Authors : Manikumari Nagappan; Swetha Jayamurugan; Aarthika Kudiarasumani

Volume/Issue : Volume 10 - 2025, Issue 2 - February

Google Scholar : https://tinyurl.com/28wkusye

Scribd : https://tinyurl.com/4a7j3kk9

DOI : https://doi.org/10.38124/ijisrt/25feb1206

Google Scholar

Note : A published paper may take 4-5 working days from the publication date to appear in PlumX Metrics, Semantic Scholar, and ResearchGate.

Note : Google Scholar may take 15 to 20 days to display the article.

Abstract : Water scarcity and inefficient irrigation practices pose significant challenges to sustainable agriculture, necessitating advanced solutions for optimizing water use. Accurate prediction of crop water requirements (CWR) is essential for efficient irrigation management, reducing water wastage, and improving crop yield. This study leverages machine learning (ML) models to estimate CWR by predicting Evapotranspiration (ETo) using meteorological data including temperature, humidity, solar radiation, wind speed and rainfall. Various ML algorithms, such as Long Short- Term Memory (LSTM), Support Vector Machine (SVM), Artificial Neural Networks (ANN), Random Forest (RF) and Gradient Boosting Machine (GBM) are implemented to develop predictive models. The dataset is collected from reliable meteorological sources, preprocessed using data normalization techniques and analyzed for feature importance to improve model efficiency. Hyperparameter tuning and cross-validation are applied to optimize model performance. The comparative study evaluates each model's accuracy using key metrics such as Mean Absolute Error (MAE), Root Mean Square Error (RMSE) and Coefficient of Determination (R2). The findings reveal that deep learning models like LSTM exhibit superior predictive accuracy due to their ability to capture complex temporal dependencies, while traditional ML models like SVM and RF perform efficiently with limited data. The proposed ML-driven irrigation management framework offers a data-driven approach to decision-making, enabling real- time water requirement prediction and sustainable agricultural practices. The study concludes that integrating machine learning with meteorological data can significantly enhance precision irrigation strategies, contributing to water conservation and improved crop productivity in diverse climatic conditions. Future research will explore hybrid ML models, real-time sensor integration, and cloud-based deployment for large-scale agricultural applications.

References :

1. Babu, K. R., Sai, V. H., Anusha, Y., Sai, M. V., & Venkatesh, K. (2024, May). Implementation of crop water requirement using machine learning. International Research Journal of Modernization in Engineering, Technology and Science.
2. Agyeman, B. T., Naouri, M., Appels, W., Liu, J., & Shah, S. L. (2023, June). Integrating machine learning paradigms and mixed-integer model predictive control for irrigation scheduling.
3. Khan, M. A., Islam, M. Z., & Hafeez, M. (2020, March). Data pre-processing and evaluating the performance of several data mining methods for predicting irrigation water requirement.
4. Efremova, N., Zausaev, D., & Antipov, G. (2019, June). Prediction of soil moisture content based on satellite data and sequence-to-sequence networks.