Techno-Economic Optimization and Assessment of Solar Photovoltaic–Battery–Hydrogen Energy Systems with Solar Tracking for Powering ICT Facility

Abstract

This paper addresses the critical issue of selecting the optimal solar tracking configuration for maximum energy generation, given the increasing demand for sustainable energy solutions in information and communication technology (ICT) facilities. The main goal is to thoroughly evaluate and compare seven different solar tracking configurations across technical, economic, and environmental dimensions: No Tracking (NT), Monthly Adjusted Horizontal Axis (MAHA), Weekly Adjusted Horizontal Axis (WAHA), Daily Adjusted Horizontal Axis (DAHA), Continuously Adjusted Horizontal Axis (CAHA), Continuously Adjusted Vertical Axis (CAVA), and Dual Axis with Continuous Adjustment (DACA). This study utilizes the HOMER simulation program to evaluate its energy and hydrogen production, emissions, and cost-effectiveness performance. Key findings indicate solar tracking improves energy efficiency, with optimal capacity factors of 18.2% and 17.7% for CAHA and DAHA configurations, respectively. Although load-following strategies increase reliability, there is a trade-off between capital costs and energy costs. In addition, an MCDM approach helps to consolidate the evaluation, resulting in CAVA being ranked as the most preferable option. The study contributes to informed decision-making for energy systems in ICT facilities by emphasizing the significance of considering a variety of criteria and evaluation techniques to address complex energy challenges.

1. Introduction

Due to its role in spurring innovation and altering the dissemination of knowledge, information, and communication technology (ICT) has become an indispensable part of today’s educational institutions. Universities and other academic institutions rely extensively on information and communication technology (ICT) infrastructure because it enhances communication, collaboration, and the delivery of education. However, the growing reliance on ICT highlights the pressing requirement for a secure and sustainable energy supply to run these ICT systems. To overcome this obstacle, universities that rely on ICT must take a holistic approach to sizing and selecting hybrid energy systems that can reliably power ICT infrastructure.

According to the United Nations Educational, Scientific, and Cultural Organization (UNESCO), over 1.7 billion students worldwide were impacted by interruptions precipitated by the COVID-19 pandemic in 2020 [1]. As a result, educational institutions have rapidly shifted to online education models that rely primarily on e-learning tools, digital platforms, and virtual classrooms [2,3]. This change highlights the importance of ICT infrastructure in maintaining classroom instruction [4]. This increased reliance on ICT systems, which are energy-intensive, however, presents significant issues, especially in terms of environmental sustainability and economic viability. Goal 4 of the Sustainable Development Goal (SDG) of the United Nations focuses on ensuring that everyone has access to high-quality education and training throughout their lives [5]. To achieve this, it is essential to ensure that people have access to quality education. Information and communication technologies are crucial to improving the quality of education because they make classrooms more dynamic and engaging places to study [6]. However, the energy production needed to power this technology must be addressed to achieve a sustainable future.

Furthermore, SDG 7 promotes universal access to cheap renewable energy [7]. Powering ICT infrastructure has become increasingly important in recent years, with institutions requiring significant energy inputs to keep their technical operations running smoothly. Data centers and other ICT networks are expected to considerably increase their energy use over the next few years, according to the International Energy Agency (IEA) [8]. Because conventional energy sources are typically linked to greenhouse gas emissions and resource depletion, this raises concerns about the environmental impact of energy generation and consumption. This situation requires a comprehensive strategy that considers energy needs, costs, and environmental impacts. Hybrid energy systems, which utilize a combination of renewable and conventional power sources along with energy storage, provide a viable alternative. Universities can reduce their carbon footprint and lessen their vulnerability to power outages by increasing the variety of sources from which they draw their energy.

Economic, technical, and environmental factors must be considered for sustainable energy solutions [9]. Considering the economics of installing hybrid energy systems in educational institutions and the environmental implications is crucial. Decision-makers must carefully weigh the initial investment expenses, ongoing operation, and the possibility of cost. Cost-benefit evaluations and optimizing system frameworks are highlighted as essential steps toward ensuring long-term economic viability, as indicated in a report by the International Renewable Energy Agency (IRENA) [10]. As universities navigate the evolving landscape of education and technology, the powering of ICT infrastructure emerges as a central challenge with profound implications for sustainability and educational continuity. An integrated energy-balance, economic, and environmental approach is crucial for sizing and selecting hybrid energy systems that effectively meet the energy demands of universities while aligning with global sustainability goals. As indicated in Figure 1, this integrated approach highlights the intersection between SDG 4 (Quality Education), SDG 7 (Affordable and Clean Energy), and SDG 9 (Industry, Innovation, and Infrastructure). As indicated in Figure 1, the intersection of SDG 4’s call for quality education and SDG 7’s focus on clean energy underscores the urgency of addressing this issue. By adopting a comprehensive approach that accounts for energy requirements, economic feasibility, and environmental impacts, universities can pave the way for a resilient and sustainable ICT-powered educational future.

An optimal integrated approach that considers techno-economic and environmental factors is imperative for selecting and sizing an appropriate hybrid energy system to power an ICT infrastructure. Due to the significance of this sector, numerous studies have been conducted, resulting in an expanding body of literature focused on developing and utilizing approaches to optimize the design of hybrid energy systems for ICT infrastructure. Recent studies in this field are reviewed and summarized in

Table 1. The literature review on the application of renewable energy in powering ICT and related applications.

References	Methodology	Purpose	Findings
[14]	Optimization Model	Optimum design of a hybrid power system (wind/PV/fuel cell/hydrogen energy/grid) for deployment of modern cellular mobile infrastructure in the fifth-generation era	The power supply’s renewable component is 98.8%, priced at AUD 0.03602 per kilowatt-hour (kWh). Additionally, energy expenditures are significantly reduced by 91.5%, resulting in an annualized return on investment of 43.9%. These favorable conditions make it feasible to enhance the reliability of the fluctuating power supply, achieving an amortization period of less than three years.
[15]	Techno-economic optimization using HOMER Pro	To provide a comprehensive 3E (Technical, Economic, Environmental) analysis of renewable-dominated hybrid standalone systems for decarbonizing the telecom sector in Pakistan and to evaluate the optimal energy configurations for BTS sites.	Different configurations were optimal depending on the site location; PV–battery–DG systems were most common, with others integrating wind, hydro, and biomass sources. LCOE ranged from 0.04547 to 0.2325 AUD/kWh. Hybrid systems have proven to be more cost-effective and sustainable than diesel-only solutions.
[16]	Hybrid energy system design	Development of a hybrid (PV/fuel cell) power system for telecom base stations in Ghana to reduce LCOE and GHG emissions	It is economically feasible, with a levelized cost of energy (LCOE) of 0.222 AUD/kWh, which is lower than the 0.25 AUD/kWh for grid-connected systems. Additionally, the LCOE is 30% less expensive than a PV/battery/diesel hybrid system and 67% less expensive than the diesel power system at the site, resulting in lower CO2 emissions per year compared with the diesel power system.
[17]	Hybrid optimization modeling using HOMER Pro	To propose and evaluate a hybrid solar/hydroenergy system with hydrogen storage for powering a remote BTS site in Nigeria.	The hybrid system achieved lower NPC (AUD 843,530), LCOE (AUD 0.4516), and operating costs (AUD 14,798) than the diesel system. It also reduced GHG emissions by 65,430 kg/year.
[18]	Simulation Model	An analysis of the PV/H/FC hybrid system’s techno-economic viability for supplying base stations in the development of green mobile communications aimed at curbing environmental deterioration and averting fossil fuel crises	The hybrid system maintained a higher quality of service (QoS) while generating approximately 17% more electricity than required, thereby enhancing system reliability.
[19]	Cost and energy management techniques	Comparison analysis of hydrogen-powered and traditional data centers with different backup power systems in China	A hydrogen-powered data center might be cost-effective if hydrogen could be produced from natural gas or H2-rich industrial water streams in chemical plants. Additionally, having a data center near hydrogen sources is advantageous.
[20]	Optimization Model	Design of reliable power system with integration of fuel cell for powering data center (computer servers and associated power control equipment and distribution network, chillers, and lighting)	Achieved reliable and sustainable power supply that is carbon-free.
[21]	Techno-economic assessment using HOMER software	To decarbonize the telecom sector by integrating photovoltaic systems into BTS infrastructure and assess their financial and environmental viability across different power-outage scenarios in Pakistan	PV–DG–battery hybrid systems reduced average LCOE by 29%, DG operational hours by 82%, and carbon emissions by 92%. Demonstrated significant improvements in operational cost and sustainability for BTS sites using solar PV systems.
[22]	Simulation model	Assessment of the techno-economic analysis of hybrid-renewable powered telecom tower in India	After government policy intervention, subsidies on the renewable energy system component reduced the energy cost from AUD 0.256/kWh to AUD 0.167/kWh compared with the diesel-powered system.
[23]	HOMER software	To assess the techno-economic and environmental benefits of integrating solar PV into BTS power supply systems in Benin	Solar integration reduced LCOE by 61.26–67.77% (off-grid) and 40.27–43.85% (on-grid), NPC by 61.24–67.71% (off-grid) and 26.77–31.34% (on-grid), and GHG emissions by over 94%.
[24]	Hybrid renewable system and comparative study	Comparison of independent solar photovoltaic and hybrid power systems for Northern Ghanaian distant outdoor communication sites	The PV/battery hybrid system is economically preferred to conventional diesel generators (DGs). It has lower environmental pollution and significantly reduced maintenance costs compared with conventional energy systems.
[25]	Simulation model	Development of hybrid energy supply for powering green data centers (GDCs) to achieve minimum net system cost and mitigate emission of greenhouse gases (GHGs)	The designed hybrid system for a new generation green data center (GDC) reduced the cost of electricity (COE) while offering preferred quality of service (QoS). The cost assessment and energy evaluation validated the system’s eco-friendly implications.
[26]	Simulation model	Techno-economic analysis of a hybrid system for powering ICT building	The findings demonstrate that the hybrid power system can efficiently and sustainably improve the current unreliable power supply.
[27]	Optimization model	Optimal hybrid power system sizing to meet the load requirements of a university laboratory in Nigeria.	The hybridized power system has the potential to reduce energy costs by more than 88% and achieve a return on investment of 41.3% in three years. In addition, using an appropriately sized hybrid system can reduce rural–urban migration and may improve the nation’s economic growth.

