INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,  
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)  
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue XII, December 2025  
Integrating Machine Learning for Crop and Energy Optimization in  
Agrivoltaic Gardening Systems  
Hampo, JohnPaul A.C.; Favour Ngoh Dibankap  
Catholic University of Cameroon, Bamenda  
DOI : https://doi.org/10.51583/IJLTEMAS.2025.1412000121  
Received: 30 December 2025; Accepted: 05 January 2026; Published: 14 January 2026  
ABSTRACT  
An Agrivoltaic gardening system offers a promising pathway to simultaneously addressing food security and  
renewable energy production through the integration of photovoltaic panels and crop cultivation. However,  
balancing crop productivity and energy generation remains a complex challenge due to variable microclimatic  
conditions and crop-specific responses to partial shading. This study presents a machine learning–based  
framework for optimizing both crop yield and photovoltaic energy output in small-scale agrivoltaics gardening  
systems. Using multi-source data including solar irradiance, soil moisture, temperature, panel configuration  
parameters, and crop growth indicators, supervised learning models were developed to predict crop yield and  
energy performance. Random Forest and Artificial Neural Network models demonstrated strong predictive  
capability, achieving coefficients of determination (R²) above 0.85 for crop yield estimation under shaded  
conditions. Multi-objective optimization revealed design configurations that improved combined land-use  
efficiency by up to 10% compared to conventional layouts. The results highlight the potential of machine  
learning to support adaptive design and real-time decision-making in agrivoltaics gardening, particularly in  
resource-constrained environments. This work contributes to emerging research on AI-enabled agrivoltaics and  
provides a scalable framework for sustainable food–energy co-production.  
Keywords: Machine learning, Agrivoltaics, Photovoltaic, Artificial intelligence, Internet of Things (IoT),  
Renewable energy, Agric-energy  
INTRODUCTION  
As the world tackles the intertwined challenges of climate change, food security, and renewable energy demand,  
agrivoltaics systems, which is a system where photovoltaic (PV) technology and agricultural production coexist  
on the same land; have emerged as a promising strategy to enhance land-use efficiency and sustainability  
(Domenico et al., 2025; Mehta & Zörner, 2025). By pairing solar power generation with crop cultivation,  
agrivoltaics can potentially reduce land competition between the energy and food sectors while providing  
environmental benefits such as reduced water stress and microclimate buffering for crops (Sebastia et al., 2026;  
Kumpanalaisatit et al., 2022). However, the complexity of these systems notably, particularly balancing crop  
performance and energy yield; presents significant design and operational challenges that call for robust  
predictive and optimization tools.  
Recent research recognizes that integrating Machine Learning (ML) and Artificial Intelligence (AI) into  
agrivoltaics system management could substantially improve decision support for farmers, system designers,  
and policy planners (Doubleday et al., 2025; Domenico et al., 2025; Dinesh & Pearce, 2016). The intersection  
of agronomy, solar technology, digital sensing, and data analytics promises new ways to optimize both  
agricultural and energy outcomes in dual-use landscapes such that agrivoltaics systems can reduce crop heat  
stress, lower evapotranspiration, and improve water-use efficiency, particularly in warm and semi-arid climates  
(Christos et al., 2021; Ikram & Aslam, 2024; Kai et al, 2025). However, these benefits are not uniform across  
crops or locations. Excessive shading can suppress photosynthesis, while poorly optimized panel configurations  
can reduce energy yield. Consequently, agrivoltaics gardening systems involve complex trade-offs between crop  
productivity and photovoltaic performance, hence the integration of ML and AI thus enhancing decision-making.  
www.ijltemas.in  
Page 1396  
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,  
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)  
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue XII, December 2025  
Machine Learning (ML) which involves implicit programming and progressively learning from experience after  
being trained; offers powerful tools to address this complexity. Unlike traditional rule-based or mechanistic  
models, ML algorithms notably but not limited to random forest, AdaBoost, support vector machine; can learn  
nonlinear relationships from multi-dimensional datasets, making them well suited for agrivoltaics environments  
characterized by dynamic interactions among climate, crops, and energy systems. This study presents a  
comprehensive framework for integrating ML into agrivoltaics gardening to optimize both crop yield and energy  
production, illustrating methods, opportunities, and evidence from recent studies.  
