Development and Performance Evaluation of a Solar-Powered IoT Based Irrigation Control System Using Random Forest Machine Learning for Autonomous Decision-Making in Semi-Arid Agriculture

Author

Ateequ Mustapha Salihi, Dr. Murtala Aminu Baba, Dr. Abdulkadir Abubakar Sadiq, & Ahmad Umar

Abstract

 Water scarcity and inefficient irrigation threaten agricultural productivity in semi-arid regions. This study developed and evaluated a solar-powered, IoT-based irrigation control system that integrates environmental sensing, LoRa wireless communication, and Random Forest machine learning for autonomous irrigation decision-making. A dual-controller ESP32 architecture used a primary node wired to soil moisture, temperature, humidity, pH, pressure, and water-level sensors, transmitting data via 433 MHz LoRa to a secondary gateway for local LCD display and ThingSpeak cloud upload, with ESP32-CAM providing visual monitoring. A dual-channel relay controlled separate irrigation and refill solenoid valves, while an ultrasonic sensor regulated overhead tank replenishment. The Random Forest model was developed using 5,400 raw sensor observations, with 4,800 high-quality observations retained after preprocessing. The model was optimized with 500 trees and a maximum depth of 12, achieving 91.15% accuracy, an F1-score of 82.96%, and an ROC-AUC of 97.60%; periodic retraining improved accuracy to 93.62%. The Random Forest model was trained on threshold-derived labels; reported accuracy indicates agreement with the programmed reference logic, not independent ground truth of crop water requirements. The 50 W solar panel and 7.5 Ah lithium-ion battery met the 33.50 Wh estimated daily energy requirement and provided approximately 46.4 hours of calculated design autonomy (approximately 48 hours measured in a controlled discharge test) at 80% depth-of-discharge. A 100-day maize field trial confirmed 98.4% LoRa reliability, 97.8% cloud logging success, and soil moisture maintained within stage-specific target ranges across five phenological stages. Soil moisture ranged from 35% during vegetative stress periods to 75% during peak tasseling/silking, with stage-specific means of 47.0% (germination), 53.0% (vegetative), 68.0% (tasseling/silking), 57.5% (grain fill), and 40.0% (maturity/harvest). Daily mean root-zone soil temperatures ranged from 19.0°C to 34.0°C, with an overall mean of 23.04°C, even as ambient temperatures reached 42.0°C. Soil temperatures remained below the 35°C threshold for significant maize root growth reduction. Because no control plot was used, this observation cannot be attributed solely to drip irrigation. The system provides a field-validated prototype for autonomous precision irrigation, supporting water conservation and clean-energy independence.

Keywords

Smart Irrigation; Random Forest; IoT; Solar-Powered; LoRa Communication; Precision Agriculture; Decision-Making; Maize Cultivation

Full Text:

Download Paper PDF

References

Abdulaziz, W. B., & Croock, M. S. (2022). Design of smart irrigation system for vegetable farms based on efficient wireless sensor network. Iraqi Journal of Computers, Communications, Control and Systems Engineering, 22(1), 72-85. https://doi.org/10.33103/uot.ijccce.22.1.8

Allen, R. G., Pereira, L. S., Raes, D., & Smith, M. (1998). Crop evapotranspiration: Guidelines for computing crop water requirements (FAO Irrigation and Drainage Paper No. 56). Food and Agriculture Organization of the United Nations.

Blasi, A. H., Abbadi, M. A., & Al-Huweimel, R. (2021). Machine learning approach for an automatic irrigation system in southern Jordan valley. Engineering, Technology & Applied Science Research, 11(1), 6609-6613. https://doi.org/10.48084/etasr.3944

Chandgude, V., & Kawade, B. (2023). Role of artificial intelligence and machine learning in decision making for business growth. International Journal of Advanced Research in Science, Communication and Technology, 3(1), 54-58. https://doi.org/10.48175/IJARSCT-8556

Doorenbos, J., & Kassam, A. H. (1979). Yield response to water (FAO Irrigation and Drainage Paper No. 33). Food and Agriculture Organization of the United Nations.

Eze, V. H. U., Eze, E. C., Alaneme, G. U., Bubu, P. E., Nnadi, E. O. E., & Okon, M. B. (2025). Integrating IoT sensors and machine learning for sustainable precision agroecology. Discover Agriculture, 3, Article 83. https://doi.org/10.1007/s44279-025-00247-y

Faruk, O. M., & Debnath, T. (2021). IoT-based smart irrigation system by exploiting distributed sensorial network. International Journal of Computer Science, Engineering and Applications, 11(5), 1-13. https://doi.org/10.5121/ijcsea.2021.11501

Innocent, A., Ezeji, E., & Prince, E. (2022). Design and construction of solar powered automated irrigation control system using Internet of Things (IoT). Engineering Research Journal, 2(9), 61-76. 