. Most studies did not consider the optimal energy production for the proposed ICT, BTS, and data center using solar tracking devices. Solar tracking systems play a crucial role in optimizing the performance of solar panels by continuously adjusting their orientation to face the sun [12]. Optimizing sunlight exposure throughout the day enhances energy output. It improves the cost-effectiveness of solar energy systems [13]—Additionally, only a few studies considered using hydrogen storage in implementing the proposed energy systems. Furthermore, none of the selected studies identified the optimal system using multiple criteria; instead, the optimal energy systems were identified based on a single criterion.

Contribution of the Present Study

Although the studies considered in the literature section offer insightful analyses of hybrid energy systems and the integration of renewable energy, there are several gaps, which include the application of batteries and hydrogen tanks together as energy storage, the effect of solar tracking on energy storage, the incorporation of multi-criteria decision-making (MCDM) in ranking energy systems, and comprehensive assessments of the technical, financial, and environmental performance of renewable energy systems. Most of the literature falls short of providing an integrated technical, economic, and environmental evaluation of hybrid renewable energy systems. Studies usually concentrate on one or two components in isolation. These elements can be combined into an integrated study to identify superior designs that maximize trade-offs. Furthermore, little research has been conducted on integrating batteries and hydrogen storage tanks as a combined energy storage solution. To fully realize the potential of these technologies in hybrid systems, evaluating the complementary trade-offs between them is necessary.

Furthermore, although various studies have been conducted to investigate the effect of solar tracking on component sizing, few have examined the impact of tracking on the technical features of storage devices, particularly those utilizing hydrogen and battery storage. Research on the effects of solar tracking on storage specifications is necessary to manage variability effectively, as tracking influences solar intermittency. Moreover, hybrid system designs could be systematically evaluated for performance across technological, economic, and sustainability considerations using multi-criteria decision-making (MCDM) methodologies. Drawing from the gaps identified in this literature and considering the importance of energy in sustaining the new paradigm shift in the use of ICT in the education sector, this study aims to do the following:

• Conduct comprehensive analyses that consider technical, economic, and environmental performance to inform the design and implementation of renewable energy systems, providing decision-makers and stakeholders with an in-depth perspective.
• Examine the synergistic utilization of hydrogen tanks and batteries as energy storage solutions in a hybrid energy system for powering an ICT facility while evaluating their performance, efficiency, and cost-effectiveness.
• This paper examines the effects of various solar tracking setups on the sizing and functioning of energy storage systems, specifically focusing on hydrogen tanks and batteries.
• Utilize multi-criteria decision-making (MCDM) frameworks to assess and prioritize solutions for hybrid energy systems, considering technical, financial, and environmental factors.

Integrating an energy-balance, economic, environmental, and multi-criteria decision-making (MCDM) approach in the sizing and selection of hybrid energy systems is crucial for ensuring optimal energy use and minimizing the carbon footprint.

2. Materials and Methods

2.1. Methodology

This section outlines the methodology employed to conduct this study. It also discusses the software used, the various solar tracking configurations considered, and the components of the energy system.

2.2. System Design

As shown in Figure 2, the proposed energy system comprises a photovoltaic (PV) panel, an energy storage system, a converter (inverter and rectifier), and the load. The storage system consists of a battery and a hydrogen storage unit. The hydrogen storage consists of its conversion unit—fuel cell, hydrogen tank, and electrolyzer. The PV and battery produce direct current, which is connected to the DC bus, while the fuel cell is connected to the AC bus system. The load is fed through the AC bus, while power from the PV and battery is converted to AC through the converter and then fed to the load. The proposed energy system is designed using HOMER software Version 2.86 beta. Utilizing software to simulate energy systems presents numerous benefits compared with experimental investigations.

To begin with, simulations offer a cost-effective alternative by mitigating substantial expenditures associated with equipment, materials, labor, and maintenance inherent to experimental approaches. Software is crucial in expediting the testing process by enabling the swift evaluation of multiple scenarios, thereby eliminating the time-consuming procedures associated with establishing and executing physical tests. The unparalleled versatility of simulations enables the effortless modeling of diverse circumstances and configurations. In contrast to practical experiments, simulations offer a secure setting for exploring diverse possibilities without the inherent hazards associated with the former. Simulation tools are highly valuable educational platforms that provide a safe environment for learning and experimentation, eliminating potential risks. Nevertheless, it is essential to remember that simulations rely on models, and it is crucial to periodically verify their validity by comparing them to empirical data to ensure their accuracy.

Currently, numerous tools are available for conducting technical and economic evaluations of integrated renewable energy (RE) systems. Some software for implementing the techno-economic and environmental features of energy systems includes TRNSYS, iHOGA 4.0, Hybrid2, PVSyst 8, RETScreen, and HOMER Pro [28]. HOMER stands out as a powerful tool for the technical and economic study and simulation of renewable energy systems [29]. With HOMER, complex energy systems with various energy sources, storage alternatives, and demand patterns can be thoroughly analyzed. It considers a variety of variables, including economics, resource availability, technical requirements, and more. The core strength of HOMER lies in its optimization algorithms. It explores numerous combinations of energy sources, storage, and other components to identify the most cost-effective and reliable system configuration based on the defined criteria. Hence, HOMER performs tasks such as modeling, optimization, sensitivity analysis, and solar tracking, among others.

