mahfooz, and Muhammad Abeer Khan for their valuable support throughout this study.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Conflicts of interest

Authors have no competing interests to declare.

Data availability statement

Data is available at relevant websites.

Author contribution statement

Muhammad Uzair Yousuf: Conceptualization, Formal analysis, Investigation, Software, Validation, Writing – original draft, Writing – review & editing.

Muhammad Hamza Malik: Conceptualization, Data curation, Formal analysis, Investigation, Software, Writing – original draft, Writing – review & editing.

Muhammad Umair: Investigation, Writing – original draft, Writing – review & editing.

References

All Tables

All Figures

	Figure 1 Sites locations for Sindh (https://globalwindatlas.info/en/area/Pakistan).
In the text

	Figure 2 Area covered by 50 MW solar farm.
In the text

	Figure 3 Area covered by 51 MW wind farm.
In the text

	Figure 4 Flowchart for the energy assessment of PVS.
In the text

	Figure 5 Flowchart for the energy assessment of WTS.
In the text

	Figure 6 Performance ratio comparison of 9 PV modules.
In the text

	Figure 7 Wake map from Bastankhah wake deficit model with weighted sum.
In the text

	Figure 8 Wind rose and weibull distribution curves for the selected sites.
In the text

	Figure 9 Annual energy generation, capacity factor, and energy efficiency.
In the text

	Figure 10 Annual output exergies.
In the text

	Figure 11 Economic evaluation of PVS, WTS, and PVWHS.
In the text

	Figure 12 Effect of interest rate on LCOE.
In the text

	Figure 13 Environmental outcomes for PVS, WTS, and PVWHS.
In the text

	Figure 14 Weighted score analysis using the Factor Rating Method (FRM).
In the text