BACKGROUND AND RELATED WORK  
Agrivoltaic Gardening Systems  
Agrivoltaics has evolved from large-scale experimental farms to small-scale gardens suitable for urban, peri-  
urban, and smallholder contexts. Gardening-scale agrivoltaics typically involve lower panel heights, diverse crop  
types, and limited mechanization. Studies betwixt 2022 to 2025 highlight that such systems can achieve land  
equivalent ratios greater than one, indicating higher combined productivity than separate land uses (Dinesh &  
Pearce, 2016).  
Nevertheless, crop responses to shading remain crop-specific. Leafy vegetables such as lettuce and spinach often  
tolerate moderate shading, while fruiting crops may require more precise light management (Koch et al, 2025).  
This variability underscores the need for adaptive, data-driven optimization.  
Machine Learning in Agri-Energy Systems  
Agriculture has increasingly adopted ML for decision-making, yield prediction, irrigation management, and  
plant disease detection, as well as in energy systems for solar forecasting and performance optimization. Recent  
research demonstrates that ensemble models (XGBoost, AdaBoost for instance) and deep learning architectures  
outperform traditional regression models in predicting crop yield under variable environmental conditions  
(Sarowaret al., 2023; Parween et al., 2021).  
Despite this progress, integrated ML applications specifically targeting agrivoltaic gardening remain limited,  
creating an opportunity for interdisciplinary advancement.  
Machine Learning for Agrivoltaic Optimization  
Agri-energy system which is a broader term for the integration of renewable energy within agricultural settings  
is different from an agrivoltaic system that intentionally integrates PV or solar panel with agriculture for the co-  
production of food and energy on the same land (Tharushi et al., 2022). The previous section shows the usage  
of ML in agri-energy systems while this section focuses on agrivoltaics systems.  
Predictive Modelling of Crop Yield under Partial Shading  
Agrivoltaic systems alter sunlight distribution and microclimatic conditions, directly affecting photosynthesis,  
growth rates, and phenological development. Traditional agronomic models often struggle to capture these  
complex interactions. ML algorithms, particularly supervised learning models such as Random Forests, Gradient  
Boosting, and Neural Networks; can learn nonlinear relationships between environmental inputs (light intensity,  
temperature, soil moisture) and crop responses. (Koch et al., 2025; Malashin et al., 2024; Md Sanzid, et al.,  
2025)  
Notably, current work in related agricultural domains demonstrates that deep learning models can accurately  
predict crop performance by integrating multi-modal inputs including remote sensing and meteorological data.  
In crop yield forecasting challenges outside agrivoltaics, transformer-based architectures have shown improved  
accuracy by capturing spatial–temporal dependencies in environmental data across growing seasons. When  
adapted to agrivoltaics settings by feeding models with shading patterns, solar irradiation maps, soil sensors, and  
www.ijltemas.in  
Page 1397  
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,  
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)  
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue XII, December 2025  
historical yields; such approaches can predict crop output more reliably than traditional statistical models.  
(Melesse, 2025; Nivethithaa, 2025)  
Optimizing Solar Panel Layout and Energy Yield  
Designing agrivoltaic installations for dual productivity involves spatial decisions about panel tilt, spacing,  
height, and orientation. These parameters influence both energy capture and light distribution to plants. Machine  
Learning methods, including genetic algorithms and reinforcement learning, have the capacity to explore large  
design spaces and identify configurations that strike a balance between photovoltaic energy harvesting and crop  
light needs (Olayiwola et al., 2025; Sebastia et al, 2026; Tharushi et al., 2022).  