Lawal, A. I., Ozturk, I., Olanipekun, I. O., & Asaleye, A. J. (2020). Examining the linkages between electricity consumption and economic growth in African economies. Energy, 208, 118363. https://doi.org/10.1016/j.energy.2020.118363

Obaideen, K., Bashria, A. A. Y., Nooman, M. A., Yong, C. T., Montaser, M., Hadi, J., & Mohamad, R. (2022). An overview of smart irrigation system using IoT. Energy Nexus, 7, 100124. https://doi.org/10.1016/j.nexus.2022.100124

Pascaris, A. S., Schelly, C., Burnham, L., & Pearce, J. M. (2021). Integrated solar energy with agriculture: Industry perspective on the market, community, and socio-political dimensions of agrivoltaics. Energy Research & Social Science, 75, 102023. https://doi.org/10.1016/j.erss.2021.102023

Praveena, K. S., Bhargavi, K., Sahana, M. S., Bhanu, H. S., & Tejaswini, S. (2021). Solar power based automatic irrigation system. International Journal of Advanced Computer Research, 11(54), 61-66. https://doi.org/10.19101/IJACR.2021.1152013

Sasi, G. K., Nagaraju, G., Rohith, D., & Vasudevarao, A. (2023). Design and development of smart irrigation system using Internet of Things (IoT): A case study. Nature Environment and Pollution Technology, 22(1), 523-526. https://doi.org/10.46488/NEPT.2023.v22i01.052

Talaviya, T., Shah, D., Patel, N., Yagnik, H., & Shah, M. (2020). Implementation of artificial intelligence in agricultural for optimization of irrigation and application of pesticides and herbicides. Artificial Intelligence in Agriculture, 4, 58-73. https://doi.org/10.1016/j.aiia.2020.04.002

Uchenna, U. H., Ogubuchi, O. K., Umoren, U. E., Chinewa, N. I., & Matthew, U. O. (2024). Enhancing precision irrigation with IoT-enabled water management mobile app. HAFED POLY Journal of Science, Management and Technology, 5(2), 98-108. https://doi.org/10.4314/hpjsmt.v5i2.8

Ullah, S., Hafeez, G., Rukh, G., Albogamy, F. R., Murawwat, S., Ali, F., Khan, F. A., Khan, S., & Rehman, K. (2022). A smart sensors-based solar-powered system to monitor and control tube well for agriculture applications. Processes, 10(9), 1654. https://doi.org/10.3390/pr10081654

Vanishree, K., Nagaraja, G. S., Bhandarkar, K., Manur, S. T., Farhan, S., & Ganesh, L. (2023). LoRa based smart agriculture monitoring and automated irrigation system. Journal of Jilin University (Engineering and Technology Edition), 42(2), 199-208.

Vohra, K., Vodonos, A., Schwartz, J., Marais, E. A., Sulprizio, M. P., & Mickley, L. J. (2021). Global mortality from outdoor fine particle pollution generated by fossil fuel combustion. Environmental Research, 195, 110754. https://doi.org/10.1016/j.envres.2021.110754

Yatnalli, V., Bhusare, S. S., Shivaleelavathi, B. G., Sheetal, C., Reshma, B., Swetha, M., & Yashaswini, H. N. (2023). Design and development of solar powered automatic irrigation system for modernization of agriculture. AGRIVITA Journal of Agricultural Science, 45(1), 173-187. https://doi.org/10.17503/agrivita.v45i1.3753

Younes, A., Zouhair, E. A. E., Othmane, E. M., Dauha, E. A. E., & Ed-dahbi, A. M. (2024). The application of machine learning techniques for smart irrigation systems: A systematic literature review. Smart Agricultural Technology, 7, 100425. https://doi.org/10.1016/j.atech.2024.100425

Yusuf, S. D., Comfort, S. L. D., Umar, I., & Loko, A. Z. (2022). Simulation and construction of solar powered smart irrigation system using Internet of Things (IoT) and Blynk mobile app. Asian Journal of Agricultural and Horticultural Research, 9(4), 136-147. https://doi.org/10.9734/AJAHR/2022/v9i4202

 

 

Share your valuable work from Social Media Buttons