At its core, HOMER operates by simulating the behavior of each system configuration over an entire year, typically on an hourly time step. The simulation accounts for the dynamic interactions between energy generation, storage, and load demand. It also incorporates real-world factors such as resource variability (e.g., fluctuations in solar irradiance), load profile variations, component degradation, and energy losses [30]. These simulations help determine how well each configuration can meet the energy demand of the defined load—in this case, the ICT facility—under the given environmental and technical constraints. Once the simulation step is complete, HOMER conducts an optimization process by evaluating thousands of system configurations across a defined solution space. The tool evaluates each configuration based on its ability to meet energy demand reliably and cost-effectively. The primary objective of the optimization is to minimize the system’s total net present cost (TNPC) over its project lifetime. However, other criteria, such as cost of energy (COE), greenhouse gas emissions, or fuel consumption, can also be considered depending on user preferences [31].

In addition to simulation and optimization, HOMER also performs sensitivity analysis to evaluate how changes in uncertain input variables—such as fuel prices, solar radiation levels, component efficiencies, or capital costs—affect the feasibility and ranking of configurations [32]. This analysis enhances the robustness of the decision-making process by revealing the conditions under which one configuration may become more favorable than others. The final outputs from HOMER include a ranked list of viable system configurations, detailed economic metrics (e.g., capital costs, replacement costs, operating and maintenance costs), energy balance results (e.g., energy production and consumption per component), reliability indicators (e.g., unmet load, capacity shortage), and environmental performance (e.g., CO₂ and NO_x emissions) [33,34]. Given the local context and resource availability, these outputs enable stakeholders to decide which system design best meets their objectives.

2.3. Solar PV

HOMER uses Equation (1) to calculate the PV array’s power output under typical test conditions and then makes any necessary adjustments for derating factors, changes in solar irradiation, and temperature variations.

$$P_{actual}=\left[1+{\alpha }_P\times \left(T_C-T_{C,STC}\right)\right]\times Y_{pv}{\times f}_{pv}\times \left({\displaystyle \frac{G}{G_{STC}}}\right)$$

(1)

$P_{actual}$ is the actual power output of the PV array (kW), $T_{C,STC}$ is the PV cell temperature under standard test conditions (25 °C), $T_C$ is the PV cell temperature in the current time step (°C), $f_{pv}$ is the PV derating factor (as a fraction, e.g., 0.90 for 90%), $Y_{pv}$ is the rated capacity of the PV array under standard test conditions (kW), $G$ is the solar radiation incident on the PV array in the current time step (kW/m²), $G_{STC}$ is the incident radiation at standard test conditions (1 kW/m², as given), and ${\alpha }_P$ is the temperature coefficient of power (percentage decrease or increase in efficiency per degree Celsius change, e.g., −0.005 or −0.5% per °C for decrease).

2.3.1. Solar Photovoltaic (PV) Tracking Systems

Solar tracking systems have been developed to maximize the energy efficiency of photovoltaic (PV) modules by precisely monitoring and adjusting their orientation in response to the sun’s trajectory throughout the day [35]. Single-axis tracking systems are widely utilized and economically efficient solar trackers. These systems can enhance the energy production of a photovoltaic (PV) array by 20–40% compared with fixed-tilt systems [36]. Solar photovoltaic (PV) tracking systems are essential for maximizing the energy output of solar panels by adjusting their orientation to match the sun’s position throughout the day and year. This is accomplished by employing various tracking strategies, each with its advantages and drawbacks, which are briefly discussed in this section [37].

The ability of solar tracking systems to enhance and stabilize solar power generation compared with stationary installations has a significant impact on properly sizing and operating battery storage, hydrogen storage, and fuel cell components in renewable energy systems. This study aims to elucidate that various configurations of solar tracking systems significantly impact the performance of battery storage, hydrogen storage tanks, and fuel cells. To prevent challenges such as oversized systems that increase costs or undersized components that reduce reliability, it is crucial to consider the effects of solar tracking when designing hybrid renewable energy systems, as in the case considered in this study. To improve the performance and cost-effectiveness of the hybrid renewable energy system, this study simulated various solar PV tracking configurations.

The solar tracking configurations considered include Monthly Adjusted Horizontal Axis (MAHA), Weekly Adjusted Horizontal Axis (WAHA), Daily Adjusted Horizontal Axis (DAHA), Continuously Adjusted Horizontal Axis (CAHA), Continuously Adjusted Vertical Axis (CAVA), and Dual Axis with Continuous Adjustment (DACA). Tracking systems are classified based on the number of axes of rotation and the frequency of modifications. The description and justification for each solar PV tracking configuration are described below. HOMER is used to implement these solar tracking configurations.

2.3.2. Solar Tracking Configurations

No Tracking: The panels’ azimuth and slope are fixed. This is the simplest and most common case.

Monthly Adjusted Horizontal Axis (MAHA)

MAHA tracking rotates solar panels once a month on a horizontal axis to account for shifting sun angles with the seasons. While this method can capture more energy than fixed systems, it does not account for daily variations in solar angles, resulting in considerable energy loss during the day.

Weekly Adjusted Horizontal Axis (WAHA)

WAHA tracking, like MAHA, modifies panel orientation weekly. This method balances the energy gains obtained by monthly tracking and the complexities of more frequent modifications. It captures more energy throughout the year than fixed installations.

Daily Adjusted Horizontal Axis (DAHA)

The DAHA tracking system adjusts the orientation of the solar panels daily to align them with the sun’s position. This technique enhances energy capture efficiency by responding to variations in sun angles on seasonal and daily time scales. Although it increases energy production, it necessitates the implementation of more complex monitoring mechanisms and control systems.

Continuously Adjusted Horizontal Axis (CAHA)

The process of CAHA tracking involves continuously adjusting the horizontal alignment of solar panels to follow the sun’s real-time movement accurately. According to the literature, this methodology presents the most significant energy benefits due to its ability to adapt to solar angle variations on a minute-by-minute basis. Nevertheless, the presence of mechanical and control complexity has the potential to increase costs and necessitate higher maintenance efforts.

Continuously Adjusted Vertical Axis (CAVA)

CAVA tracking entails dynamically altering the vertical orientation of solar panels to track the sun’s height angle. This method proves particularly beneficial in places where the solar elevation varies significantly throughout the day. CAVA tracking can enhance performance during low solar angles, such as in the mornings and evenings.

Dual Axis with Continuous Adjustment (DACA)

DACA tracking uses horizontal and vertical movements to ensure solar panels are always in the best position to extract energy. This method gives the most energy gains throughout the year because it considers both the angle of the sun and how high it is in the sky. However, it is plagued with complexity regarding its architecture, higher implementation costs, and potential maintenance problems.

2.3.3. Choice of Tracking Strategy

Determining the optimal solar tracking method involves achieving an equilibrium between factors such as energy output, cost analysis, sun angle variations particular to a site, space availability, and maintenance capabilities. Each strategy helps bring solar energy closer to realizing its full potential as a sustainable energy option. The choice of tracking strategies is discussed as follows [12,38]:

Energy output goals: more complex and robust tracking systems, like CAHA and DACA, provide more energy output but at a higher cost of implementation and maintenance.

Location: Solar angle fluctuations are determined by geographic location. Regions located at higher latitudes undergo more significant variations in sun angles over a year, making tracking systems more advantageous.

Budget and maintenance: Complex tracking systems require sophisticated technology and periodic maintenance. When choosing a solar tracking mechanism, consider the potential long-term expenses and the level of technical proficiency available.

Space requirements: because tracking systems require mobility, their spatial requirements may be greater than those of fixed installations.

2.4. Battery Storage

HOMER determines the maximum amount of electricity the battery bank can take in for each hour of the simulation. When determining whether the battery can take all available surplus renewable power or how much surplus power a cycle charging generator should provide, it uses this “maximum charge power” as a guide. Depending on the battery’s charge level and recent history of charges and discharges, the maximum charge power changes from hour to hour. The maximum charging power of the battery bank is subject to three different constraints set by HOMER. HOMER restricts the maximum charging power of the battery bank in three different ways. The kinetic battery model imposes the first constraint; the maximum amount of power that the two-tank system can absorb is given by the following equation [37]:

$$P_{batt,cmax,kbm}={\displaystyle \frac{QKC(1-e^{-k∆t})+K{Q}_1{e}^{-k∆t}}{C(K∆t-1+e^{-k∆t})+1-e^{-k∆t}}}$$

(2)

where $Q$ is the battery’s total energy at the start of the time step, ${ Q}_1$ is the energy available at that moment, and $∆t$ is the duration of the time step. $K$ is the rate constant, $\mathrm{C}$ is the capacity ratio, and $Q_{max}$ is the total energy/maximum battery capacity.