Emerging simulation frameworks (Sebastian et al., 2025; Ding et al., 2025) already integrate crop response  
curves with PV production models, testing dozens of configurations and performing sensitivity analyses. These  
tools can be augmented with ML-driven surrogate models to speed up optimization by predicting outcomes for  
untested design parameters based on learning from simulation datasets.  
IoT, Sensors, and Real-Time Decision Support  
Modern agrivoltaic systems increasingly employ Internet of Things (IoT) sensor networks to collect continuous  
data streams on light intensity, soil moisture, temperature, and plant physiological indicators. Coupling IoT with  
ML enables real-time adaptive decision support, such as dynamic irrigation scheduling and shade management  
(Md Sanzid et al., 2025; Nguyen et al., 2025; Parween et al., 2021). A recent study showed the feasibility of  
integrating IoT and ML for monitoring fruit ripening by analysing RGB camera data, highlighting how ML  
algorithms can enhance precision agriculture even under agrivoltaic conditions (Alsalami et al., 2025; Alsalami  
et al., 2025; Zito et al., 2024).  
Agrivoltaic Systems: Opportunities and Challenges  
An Agrivoltaic system aims to maximize land productivity by co-locating photovoltaic (solar) panels above  
crops. Studies show that agrivoltaics can influence both crop microclimates and plant growth outcomes, with  
variabilities across crop types and locations (Widmer et al., 2024; Sebastia et al., 2025; Zhang et al., 2025). Some  
crops benefit from microclimate moderation—such as reduced midday heat stress and improved water  
retention—while others may experience yield reductions under excessive shading. For example, crop yield  
responses to PV shade vary widely: some exhibit modest decreases, others maintain yields similar to traditional  
systems, and a few even thrive under partial shade conditions. These variations underscore the need for site-  
specific design and control strategies to support both crop productivity and energy generation.  
Case Studies and Evidence from Recent Research  
Although empirical adoption of ML in agrivoltaics is still emerging, foundational research in agrivoltaic  
modelling and optimization indicates clear pathways for AI integration:  
A comprehensive simulation framework concurrently analyses PV energy performance and crop  
productivity metrics across multiple design configurations, forming a basis for ML-enabled optimization.  
(Mehta & Zörner, 2025; Sebastian et al., 2025; Tharushi et al., 2022)  
Field experiments demonstrating that inter-panel spacing and other parameters can affect yields  
differently for various crops, underscoring the utility of optimized configurations informed by data-  
driven methods. (Ikram & Aslam, 2024; Doubleday et al, 2025; Channing, 2025; Sebastian et al, 2026)  
Digital twin and IoT frameworks have been proposed that predict crop growth under partial shading,  
opening opportunities for ML to refine models and improve predictive accuracy. (Zhang et al, 2025;  
Channing, 2025; Md Sanzid et al., 2025; Awais et al., 2025; Abdelali et al., 2025)  
www.ijltemas.in  
Page 1398  
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,  
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)  
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue XII, December 2025  
METHODOLOGY  
System Architecture  
Figure 1: Agricvoltaic ML-Enabled Optimization Framework  
Figure 1 illustrates the proposed ML-enabled agrivoltaic optimization framework which integrates  
environmental sensing, crop monitoring, and photovoltaic performance data into predictive and optimization  
models. The ML model, having taken inputs from environmental sensors, crop monitoring devices and PV  
system monitoring devices, supports decision through the model’s outcome. The various components in Figure  
1 are: Environmental sensors (namely solar irradiance, temperature, humidity, soil moisture), Crop monitoring  
(such as RGB imagery, phenological data), PV system monitoring (for instance power output, panel  
temperature), ML prediction and optimization layer and Decision support outputs.  
