Home / Current Issue / Paper 1717982
Energy Consumption Forecasting in Organizational Buildings Using Machine Learning
Subject area: Science,Engineering and Technology · Area of research: Machine Learning
DOI: https://doi.org/10.64388/IREV9I11-1717982
Abstract
Energy consumption forecasting has become an important research area due to the increasing demand for efficient energy management in organizational buildings such as offices, universities, hospitals, and commercial infrastructures. These buildings consume a large amount of energy daily for lighting, heating, ventilation, cooling systems, and electrical equipment. Because of changing occupancy patterns, flexible working schedules, environmental conditions, and operational uncertainties, accurately predicting energy consumption has become a complex task. Traditional forecasting approaches such as statistical regression and time-series methods are simple to implement but often fail to capture complex nonlinear energy consumption patterns. In recent years, Machine Learning and Deep Learning techniques have improved forecasting performance by identifying hidden relationships between environmental conditions, occupancy behavior, and energy usage. However, existing systems still suffer from several challenges such as limited data availability, poor adaptability, lack of integration between forecasting models, and difficulty handling real-time variations. To overcome these challenges, this research proposes an integrated hybrid framework called GAN-GBRT-LSTM. The proposed system combines Generative Adversarial Networks (GAN) for synthetic data generation, Gradient Boosted Regression Trees (GBRT) for feature extraction, and Long Short-Term Memory (LSTM) networks for time-series forecasting. The framework aims to improve prediction accuracy, increase scalability, and enhance adaptability in organizational buildings. The study analyzes existing forecasting techniques, identifies major research gaps, and proposes a scalable intelligent forecasting system suitable for dynamic real-world environments. The proposed model contributes to efficient energy management, reduced operational costs, and sustainable energy utilization.
Keywords
Energy Forecasting, Organizational Buildings, Machine Learning, Deep Learning, LSTM, GAN, GBRT, Hybrid Models, Energy Management, Time-Series Prediction
References
[1] Smith et al. (2020) – Energy prediction using regression models.
[2] Lee et al. (2021) – XGBoost for load forecasting.
[3] Kumar et al. (2022) – LSTM-based energy prediction.
[4] Zhang et al. (2023) – GAN for synthetic data generation.
[5] Chen et al. (2022) – GBRT for feature extraction in energy data.
[6] Wang et al. (2024) – Hybrid deep learning models for smart buildings.
[7] Patel et al. (2021) – Machine learning approaches for energy consumption prediction.
[8] Sharma et al. (2020) – Time-series forecasting using ARIMA models.
[9] Gupta et al. (2022) – Comparative study of ML models in energy forecasting.
[10] Singh et al. (2023) – Deep learning techniques for smart energy systems.
How to cite this paper
@article{1717982,
author = {Dinesh Kumar S, Dr. Nachappa N},
title = {Energy Consumption Forecasting in Organizational Buildings Using Machine Learning},
journal = {Iconic Research And Engineering Journals},
year = {2026},
volume = {9},
number = {11},
pages = {2973-2979},
issn = {2456-8880},
url = {https://www.irejournals.com/formatedpaper/1717982.pdf},
abstract = {Energy consumption forecasting has become an important research area due to the increasing demand for efficient energy management in organizational buildings such as offices, universities, hospitals, and commercial infrastructures. These buildings consume a large amount of energy daily for lighting, heating, ventilation, cooling systems, and electrical equipment. Because of changing occupancy patterns, flexible working schedules, environmental conditions, and operational uncertainties, accurately predicting energy consumption has become a complex task.
Traditional forecasting approaches such as statistical regression and time-series methods are simple to implement but often fail to capture complex nonlinear energy consumption patterns. In recent years, Machine Learning and Deep Learning techniques have improved forecasting performance by identifying hidden relationships between environmental conditions, occupancy behavior, and energy usage. However, existing systems still suffer from several challenges such as limited data availability, poor adaptability, lack of integration between forecasting models, and difficulty handling real-time variations.
To overcome these challenges, this research proposes an integrated hybrid framework called GAN-GBRT-LSTM. The proposed system combines Generative Adversarial Networks (GAN) for synthetic data generation, Gradient Boosted Regression Trees (GBRT) for feature extraction, and Long Short-Term Memory (LSTM) networks for time-series forecasting. The framework aims to improve prediction accuracy, increase scalability, and enhance adaptability in organizational buildings.
The study analyzes existing forecasting techniques, identifies major research gaps, and proposes a scalable intelligent forecasting system suitable for dynamic real-world environments. The proposed model contributes to efficient energy management, reduced operational costs, and sustainable energy utilization.},
keywords = {Energy Forecasting, Organizational Buildings, Machine Learning, Deep Learning, LSTM, GAN, GBRT, Hybrid Models, Energy Management, Time-Series Prediction},
month = {May},
doi = {https://doi.org/10.64388/IREV9I11-1717982}
}