The second restriction relates to the battery’s maximum charge rate, which is shown by the A/Ah value. The following equation provides the battery charge power equivalent to this maximum charge rate [37]:

$$P_{batt,cmax,mc\gamma }={\displaystyle \frac{(1-e^{-{\propto }_c∆t})\left(Q_{max}-Q\right)}{∆t}}$$

(3)

${\propto }_c$ is the battery’s maximum charge rate [A/Ah], and $Q_{max}$ is the total capacity of the battery bank [kWh]

The battery’s maximum charge current is the subject of the third restriction. The following equation provides the maximum battery bank charge power that corresponds to this maximum charge current [39]:

$$P_{batt,cmax,mcc}=I_{max}\times V_{nom}\times N_{batt}\times {10}^{-3}$$

(4)

$N_{batt}$ is the number of batteries in the battery bank, $I_{max}$ is the battery’s maximum charge current [A], and $V_{nom}$ is the battery’s nominal voltage [V].

Therefore, assuming each limitation applies after charging losses, the maximum battery charge power is equal to the least of these three values and represented as [39]:

$$P_{batt,cmax}={\displaystyle \frac{min\left(P_{batt,cmax,mcc},P_{batt,cmax,mc\gamma },P_{batt,cmax,kbm}\right)}{{\eta }_{batt,c}}}$$

(5)

where ${\eta }_{batt,c}$ is the battery charge efficiency.

The cyclic energy efficiency of a battery is a crucial factor to consider when selecting a battery since it necessitates an enormous input source for replenishment in the event of energy loss. In addition to the capital cost, several other crucial aspects contribute to determining lifecycle costs. These factors encompass the number of cycles provided at a specific discharge depth, the battery’s operational lifespan (often ranging from 3 to 7 years), and the prevailing maintenance practices. Equations (6) and (7) in HOMER estimate the battery autonomy and battery bank life, respectively [40].

$${Batt}_{aut}={\displaystyle \frac{n_b{V}_n{Q}_n\left(\frac{100-{SOS}_{min}}{100}\right)\left(24 \mathrm{h}/\mathrm{d}\mathrm{a}\mathrm{y}\right)}{D_{pry,ave}\left(1000 \mathrm{W}\mathrm{h}/\mathrm{k}\mathrm{W}\mathrm{h}\right)}}$$

(6)

where $D_{pry,ave}$ is average primary energy demand, ${SOS}_{min}$ is minimum state of charge (%), $Q_n$ is nominal capacity of a single battery (Ah), $V_n$ is the nominal voltage of a single battery (V), and $n_b$ is the number of battery [40]

$${Batt}_{life}=\left\{\begin{array}{c}{\displaystyle \frac{n_b\times Q_{lifetime}}{Q_{thrpt}}} if limited by throughput\\ c if limited by time\\ min\left({\displaystyle \frac{n_b\times Q_{lifetime}}{Q_{thrpt}}},B_{fl}\right) if limited by throughput and time\end{array} \right.$$

(7)

$n_b$ is the number of batteries, $Q_{lifetime}$ is the lifetime throughput of a single battery (kWh), $Q_{thrpt}$ is the annual throughput of the battery $(\mathrm{k}\mathrm{W}\mathrm{h}/\mathrm{y}\mathrm{r})$, $B_{fl}$ is battery float life (yr) [40]

$$Q_{lifetime}=f_c\times DOD\left({\displaystyle \frac{V_n{q}_{max}}{1000 \mathrm{W}/\mathrm{k}\mathrm{W}}}\right)$$

(8)

where $f_c$ is the number of cycles to failure $,DOD$ is the depth of discharge (%), and $q_{max}$ is the maximum capacity of the battery (Ah).

2.5. Hydrogen Storage

The hydrogen storage consists of the fuel cell, hydrogen tank, and the electrolyzer. The electrolyzer’s work is to turn water into hydrogen gas via electrolysis. The water molecules separate into hydrogen and oxygen when an electric current is sent through them (Equation (9)). The hydrogen gas generated in the electrolyzer is generally captured and stored for subsequent uses. Electrolyzers use sustainable energy sources, such as solar or wind power, to generate environmentally friendly hydrogen without carbon emissions [37].

$$H_{2}O\to {\displaystyle \frac{1}{2}}{O}_2+H_2$$

(9)

The hydrogen gas produced by the electrolyzer is kept in either high-pressure or cryogenic tanks, depending on the physical state of the hydrogen, whether it is gaseous or liquid. The tanks have been specifically engineered to securely store hydrogen at the necessary pressure or temperature until it is required for diverse uses.

A fuel cell is an electrochemical device that produces electrical energy by facilitating a reaction between hydrogen and oxygen [41]. When hydrogen is introduced from the storage tank to the fuel cell’s anode and oxygen is provided to the cathode, a chemical reaction occurs, generating electricity, heat, and water as secondary products (Equation (10)). The generated electricity has the potential to be utilized in a wide range of applications, encompassing the powering of vehicles, stationary power generation, and several other uses. Fuel cells are characterized by their outstanding efficiency and ability to generate energy without emitting harmful substances, positioning them as a sustainable and environmentally friendly technology for energy conversion [42].

$${\displaystyle \frac{1}{2}}{O}_2+H_2\to H_{2}O+Electricity+Heat$$

(10)

HOMER can simulate a system for hydrogen-based electrical storage. This system involves an electrolyzer that makes hydrogen from surplus energy, a storage tank for the hydrogen, and a hydrogen-fueled generator that transforms the stored hydrogen back into electricity when there is a shortage of electrical supply. The hydrogen tank autonomy is obtained by dividing the energy capacity of the hydrogen tank by the electricity demand; it is estimated using Equation (11) [42].

$$H_{aut}={\displaystyle \frac{C_{htank}{LHV}_h\left(24 \mathrm{h}/\mathrm{day}\right)}{D_{pry,ave}\left(3.6 \mathrm{M}\mathrm{J}/\mathrm{k}\mathrm{W}\mathrm{h}\right)}}$$

(11)

$C_{htank}$ is the size of the hydrogen tank, and ${LHV}_h$ is hydrogen’s energy content (lower heating value) [120 MJ/kg].

2.6. Economic Details of the Components

HOMER determines the optimal mix of components with the lowest cost that can fulfill electrical and thermal loads. HOMER utilizes a simulation process to model numerous system configurations, focusing on minimizing lifecycle costs (Equation (12)). Additionally, it provides the outcomes of sensitivity analyses for most input variables. The energy system with the lowest total net present cost (TNPC) is the most competitive. The levelized cost of energy for the technically possible systems is also determined using Equation (13) [42].

$$Min. C_{acc}=TNPC\times \left({\displaystyle \frac{i.[1+i{]}^Y}{[1+i{]}^Y-1}}\right)$$

(12)

$$LCOE={\displaystyle \frac{C_{anc}}{E_{Total}}}$$

(13)

$$LCOPV={\displaystyle \frac{C_{anc,PV}}{E_{Total}}}$$

(14)

$$LCOH={\displaystyle \frac{C_{anc,H_2}}{E_{Total}}}$$

(15)

where $i$ is the real discounted rate, $Y$ is the project lifetime, $C_{anc}$ annualized cost of the system, $C_{anc, PV}$ annualized cost of PV, $C_{anc,{ H}_2}$ annualized cost of the hydrogen tank, and $E_{Total}$ is the total energy production.