Data Collection  
Data were synthesized from recent agrivoltaic field studies and validated simulation datasets reported in  
literature between 2022 to 2025. Input variables included:  
Daily solar radiation (W/m²)  
Panel tilt angle and height  
Soil moisture (%)  
Air temperature (°C)  
Crop type and growth stage  
Measured crop yield (kg/m²)  
PV energy output (kWh)  
www.ijltemas.in  
Page 1399  
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,  
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)  
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue XII, December 2025  
Machine Learning Models  
Three ML models trained on the dataset and were evaluated:  
1. Random Forest Regression – for crop yield prediction under partial shading  
2. Artificial Neural Networks (ANN) – for modelling nonlinear crop–energy interactions  
3. Multi-objective optimization algorithms – for balancing energy yield and crop productivity  
Model performance was evaluated using R², RMSE, and mean absolute error.  
Table 1. Input Variables Used in ML Models  
Category  
Climate  
Climate  
Soil  
Variable  
Unit  
W/m²  
°C  
Solar irradiance  
Air temperature  
Soil moisture  
Panel tilt angle  
Panel height  
Leaf area index  
Crop yield  
%
PV  
Degrees  
m
PV  
Crop  
–
Output  
Output  
kg/m²  
kWh  
Energy yield  
RESULTS  
Crop Yield Prediction  
ML models achieved high predictive accuracy for crop yield under agrivoltaic conditions. Random Forest  
models consistently outperformed linear regression, capturing nonlinear shading effects.  
Figure 2: Predicted vs. observed crop yield under agrivoltaic shading  
www.ijltemas.in  
Page 1400  
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,  
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)  
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue XII, December 2025  
The Random Forest model demonstrates strong agreement with measured yields. Table 2 and figure 3 depicts  
that from the three models experimented with, random forest performs strongly having 0.86 and 0.23 for R2 and  
RMSE metrics respectively, however; the ANN model has a better R2 and RMSE metric.  
Energy Production Optimization  
Optimized panel configurations identified by the ML framework improved energy output by 6–12% compared  
to baseline fixed designs while maintaining acceptable crop yield thresholds.  
Table 2. Performance Comparison of ML Models  
Model  
R²  
RMSE  
0.48  
Linear Regression  
Random Forest  
ANN  
0.62  
0.86  
0.88  
0.23  
0.21  
Figure 3: Comparison of machine learning model performance in predicting crop yield under agrivoltaic  
conditions  
Food–Energy Trade-off Analysis  
Pareto front analysis revealed multiple optimal solutions where moderate panel spacing produced balanced  
outcomes. These results align with recent agrivoltaic field studies reporting improved land-use efficiency  
through adaptive designs.  
www.ijltemas.in  
Page 1401  
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,  
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)  
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue XII, December 2025  
Figure 4: Pareto front showing trade-offs between crop yield and energy production.  
DISCUSSION  
The results demonstrate that ML can serve as a powerful decision-support tool for agrivoltaic gardening systems.  
By integrating environmental sensing and predictive modelling, system designers can move beyond static  
designs toward adaptive configurations tailored to local conditions and crop requirements.  
Importantly, the framework supports low-cost deployment using lightweight models and basic sensors, making  
it suitable for smallholder and urban gardening contexts. The inclusion of explainable AI techniques would  
further enhance transparency and adoption among practitioners. The future work, will incorporate a  
comprehensive cost-benefit and return on investment (ROI) anaylsis, accounting for sensor procurement,  
maintenance, photovoltaic installation costs, and potential yield gains. Hence, providing a clearer guidance on  
system scalability and affordability across different socio-economic contexts. Also, since this current study  
priotizes predictive accuracy over interpretability, future iterations of the framework will integrate Explainable  
AI (XAI) techniques, such as feature importance analysis.  
CONCLUSION  
Agrivoltaics systems are a cornerstone of sustainable land-use strategies that address energy and food production  
simultaneously. However, their complexity requires advanced computational tools for effective design and  
operation. Machine Learning offers robust, adaptable, and data-driven pathways to predict crop responses to  
shading, optimize PV layouts, and enable real-time control through IoT integration. By bridging agronomy,  
renewable energy engineering, and artificial intelligence, research in this area can help realize the full potential  
of agrivoltaics for resilient and productive gardening systems.  
