Nagaraj et al. (2026) AI-Based Energy Demand Forecasting for a Smart Farm

Identification

• Journal: INTERANTIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT
• Year: 2026
• Date: 2026-04-10
• Authors: P. Nagaraj, Kotra Sahithi, Neha Samira, Yagnasri Ashwini, G. Vikas Reddy, V.M.Hannuma Sai
• DOI: 10.55041/ijsrem59824

Research Groups

• Department of Computer Science and Engineering, Anurag University, Hyderabad, India

Short Summary

This paper designs and implements an AI-driven Smart Farm Energy Forecasting System that integrates machine learning and deep learning for real-time energy demand prediction with smart irrigation advisory modules. The system successfully provides environmental-aware irrigation recommendations and adaptive scheduling, establishing a scalable framework for sustainable smart farm management.

Objective

• To design and implement an AI-driven Smart Farm Energy Forecasting System to address inefficient resource utilization, rising energy costs, and the lack of predictive decision-support tools in modern agriculture.

Study Configuration

• Spatial Scale: Farm-level
• Temporal Scale: Real-time forecasting and adaptive scheduling

Methodology and Data

• Models used: XGBoost and Long Short-Term Memory (LSTM) Hybrid model, Explainable AI (XAI) using SHAP values.
• Data sources: Environmental parameters including temperature, humidity, soil moisture, rainfall, and sunlight hours (likely sensor observations).

Main Results

• A fully functional AI-driven system was developed capable of real-time energy demand forecasting for smart farms.
• The system provides environmental-aware irrigation recommendations, including automated classification of irrigation types (drip, sprinkler, manual) based on environmental conditions.
• Adaptive scheduling identifies optimal irrigation periods and watering durations.
• Explainable AI (XAI) components using SHAP values enhance model interpretability by providing insights into feature contributions.
• The platform features a Streamlit-based web application for role-based user interaction, offering instant energy predictions and actionable guidance.

Contributions

• Development of a novel hybrid XGBoost and LSTM model for integrated energy demand forecasting and smart irrigation advisory in agricultural settings.
• Introduction of Explainable AI (XAI) using SHAP values to enhance the transparency and trustworthiness of AI-driven decisions in smart farming.
• Creation of a user-friendly, Streamlit-based web application enabling real-time, role-based access to energy predictions and irrigation recommendations for farmers.
• Provides a scalable framework for improving energy efficiency, reducing costs, and promoting sustainable practices in smart farm management.

Funding

• Not specified in the provided text.

Citation

@article{Nagaraj2026AIBased,
  author = {Nagaraj, P. and Sahithi, Kotra and Samira, Neha and Ashwini, Yagnasri and Reddy, G. Vikas and Sai, V.M.Hannuma},
  title = {AI-Based Energy Demand Forecasting for a Smart Farm},
  journal = {INTERANTIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT},
  year = {2026},
  doi = {10.55041/ijsrem59824},
  url = {https://doi.org/10.55041/ijsrem59824}
}