The research examined a 12 V battery with a capacity of 200 ampere-hours, initial costs amounting to USD 357, and a replacement cost of USD 350. The annual maintenance cost for the battery is USD 1.5, and it is anticipated to be replaced on a four-year cycle. A 1 kW fuel cell system with an initial cost of USD 4000 and a replacement cost of USD 3000 is being considered. The operating cost is USD 0.01 per hour, estimated to last 40,000 h. Furthermore, an 8 kW electrolyzer with an USD 2700 initial cost and no difference in replacement cost is included in the analysis. The electrolyzer has an annual maintenance cost of USD 3 and is anticipated to last 15 years. Also included in the analysis is an 8 kg hydrogen storage tank, which costs USD 3100 to buy and replace and has an annual maintenance cost of USD 4; the lifespan of the hydrogen storage tank is 2 years. Other economic details of the components are given in

Table 2. Economic details of the components [37,43].

Component Parameter	Rated Capacity	Capital Cost (USD)	Replacement Cost (USD)	O and M Cost	Lifespan
Battery	12 V, 200 Ah	357	350	USD 1.5/yr	4 yrs
Fuel cell	1 kW	4000	3000	USD 0.01/h	40,000 h
Electrolyzer	8 kW	2700	2700	USD 3/yr	15 yrs
Hydrogen tank	8 kg	3100	3100	USD 4/yr	25 yrs
Converter	1 kW	37	36	USD 2/yr	10 yrs
Solar PV and Sun-Tracking Technologies
NT	0.25 kW	295	290	USD 0/yr	20 yrs
MAHA	0.25 kW	435.75	413.96	3.26	20 yrs
WAHA	0.25 kW	435.75	413.96	3.26	20 yrs
DAHA	0.25 kW	435.75	413.96	3.26	20 yrs
CAHA	0.25 kW	512.5	486.87	3.84	20 yrs
CAVA	0.25 kW	407.5	387.12	3.05	20 yrs
DACA	0.25 kW	545	517.75	4.08	20 yrs

.

2.7. Study Area

The study was conducted at Anyigba, Kogi State University, in Anyigba, Nigeria (7.4934° N and 7.1736° E). The university is home to a wide variety of academic departments and programs. These include the College of Agriculture, the College of Arts and Humanities, the College of Law, the College of Management Science, the College of Natural Science, the College of Social Science, the College of Medicine, and the College of Veterinary Medicine. Solar radiation at the location varies throughout the year in this region, with the highest levels recorded in March (5.484 kWh/m²/day) and the lowest recorded in July (3.70 kWh/m²/day), all of which contribute to the scaled annual average of roughly 4.87 kWh/m²/day (Figure 3). The university’s climate and environmental dynamics are impacted by the year-round variations in temperature, with February seeing the greatest at 27 °C and July and August seeing the lowest at an average of 24.2 °C. The meteorological data were obtained from the National Renewable Energy Laboratory.

2.8. Electricity Demand

Figure S1 shows a detailed hourly breakdown of power consumption (in watts) for various equipment and appliances in an ICT facility at Kogi State University over 24 h. IT equipment such as servers, switches, and computers are included, as are environmental control systems such as air conditioners and fans, as well as electrical gadgets such as refrigerators and televisions. The data show both steady power consumption, most notably by IT equipment, and scheduled and intermittent consumption patterns, most notably in air conditioning devices. It is worth mentioning that the highest electricity demand occurs in the morning hours, between 9 and 2 p.m. (Figure S1), due to the operation of air conditioning systems and significant computer activity. In contrast, power usage drops significantly throughout the night, corresponding to lower activity levels. The total hourly consumption represents the overall daily power usage, roughly 358.48 kilowatt-hours (kWh). A day-to-day and time-step-to-time-step random variability of 10% each is added to the power consumption to make it more realistic.

2.9. Multi-Criteria Decision-Making (MCDM) Framework

This study employed a multi-criteria decision-making (MCDM) framework to rank the performance of different solar tracking configurations based on technical, economic, and environmental attributes. The MCDM methods used include CODAS (COmbinative Distance-based Assessment) [45], ARAS (Additive Ratio Assessment) [46], EDAS (Evaluation based on Distance from Average Solution) [40], and MOORA (Multi-Objective Optimization based on Ratio Analysis) [47]. These methods were selected due to their robustness in handling decision problems with multiple conflicting criteria. The application of multiple MCDM methods provides a robust framework for decision-making, allowing for cross-validation of results and identification of any methodological biases.

2.9.1. Entropy Weighing Method

Prior to applying the MCDM methods, criteria weights were determined using the entropy weighing method, which objectively assigns weights based on the information provided by each criterion. The entropy weighing method follows these steps [45]:

• Normalize decision matrix:

$$r_{ij}={\displaystyle \frac{x_{ij}}{\sum _{i=1}^m{x}_{ij}}}$$

where $x_{ij}$ represents the performance of the $ith$ alternative with respect to the $jth$ criterion.

2.

Calculate the entropy value for each criterion:

$$e_j=-{\displaystyle \frac{1}{\ln \left(m\right)}}\sum _{i=1}^m{r}_{ij}\ln \left(r_{ij}\right)$$

3.

Calculate the degree of diversification:

$$d_j=1-e_j$$

4.

Calculate the final weight:

$$w_j={\displaystyle \frac{d_j}{\sum _{i=1}^n{d}_j}}$$

where $m$ is the number of alternatives and $n$ is the number of criteria.

2.9.2. CODAS Method

The CODAS method ranks alternatives based on a combination of Euclidean and Taxicab distances from the negative-ideal solution. The procedure includes the following [45]:

• Normalize the decision matrix using a linear normalization method.
• Calculate the weighted normalized decision matrix.
• Determine the negative-ideal solution for each criterion
• Compute each alternative’s Euclidean $\left(E_i\right)$ and Taxicab $\left(T_i\right)$ distances from the negative-ideal solution.
• Construct the relative assessment matrix using the following:

  $$h_{ik}=\left(E_i-E_k\right)+\left(\psi \times \left(T_i-T_k\right)\right)$$

  where $\psi$ is a threshold parameter (typically set to 0.02) and:

  $$\psi \times \left(T_i-T_k\right)=\left\{\begin{array}{c}1 if \left|E_i-E_k\right|<\tau \\ 1 if \left|E_i-E_k\right|\ge \tau \end{array}\right.$$

  with $\tau$ being a threshold parameter (typically set to 0.01).

• Calculate the assessment score for each alternative:

  $$H_i=\sum _{k=1}^m{h}_{ik}$$
• Rank the alternatives based on assessment scores, with higher scores indicating better alternatives.

2.9.3. ARAS Method

The ARAS method evaluates alternatives by comparing their utility function values relative to the optimal alternative. The steps include the following [46]:

• Normalize the decision matrix:
  • For benefit criteria:

    $$r_{ij}={\displaystyle \frac{x_{ij}}{\sum _{i=0}^m{x}_{ij}}}$$
  • For cost criteria:

    $$r_{ij}={\displaystyle \frac{1/x_{ij}}{\sum _{i=0}^{m}1/x_{ij}}}$$
• Calculate the weighted normalized decision matrix.
• Determine the optimality function for each alternative:

  $$S_i=\sum _{j=1}^n{w}_j\times r_{ij}$$
• Compute the utility degree of each alternative:

  $$K_i={\displaystyle \frac{S_i}{S_o}}$$

  where $S_o$ is the optimality function value for the optimal alternative.
• Rank the alternatives based on utility degrees, with higher values indicating better alternatives.