Strong predictive performance and optimization potential has been demonstrated by the proposed framework,  
however, several limitations must be acknowledged to contextualize the findings and guide future work. Recall  
www.ijltemas.in  
Page 1402  
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,  
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)  
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue XII, December 2025  
that synthesized and simulated datasets derived from well-established agrivoltaic and crop-growth literature is  
primarily relied on by this study.  
This study presents a comprehensive ML-based framework for optimizing crop yield and energy production in  
agrivoltaics gardening systems. Results indicate that data-driven models significantly improve predictive  
accuracy and enable balanced food–energy outcomes. As agrivoltaics continues to expand globally, integrating  
machine learning will be critical for scaling efficient, resilient, and sustainable dual-use land systems.  
Future Directions  
Integrating ML into agrivoltaics gardening research presents several promising directions:  
1. Multi-Objective Optimization: Developing ML models that simultaneously optimize crop yield and  
energy production while accounting for climatic variability and economic constraints.  
2. Transfer Learning for Crop Types: Training models on diverse agrivoltaics datasets to enable  
transferability across locations and crops with limited data availability.  
3. Explainable AI (XAI): Incorporating interpretability techniques to ensure that agrivoltaics stakeholders  
trust model outputs and understand drivers of performance.  
4. Edge AI for Remote Gardens: Implementing lightweight ML models that run on low-cost devices in rural  
or resource-limited settings to support decentralized decision-making.  
5. Long-term field validation across multiple climates: Assessing ML-enabled agrivoltaics systems over  
extended periods in diverse climatic regions to ensure reliability, robustness, and consistent performance  
under varying weather patterns and seasonal conditions.  
6. Real-time Adaptive Control Using Edge Computing: Leveraging edge-based computation to procedure  
sensor data locally and dynamically adjust agrivoltaics operations, enabling faster responses, reduced  
latency, and improved system efficiency without reliance on constant cloud connectivity.  
7. Socio-economic Assessment of ML-enabled Agrivoltaics: Analysing the social and economic impacts of  
ML agrivoltaics systems, including effects on farmer livelihoods, energy access, cost-effectiveness,  
acceptance and adoption barriers, and communal resilience.  
As research and field deployments expand, ML will likely become a central tool for agrivoltaics system design,  
management, and scaling, thus transforming how we optimize land use for both food and clean energy in a  
changing climate.  
REFERENCES  
1. Abdelali, A., Hicham, G., Nadège, T., Brahim, E, Mounir, A., et al. (2025). Digital twin technology and  
artificial intelligence in energy transition: A comprehensive systematic review of applications. Energy  
Reports, Volume 13, pp.5196 - 5218. https://doi.org/10.1016/j.egyr.2025.04.060ff. ffhal-05055048  
2. Alsalami, Z., Mohammed, G. & Srinivas, T. (2025). Enhancing Crop Yield Prediction Using IoT-Based  
Soil  
Moisture  
and  
Nutrient  
Sensors.  
SHS  
Web  
of  
Conferences,  
https://doi.org/10.1051/iciaites/2025042701029  
3. Awais, M., Wang, X., Hussain, S., Aziz, F., & Mahmood, M. Q. (2025). Advancing Precision Agriculture  
Through Digital Twins and Smart Farming Technologies: A Review. AgriEngineering, 7(5),137.  
https://doi.org/10.3390/agriengineering7050137  
4. Channing, P. (2025, August 6). Digital Twin Models for Crop Yield Prediction. Anvil Labs. URL:  
https://anvil.so/post/digital-twin-models-for-crop-yield-prediction, accessed on December 28th, 2025  
5. Channing, P. (2025, May 27). Digital Twins for Crop Yield Forecasting. Anvil Labs. URL:  
https://anvil.so/post/digital-twins-for-crop-yield-forecasting, accessed on December 29th, 2025  
www.ijltemas.in  
Page 1403  
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,  
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)  
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue XII, December 2025  
6. Christos, P., Sjoukje, O. & Ioannis, N.A. (2021). Introducing digital twins to agriculture. Computers and  
Electronics in Agriculture, Volume 184, https://doi.org/10.1016/j.compag.2020.105942  
7. Dinesh, H., & Pearce, J. M. (2016). The potential of agrivoltaic systems. Renewable and Sustainable  
Energy Reviews, volume 54, https://doi.org/10.1016/j.rser.2015.10.024  
8. Ding, H., Tao, S., Zhang, L., Li, Y., Wu, X., Zhang, J., Guo, J., Bao, E., & Cao, K. (2025). Optimization  
Design of Agrivoltaic Systems Based on Light Environment Simulation. Agriculture, 15(23), 2437.  
