Home / Current Issue / Paper 1717926
Energy Consumption Forecasting in Government Buildings Using Machine Learning
Subject area: Science,Engineering and Technology · Area of research: Machine Learning and Energy Forecasting
DOI: https://doi.org/10.64388/IREV9I11-1717926
Abstract
Government buildings represent a significant yet underexplored domain in machine learning-based energy forecasting. Existing studies predominantly target residential or commercial facilities, leaving a critical contextual gap for public sector infrastructure, which operates under rigid occupancy schedules, multi-department load diversity, and policy-driven operational constraints. This paper presents a structured review of 15 IEEE-indexed publications (2021–2026) on ML-based building energy consumption forecasting, analysing methodologies, datasets, model architectures, and key limitations. Thematic classification identifies five major research directions: classical regression ML, deep learning and LSTM-based approaches, hybrid and ensemble methods, IoT-integrated systems, and neuro-fuzzy approaches. Systematic gap analysis reveals ten critical deficiencies — including absence of government building datasets, exclusion of occupancy features, lack of real-time deployment validation, limited interpretability, and insufficient multi-department modelling — and maps targeted research questions and objectives. A novel intelligent occupancy-aware ML-based energy consumption forecasting framework is proposed, aimed at improving forecast accuracy, enabling near-real-time energy management, and supporting data-driven sustainability decisions in government infrastructure.
Keywords
Energy Consumption Forecasting, Government Buildings, Machine Learning, LSTM, Random Forest, Hybrid Ensemble, Occupancy-Aware Features, Smart Buildings, Time-Series Forecasting, Sustainable Energy.
How to cite this paper
@article{1717926,
author = {Akalya P, Dr. Haripriya V},
title = {Energy Consumption Forecasting in Government Buildings Using Machine Learning},
journal = {Iconic Research And Engineering Journals},
year = {2026},
volume = {9},
number = {11},
pages = {2558-2567},
issn = {2456-8880},
url = {https://www.irejournals.com/formatedpaper/1717926.pdf},
abstract = {Government buildings represent a significant yet underexplored domain in machine learning-based energy forecasting. Existing studies predominantly target residential or commercial facilities, leaving a critical contextual gap for public sector infrastructure, which operates under rigid occupancy schedules, multi-department load diversity, and policy-driven operational constraints. This paper presents a structured review of 15 IEEE-indexed publications (2021–2026) on ML-based building energy consumption forecasting, analysing methodologies, datasets, model architectures, and key limitations. Thematic classification identifies five major research directions: classical regression ML, deep learning and LSTM-based approaches, hybrid and ensemble methods, IoT-integrated systems, and neuro-fuzzy approaches. Systematic gap analysis reveals ten critical deficiencies — including absence of government building datasets, exclusion of occupancy features, lack of real-time deployment validation, limited interpretability, and insufficient multi-department modelling — and maps targeted research questions and objectives. A novel intelligent occupancy-aware ML-based energy consumption forecasting framework is proposed, aimed at improving forecast accuracy, enabling near-real-time energy management, and supporting data-driven sustainability decisions in government infrastructure.},
keywords = {Energy Consumption Forecasting, Government Buildings, Machine Learning, LSTM, Random Forest, Hybrid Ensemble, Occupancy-Aware Features, Smart Buildings, Time-Series Forecasting, Sustainable Energy.},
month = {May},
doi = {https://doi.org/10.64388/IREV9I11-1717926}
}