2.9.4. EDAS Method

The EDAS method evaluates alternatives based on their distance from the average solution [40]:

• Compute the average solution for each criterion:

  $${AV}_j={\displaystyle \frac{\sum _{i=1}^m{x}_{ij}}{m}}$$
• Calculate the positive distance from average (PDA) and negative distance from average (NDA):
  • For benefit criteria:

    $${PDA}_{ij}={\displaystyle \frac{\mathrm{m}\mathrm{a}\mathrm{x}(0,(x_{ij}-A_j\left)\right)}{A_j}}$$

    $${NDA}_{ij}={\displaystyle \frac{\mathrm{m}\mathrm{a}\mathrm{x}(0,(A_j-x_{ij}\left)\right)}{A_j}}$$
  • For cost criteria

    $${PDA}_{ij}={\displaystyle \frac{\mathrm{m}\mathrm{a}\mathrm{x}(0,(A_j-x_{ij}\left)\right)}{A_j}}$$

    $${NDA}_{ij}={\displaystyle \frac{\mathrm{m}\mathrm{a}\mathrm{x}(0,(x_{ij}-A_j\left)\right)}{A_j}}$$
• Calculate the weighted sum of PDA and NDA:

  $${SP}_i=\sum _{j=1}^n{w}_j\times {PDA}_{ij}$$

  $${SN}_i=\sum _{j=1}^n{w}_j\times {NDA}_{ij}$$
• Normalize the values of SP and SN:

  $${NSP}_i={\displaystyle \frac{S{P}_i}{{max}_i\left(S{P}_i\right)}}$$

  $${NSN}_i=1-{\displaystyle \frac{S{N}_i}{{max}_i\left(S{N}_i\right)}}$$
• Calculate the appraisal score:

  $${AS}_i={\displaystyle \frac{1}{2}}\left({NSP}_i+{NSN}_i\right)$$
• Rank the alternatives based on AS values, with higher values indicating better alternatives.

2.9.5. MOORA Method

The MOORA method evaluates alternatives using a ratio system approach [47]:

• Normalize the decision matrix:

  $$r_{ij}={\displaystyle \frac{x_{ij}}{\sqrt{\sum _{i=1}^m{x_{ij}}^2}}}$$
• Calculate the benefit and cost criteria values:

  $$y_i=\sum _{j=1}^n{w}_j\times {r_{ij}}^{+}$$

  $$z_i=\sum _{j=1}^n{w}_j\times {r_{ij}}^{-}$$
• Compute the MOORA performance score:

  $$M_i=y_i-z_i$$
• Rank the alternatives based on $M_i$, where higher scores indicate more favorable alternatives.

3. Results and Discussion

This section presents the results of the techno-economic simulation of the proposed energy systems depicted in Figure 2 and the seven solar tracking configurations considered. It also presents the outcome of the MCDM.

3.1. Technical

This section discusses the technical details of the 7 PV tracking configurations considered during the simulation of the optimal energy system for the ICT facility (

Table 3. Optimal energy system for the ICT facility.

Configuration	Rank	PV (kW)	FC (kW)	Battery (no)	Converter (kW)	Electrolyzer (kW)	H2 Tank (kg)	Dispatch Strategy	PV Production (kWh/yr)	FC Production (kWh/yr)	Tot. Electrical Production (kWh/yr)	Cap. Shortage (kWh/yr)	Unmet Load (kWh/yr)	Excess Electricity (kWh/yr)
NT	1	150	50	70	50	80	100	LF	223,954	21,868	245,822	1239	1078	36,413
	2	200	50	0	50	80	100	CC	298,605	31,870	330,475	97	86	96,986
CAHA	1	150	50	60	50	80	100	CC	238,897	23,521	262,418	414	364	47,666
	2	150	50	0	50	80	120	CC	238,897	33,569	272,466	1422	1239	34,891
CAVA	1	150	50	60	50	80	100	CC	232,127	24,323	256,450	104	91	39,256
	2	150	50	0	50	80	100	CC	232,127	34,733	266,860	711	624	26,421
DACA	1	120	50	80	50	80	150	LF	225,024	23,233	248,257	1391	1224	33,294
	2	200	50	0	50	80	120	CC	375,040	30,348	405,388	79	68	175,084
DAHA	1	150	50	60	50	80	100	CC	235,549	23,770	259,319	820	722	44,276
	2	150	50	0	50	80	150	CC	235,549	34,243	269,792	1252	1086	28,993
MAHA	1	150	50	70	50	80	100	LF	223,954	21,868	245,822	1239	1078	36,413
	2	200	50	0	50	80	100	CC	298,605	31,870	330,475	97	86	96,986
WAHA	1	150	50	60	50	80	100	CC	225,323	24,383	249,705	711	625	32,890
	2	150	50	0	50	80	150	CC	225,323	34,884	260,207	1199	1044	17,742

). As discussed in the methodology section, the 7 PV tracking arrangements include NT, CAHA, CAVA, DACA, DAHA, MAHA, and WAHA. The optimal configuration for the NT and MAHA tracking setup is a load-following PV system consisting of a 150 kW solar panel with a 50 kW FC, 70 batteries, a 50 kW converter (comprising a 50 kW inverter and a 37.5 kW rectifier), an 80 kW electrolyzer, and a 100 kg hydrogen tank. The optimal configuration of the CAVA tracking arrangements consists of a 150 kW PV, 50 kW FC, 60 batteries, a 50 kW converter, an 80 kW electrolyzer, and a 100 kg hydrogen tank in a cycle charging format. The simulation results for the DACA tracking arrangement indicate that a 120 kW PV array, 50 kW FC, 80 batteries, 50 kW converter, 80 kW electrolyzer, and 150 kg hydrogen tank in a load-following format would be needed to satisfy the energy demand of the ICT facility. The optimal configuration for the DAHA and WAHA tracking arrangements is a charging configuration PV system consisting of a 150 kW solar panel with 50 kW FC, 60 batteries, 50 kW converter (50 kW inverter and 37.5 kW rectifier), 80 kW electrolyzer, and 100 kg hydrogen tank.

Regarding solar energy production (Figure 4), the CAHA had the highest production, totaling 238,897 kWh/year. The capacity factor of the solar panel connected to the CAHA tracking configuration is 18.2%. The capacity factor of the solar PV is calculated by obtaining the average power output of the PV array (in kW) divided by its rated power. The second tracking configuration with the highest PV production is the DAHA, with an energy production of 235,549 kWh/yr and a capacity factor of 17.9%.

Figure 5 presents the contribution of the PV and FC to the total electrical production for each solar tracking configuration. In the NT, CAHA, CAVA, DACA, DAHA, and MAHA configurations, the PV and FC contributed 91% and 9%, respectively. In the WAHA solar tracking configuration, the PV contributed 90% of the total energy production, and the FC contributed 10%.

The least PV energy production is observed from the non-tracking energy configuration and MAHA; their capacity factor is almost the same (17% and 17.1%, respectively). The FC production and capacity factor (Figure 6) show a direct relationship. The highest FC electrical production is attributed to the WAHA tracking arrangement, which produces 24,383 kWh of energy annually; its 5.57% capacity factor is also the highest among the tracking configurations. Again, the lowest FC energy production (21,868 kWh/yr) is observed from the non-tracking energy configuration and MAHA, while their capacity factor is 4.99%. From the foregoing, because the energy conversion efficiency of tracking systems was higher than that of non-tracking systems, the tracking systems have higher power generation.

According to Table 3, when comparing the optimal systems for the seven solar tracking configurations, CAHA has the highest excess electricity production (47,666 kWh/yr), DACA has the highest unmet load (1224 kWh/yr), and, consequently, the highest capacity shortage (1391 kWh/yr).

When decomposed to the monthly energy production (PV and FC), as seen in Figure S2, the highest PV energy generation for each solar tracking configuration occurs in December, while the lowest generation is observed in July when the rainy season is at its peak. The CAHA produced the highest amount of energy, with 33.721 kWh, while DACA produced the least (18.204 kWh). The expression “$NT<MAHA<DACA<WAHA<CAVA<DAHA<CAHA$” describes the comparison of the maximum energy produced by each configuration relative to each other. The highest production of FC (Figure S3) can be attributed to the CAHA solar tracking configuration in September, while the least production is credited to MAHA in February. Furthermore, as shown in Figure S4, the CAHA configuration produced the least hydrogen in February, while the highest hydrogen production was observed in January; this was produced by the DACA solar tracking configuration.

Regarding battery energy storage, the study compared the results related to annual throughput, autonomy, average energy cost, battery wear cost, energy in, energy out, expected life, throughput, losses, nominal capacity, storage depletion, and usable nominal capacity for the solar tracking configurations.

Based on the findings presented in

Table 4. Battery and hydrogen tank storage simulation outputs.

