Open Access

ARTICLE

Smart Load Forecasting and Load Scheduling in Agriculture Irrigation Using Deep Learning Techniques

Bindu Vadlamudi¹, Subhojit Dawn^1,*, Ishwarya Devarakonda¹, Sri Hari Priya Lanka¹, Sujan Turaka¹, Taha Selim Ustun^2,*

1 Department of Electrical and Electronics Engineering, Siddhartha Academy of Higher Education, Deemed to be University, Vijayawada, Andhra Pradesh, India
2 Fukushima Renewable Energy Institute, Koriyama, Japan

* Corresponding Authors: Subhojit Dawn. Email: ; Taha Selim Ustun. Email:

Energy Engineering 2026, 123(9), 5 https://doi.org/10.32604/ee.2026.083555

Received 06 April 2026; Accepted 05 June 2026; Issue published 06 August 2026

Abstract

Agricultural irrigation consumes a large share of electricity in rural areas, creating predictable peak conditions on distribution systems that can lead to grid instability and unreliability. Classic load-forecasting and scheduling methods are time-consuming and unable to respond rapidly to fluctuating irrigation demand. Additionally, most traditional methods require a stable internet connection to function and therefore cannot readily adapt to seasonal changes or crop-specific irrigation requirements. This creates inefficiencies in energy consumption and inconsistencies in water delivery to consumers. To reduce these drawbacks, this research proposes a framework for irrigation forecasting and dynamic scheduling for agricultural systems. A hybrid deep-learning algorithm combining convolutional neural networks (CNN), bidirectional long short-term memory (BiLSTM), and particle swarm optimization (PSO) was developed. The CNN component identifies spatial features from historical electricity consumption, soil moisture, weather data, and crop growth-stage information to map when and how much water is needed. The BiLSTM component captures temporal trends in irrigation demand predicted by the CNN. Finally, the PSO component reduces simultaneous agricultural load demand and constrains scheduled demand within a predefined grid-capacity threshold, helping mitigate overloading risks in constrained supply systems. This integrated approach supports improved energy efficiency and agricultural sustainability through better irrigation practices, enhanced management of rural power systems, and increased reliability of irrigation services.

---

Keywords

---