https://doi.org/10.3390/agriculture15232437  
9. Domenico, M., Andrea, D.Z., Claudio, P. & Sonia, L. (2025). Optimizing agrivoltaic systems: A  
comprehensive analysis of design, crop productivity and energy performance in open-field  
configurations. Applied Energy, Volume 390, https://doi.org/10.1016/j.apenergy.2025.125750  
10. Doubleday, K., Oleskewicz, K., Ovaitt, S., Hickey, T., Herbert, S. & Macknick, J. (2025). Impacts of  
year-to-year weather variability and inter-panel spacing on agrivoltaic crop yields in Massachusetts.  
Agroforestry Systems. Volume 99. https://doi.org/10.1007/s10457-025-01248-y  
11. Ikram, A. & Aslam, W. (2024). Enhancing Intercropping Yield Predictability. IEEE Access, Volume 12,  
https://doi.org/10.1109/ACCESS.2024.3486101  
12. Kai, L., Hanna, F., Chong, S.C., Thomas, H., Benny, T., Brittany, S., James, M., Julia, C. & Jordan, M.  
(2025). Comprehensive Evaluation of Agrivoltaics Research: Breadth, Depth, and Insights for Future  
Research. Energies, MDPI, vol. 18(17), pages 1-42  
13. Koch, O., Moore, J., Hörl, J. et al. (2025). Sheltered by trees: long-term yield dynamics in temperate  
alley cropping agroforestry with changing water availability. Agron. Sustain. Dev. 45(27),  
https://doi.org/10.1007/s13593-025-01022-5  
14. Kumpanalaisatit, M., Setthapun, W., Sintuya, H., Pattiya, A. & Jansri, N.S. (2022). Current Status of  
Agrivoltaic Systems and Their Benefits to Energy, Food, Environment, Economy, and Society.  
Sustainable Production and Consumption, Volume 33, https://doi.org/10.1016/j.spc.2022.08.013  
15. Malashin, I., Tynchenko, V., Gantimurov, A., Nelyub, V., Borodulin, A., & Tynchenko, Y. (2024).  
Predicting Sustainable Crop Yields: Deep Learning and Explainable AI Tools. Sustainability, 16(21),  
9437. https://doi.org/10.3390/su16219437  
16. Md Sanzid, H.E., Mahbub, E.S., Shamima, N., Abu S.A R & A.K.M. Muzahidul, I. (2025). CropSynergy:  
Harnessing  
IoT  
Solutions  
for  
Smart  
and  
Efficient  
Crop  
Management.  
Crop  
Design,  
https://doi.org/10.1016/j.cropd.2025.100127.  
17. Mehta, K., & Zörner, W. (2025). Optimizing Agri-PV System: Systematic Methodology to Assess Key  
Design Parameters. Energies, 18(14), 3877. https://doi.org/10.3390/en18143877  
18. Melesse, T.Y. (2025). Digital twin-based applications in crop monitoring. Heliyon, Volume 11(2),  
https://doi.org/10.1016/j.heliyon.2025.e42137  
19. Nguyen, D.C., Lo Thi, H.V., Liudmila, A.G. & Tran, T.T. (2025). IoT and the prospect of smart  
agriculture development. E3S Web Conf. 646 00048, https://doi.org/10.1051/e3sconf/202564600048  
20. Nivethithaa, K.K. (2025). Crop Digital Twin: A Smart Framework for Predictive and Sustainable  
Agriculture. International Research Journal of Modernization in Engineering Technology and Science.  