Battery Storage
System Variables	NT	CAHA	CAVA	DACA	DAHA	MAHA	WAHA
Usable nominal capacity (kWh)	101	86.4	88.4	115	86.4	101	86.4
Storage depletion (kWh/yr)	24	24	24	24	24	24	24
Nominal capacity (kWh)	168	144	144	192	144	168	144
Losses (kWh/yr)	3694	3683	3698	3687	3693	3694	3737
Lifetime throughput (kWh)	64,190	55,020	55,020	73,360	55,020	64,190	55,020
Expected life (yr)	3.83	3.29	3.28	4.39	3.29	3.83	3.25
Energy out (kWh/yr)	14,979	14,936	14,995	14,950	14,974	14,979	15,153
Energy in (kWh/yr)	18,697	18,643	18,718	18,661	18,691	18,697	18,914
Battery wear cost (AUD/kWh)	0.427	0.427	0.427	0.427	0.427	0.427	0.427
Average energy cost (AUD/kWh)	0	0	0	0	0	0	0
Autonomy (hrs)	6.8	5.82	5.82	7.7	5.82	6.8	5.82
Annual throughput (kWh/yr)	16,747	16,699	16,765	16,714	16,742	16,747	16,941
Hydrogen tank storage
System variables	NT	CAHA	CAVA	DACA	DAHA	MAHA	WAHA
Hydrogen generation (kg/yr)	1399	1499	1546	1526	1514	1399	1550
Hydrogen consumption (kg/yr)	1312	1441	1459	1394	1426	1312	1463
Hydrogen autonomy (hours)	225	225	225	337	225	225	225

, it can be observed that all the systems have comparable yearly energy throughput, with values ranging from roughly 16,699 kWh/yr to 16,941 kWh/yr. This suggests that all the systems can effectively harness solar energy. Furthermore, it can be observed that the mean energy cost across these systems constantly registers at zero, indicating that the energy origin is devoid of cost or self-produced, in this case—the utilization of solar power. This attribute confers a substantial benefit in terms of economic viability. Nevertheless, there are significant differences in other important factors. With its 7.7 h autonomy, for example, DACA stands out as the most suitable option for cases when prolonged periods of energy independence are critical. Increased autonomy may ensure a reliable energy supply even during prolonged periods of reduced sunshine or grid disruptions, which might be vital in off-grid or essential power backup applications. The WAHA system demonstrates a comparatively elevated annual throughput. However, it exhibits limitations in terms of autonomy, with a duration of 3.25 h and an anticipated lifespan of 3.28 years when compared with the DACA system. This suggests that although it may possess commendable energy generation capabilities, it may not be the optimal selection for applications necessitating prolonged energy self-sufficiency or long-term reliability.

Moreover, DACA has the highest nominal capacity of 192 kWh and the longest anticipated life of 4.39 years, making it an excellent option for applications requiring extended durability and significant energy storage. The prolonged lifespan of the system can result in reduced total cost of ownership over its lifecycle since it may necessitate fewer replacements or maintenance endeavors. In contrast, both MAHA and NT exhibit comparable attributes regarding their annual throughput, energy input/output, and anticipated lifespan. This observation suggests that in cases where cost is a central consideration and a moderate level of energy production is sufficient, these systems could be viable alternatives. Nevertheless, the MAHA exhibits a slightly greater nominal capacity of 168 kWh, which may render it a somewhat superior option when evaluating storage capacity within budgetary limitations. CAVA and CAHA share similar attributes, such as annual throughput and losses. These tracking systems are especially suitable for regions that experience considerable variations in solar elevation throughout the day, such as areas characterized by distinct morning and evening sunshine. The system achieves a harmonious equilibrium between energy capture efficiency and system complexity.

3.2. Economic

The economic details of the optimal systems using the seven solar tracking configurations are given in

Table 5. Economic details of the optimal systems.

	Rank	Total Capital Cost (USD)	Total NPC (USD)	Tot. Ann. Cap. Cost (USD/yr)	Tot. Ann. Repl. Cost (USD/yr)	Total O and M Cost (USD/yr)	Total Ann. Cost (USD/yr)	Operating Cost (USD/yr)	COE (AUD/kWh)
NT	1	419,590	575,616	18,864	6215	800	25,879	7015	0.201
	2	453,600	907,033	20,393	17,522	2864	40,779	20,386	0.314
CAHA	1	546,820	781,706	24,585	7404	3157	35,145	10,560	0.271
	2	533,150	1,057,168	23,970	18,271	5288	47,529	23,559	0.369
CAVA	1	483,220	694,872	21,725	6906	2610	31,241	9516	0.241
	2	461,800	968,595	20,762	18,020	4765	43,547	22,785	0.337
DACA	1	527,135	745,770	23,700	7042	2788	33,529	9830	0.26
	2	661,350	1,214,410	29,734	18,869	5996	54,599	24,865	0.42
DAHA	1	500,620	711,696	22,507	6935	2555	31,997	9490	0.248
	2	498,575	1,003,577	22,415	17,978	4727	45,120	22,704	0.35
MAHA	1	419,590	575,616	18,864	6215	800	25,879	7015	0.201
	2	453,600	907,033	20,393	17,522	2864	40,779	20,386	0.314
WAHA	1	500,620	715,789	22,507	7105	2569	32,181	9674	0.249
	2	498,575	1,008,428	22,415	18,171	4752	45,338	22,923	0.352

. These economic attributes include total capital cost, total net present cost, total annual capital cost, total annual replacement cost, total operation and maintenance cost, total annual cost, operating cost, and the cost of energy. After the energy system simulation, HOMER uses the TNPC to select the optimal energy system; the energy system with the lowest TNPC among the technically feasible systems is selected as the optimal. Consequently, HOMER uses TNPC minimization as its objective function. Among the optimal systems, the CAHA has the highest total capital cost, total net present cost, total annual capital cost, total annual replacement cost, total operation and maintenance cost, total annual cost, and operating cost. The solar tracking configuration with the lowest COE is the MAHA and the NT, which stands at USD 0.201/kWh, while the CAHA returned the highest COE of USD 0.271/kWh. If a choice were made entirely based on TNPC, the best choices would be TN and MAHA. Although CAHA has the highest cost concerning all the economic metrics presented in Table 5, it also produced the highest total electricity production and the least capacity shortage. Conversely, the NT and MAHA, with the least COE, also produced the least energy electricity and, consequently, the highest annual unmet load. This means the economic parameters are directly proportional to the total electricity produced. If more money is spent on acquiring the CAHA solar tracking configuration instead of the NT or MAHA, the cost would be recouped/recovered from enhanced reliability (since there is a lower capacity shortage) and sales of excess electricity.

Figure 7 shows the relationship between the TNPC and COE; it is evident that there is a direct relationship between the two for each of the solar tracking alternatives. A comparison between the present and annual worth of the optimal system for each solar tracking configuration and the second-ranked system for each is also presented. The values of the present worth and annual worth for each configuration are positive; the sign of the present worth indicates whether the optimal system compares favorably as an investment option with the second-ranked system. A positive value means the optimal systems save money compared with the second-ranked systems over the project’s life. When compared with each other, the present worth and annual worth of the optimal systems are expressed as follows: $CAVA<CAHA<DAHA<WAHA<NT<MAHA<DACA$ (Figure S5). The study also considered the levelized cost of hydrogen and the levelized cost of PV. Figure S6 shows that the LCOH and LCOPV follow the same pattern for all the solar tracking configurations. Furthermore, the NT and MAHA have the lowest LCOH and LCOPV, while CAHA has the highest LCOH and LCOPV (Figure S6).

3.3. Environmental

Studies typically use emissions and renewable fractions as metrics to quantify the environmental implications of adopting energy systems. For the proposed energy systems and solar tracking configurations, the renewable fraction is 100% in all cases, as no conventional energy sources were integrated into the proposed system. Additionally, regarding emissions, none of the configurations emitted CO₂ during the operation of the energy systems (Figure 8). The significant energy emissions systems during operation are NOx and CO (Table S1). Fuel cells that operate at low temperatures, such as proton exchange membrane fuel cells (PEMFCs) or alkaline fuel cells (AFCs), typically run at temperatures below 100 degrees Celsius. The low temperatures benefit specific applications but can also present challenges, such as the potential for producing pollutants like nitrogen oxides (NOx) and carbon monoxide (CO). The leading cause of NOx and CO emissions in low-temperature fuel cells is directly linked to the electrochemical reactions at the electrodes. The CAVA and WAHA emitted the highest level of NOx, while the NT and MAHA emitted the lowest.

3.4. Multi-Criteria Decision Analysis

Traditionally, studies have adopted technical, economic, social, policy, or environmental metrics to identify the most preferred energy systems and solar tracking alternatives in the literature. To make a comprehensive decision, all aspects of the attributes that contribute to the features of the energy systems and solar tracking must be considered; although they may be conflicting, they can offer a robust alternative. The techno-economic results show that the CAHA configuration is the best in terms of electrical production. In terms of COE and TNPC, the best NT and MAHA configurations are identified as the most optimal. Regarding NOx emissions, NT and MAHA are also considered the best options for sun tracking. To combine the effects of the technical, economic, and environmental attributes in selecting the most preferred energy solutions for the ICT facility, this study employed the MCDM approach. The CODAS [48], ARAS [49], EDAS [50], and MOORA [47] MCDM methods were employed in this study, which have been extensively discussed in the literature.

The attributes considered in ranking the energy alternatives include unmet load, total NPC, total capital cost, total electrical production, PV production, NOx emissions, levelized cost of hydrogen, FC production, excess electricity, CO emissions, and capacity shortage (Table S2). During the implementation of MCDM, it is essential to allocate weights to the criteria used in the selection process. In this study, the entropy weighting method is employed to determine the weights of the criteria. The result of the entropy weighting method is given in Table S3. The results of the weight analysis show that capacity shortage and unmet load are the most significant criteria. The result of the CODAS (

Table 6. Multi-criteria decision-making results for the proposed systems.

Configuration	CODAS Score (Hi)	Rank	ARAS Score (KI)	Rank	EDAS Score (Asi)	Rank	MOORA (yi)	Rank
NT	−0.7350	5	0.1161	5	0.1535	5	−4768.03	1
CAHA	−0.0063	2	0.2736	2	0.7875	2	−7805.81	6
CAVA	3.4251	1	0.9944	1	0.9995	1	−6125.02	3
DACA	−0.7939	7	0.1013	7	2.22 ×10−5	7	−7992.55	7
DAHA	−0.5641	4	0.1549	4	0.5163	4	−6631.36	4
MAHA	−0.7350	5	0.1161	5	0.1535	5	−4768.03	1
WAHA	−0.4798	3	0.1706	3	0.5848	3	−6844.34	5

) for the ranking of the solar tracking method is given as $DACA<NT<MAHA<DAHA<WAHA<CAHA<CAVA$. The most preferred is the CAVA, and the least preferred is the DACA.

When the results of the CODAS were compared with those obtained from ARAS and EDAS, the same ranking was obtained. However, a different result was obtained compared with the results from MOORA. The NT is the best system for the MOORA method, while the least ranked from MOORA correlates with DACA. MOORA’s unique method, which focuses on the ratio of the geometric mean of the positive attributes to the geometric mean of the negative attributes, might have favored the attributes of the NT configuration, leading to the observed difference in the ranking compared with other MCDM techniques.

4. Conclusions

The objective of this study was to conduct a comprehensive assessment and comparison of different solar tracking configurations to determine their appropriateness for supplying power to an Information and Communication Technology (ICT) facility. The study provides valuable insights into the technical, economic, and environmental dimensions of these configurations, thereby enabling decision-makers to make informed choices about the most effective solution. The research examined seven solar tracking arrangements, each affecting energy generation. These configurations included Non-Tracking (NT), Monthly Adjusted Horizontal Axis (MAHA), Weekly Adjusted Horizontal Axis (WAHA), Daily Adjusted Horizontal Axis (DAHA), Continuously Adjusted Horizontal Axis (CAHA), Continuously Adjusted Vertical Axis (CAVA), and Dual Axis with Continuous Adjustment (DACA).

These solar tracking systems were evaluated for their ability to generate electricity, cost-effectiveness, and environmental impact. A systematic approach was used to achieve the research goals. HOMER was used to model and assess energy systems under different solar tracking arrangements. Technical assessments of electricity and hydrogen generation, emissions, and cost implications were implemented with HOMER. Using Multi-Criteria Decision Analysis (MCDM) methods, such as CODAS, EDAS, ARAS, and MOORA, the solar tracking arrangements and energy systems were ranked based on multiple technical, economic, and environmental criteria. The significant findings of the research include the following:

• Tracking systems enhance energy generation efficiency: In terms of overall energy production efficiency, the research found that photovoltaic (PV) tracking configurations outperformed non-tracking systems. Using tracking systems resulted in consistently better energy production, demonstrating their ability to extract more power from solar panels.
• Capacity factor variation: The capacity factor, a crucial metric for evaluating the efficacy of PV panels, varied across tracking configurations considered in this research. CAHA and DAHA configurations exhibited the highest energy production, demonstrating their capacity to maximize PV panel utilization.
• Energy storage: Solar tracking systems optimize energy production. Regarding battery energy storage, DACA offers high autonomy and capacity, ideal for reliability, while MAHA and NT are cost-effective. CAHA produces the most FC energy in September, whereas MAHA produces the least in February. DACA produces the most hydrogen in January, whereas CAHA produces the least in February.
• Load-following strategies improve reliability: Regarding capacity deficit and excess power output, configurations employing load-following algorithms performed exceptionally well. This indicates that load-following strategies improve reliability and economic sustainability by more closely matching energy supply with demand.
• Environmental impact: All configurations achieved a 100% renewable fraction, demonstrating their commitment to sustainability. Furthermore, none of the configurations emitted CO₂, indicating their environmental friendliness. However, NOx emissions varied, with CAVA and WAHA producing the most and NT and MAHA emitting the least.
• Multi-Criteria Decision Analysis (MCDM): The research utilized the multi-criteria decision-making (MCDM) methodology to evaluate and prioritize various energy choices holistically. The study’s findings indicate that the CAVA energy system emerged as the most preferred option, demonstrating its superior performance across multiple criteria categories. On the other hand, DACA was deemed the least desirable option because of its performance shortcomings compared with other options. The inconsistency in MOORA’s findings highlights the need to employ multiple decision-making frameworks to achieve a comprehensive assessment. While each approach has merit, inconsistencies between them highlight the need for a comprehensive study that considers various evaluation methods and criteria before drawing conclusions in complex scenarios, such as energy system selection.
• Solar-tracking configuration rankings (using multiple metrics): Based on electricity production, the systems are ranked as $CAHA>DAHA>CAVA>WAHA>DACA>\mathrm{“}NT=MAHA\mathrm{”}$. Based on cost of energy (COE), they are ranked as $MAHA and NT \left(lowest\right)<DACA<DAHA<WAHA<CAVA<CAHA \left(highest\right)$. Also, using the CODAS, EDAS, and ARAS methods, the systems are ranked as $CAVA>CAHA>WAHA>DAHA>NT>MAHA>DACA$.

5. Future Work

The economic analysis primarily focused on capital costs, net present value, and operational expenses; however, more comprehensive economic models that incorporate components such as financing choices, tax incentives, and energy price variations could benefit future research. These models aim to ascertain the tracking systems that offer the highest degree of cost-effectiveness over extended durations. Furthermore, future research endeavors may undertake a comprehensive lifecycle analysis (LCA) to assess the long-term environmental impacts of solar tracking systems. This study would encompass various aspects, including but not limited to energy payback time, embodied energy, and concerns about end-of-life disposal. Finally, future work should focus on experimental validation of the most promising configurations (e.g., CAHA and CAVA) in field conditions. Additionally, the integration of innovative grid technologies and real-time energy management systems will be explored to enhance operational efficiency. Further studies should also assess the adaptability of these configurations in diverse climatic regions and ICT load profiles.