Volume 7(12), https://www.doi.org/10.56726/IRJMETS86475  
21. Olayiwola, O., Cali, U., Elsden, M., & Yadav, P. (2025). Enhanced Solar Photovoltaic System  
Management  
and  
Integration:  
The  
Digital  
Twin  
Concept. Solar, 5(1),  
7.  
https://doi.org/10.3390/solar5010007  
22. Parween, S., Pal, A., Snigdh, I., Kumar, V. (2021). An IoT and Machine Learning-Based Crop Prediction  
System for Precision Agriculture. In: Bora, P.K., Nandi, S., Laskar, S. (eds) Emerging Technologies for  
Smart  
Cities.  
Lecture  
Notes in Electrical  
Engineering,  
vol 765.  
Springer,  
Singapore.  
https://doi.org/10.1007/978-981-16-1550-4_2  
23. Sarowar, M.S., Falguny, B.E., Asura, K.M. & Md. Mohsin, S.R. (2023). Crop Yield Prediction: Robust  
Machine Learning Approaches for Precision Agriculture. 2023 26th International Conference on  
Computer and Information Technology (ICCIT). https://doi.org/10.1109/ICCIT60459.2023.10441634  
24. Sebastia, Z., Silvia, M.L., Yuri, B. & Pietro, E.C. (2026). Optimisation of agrivoltaic systems within the  
water-energy-food  
nexus.  
Journal  
of  
Cleaner  
Production,  
Volume  
538,  
https://doi.org/10.1016/j.jclepro.2025.147275.  
25. Sebastian, Z., Silvia, M.L., Álvaro, F., Alejandro, C., Eduardo, F.F., Tekai, E.K.Z., Erlend, H.H.,  
Magnus, M.N., Jonathan, L., Matthew, B., Max, T., Stefano, A., Shiva, G. & Pietro, E.C. (2025).  
www.ijltemas.in  
Page 1404  
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,  
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)  
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue XII, December 2025  
Modelling, simulation, and optimisation of agrivoltaic systems: A Comprehensive Review. Applied  
Energy, Volume 386, https://doi.org/10.1016/j.apenergy.2025.125558  
26. Tharushi, I.S., W.M., Pabasara U.W., Rebecca, J.Y. & Ron, W. (2022). A multi-objective optimization  
framework for building-integrated PV envelope design balancing energy and cost. Journal of Cleaner  
Production, Volume 342, https://doi.org/10.1016/j.jclepro.2022.130930  
27. Widmer, J., Christ, B., Grenz, J. & Norgrove, L. (2024). Agrivoltaics, a promising new tool for electricity  
and food production: A systematic review. Renewable and Sustainable Energy Reviews, Volume 192,  
https://doi.org/10.1016/j.rser.2023.114277  
28. Zhang, R., Zhu, H., Chang, Q., & Mao, Q. (2025). A Comprehensive Review of Digital Twins  
Technology in Agriculture. Agriculture, 15(9), 903. https://doi.org/10.3390/agriculture15090903  
29. Zito, F., Giannoccaro, N. I., Serio, R., & Strazzella, S. (2024). Analysis and Development of an IoT  
System for an Agrivoltaics Plant. Technologies, 12(7), 106.  
https://doi.org/10.3390/technologies12070106  
Ethical Statement  
This study did not involve human participants, animals, or personal data. All data used in this research were  
obtained from publicly available literature sources and simulated datasets derived from peer-reviewed  
agrivoltaics studies. No ethical approval was required for this research in accordance with institutional and  
national guidelines.  
Data Availability Statement  
The data supporting the findings of this study are derived from published sources and simulation-based datasets  
generated during the study. Derived data and model scripts used for analysis are available from the corresponding  
author upon reasonable request and if they have become available upon agreement by the authors and  
researchers. Future work will include open-access field datasets as they become available.  
www.ijltemas.in  
Page 1405