TSB-Forecast: A Short-Term Load Forecasting Model in Smart Cities for Integrating Time Series Embeddings and Large Language Models

In smart cities, energy management systems are essential for efficient resource utilization, enhanced operational efficiency, and sustainability promotion. This work presents a novel load forecasting model, TSB-Forecast (Time Series BERT), a hybrid machine learning model aiming to improve short-term...

Full description

Saved in:
Bibliographic Details
Main Authors: Mohamed Mahmoud Hasan, Neamat El-Tazi, Ramadan Moawad, Amany H. B. Eissa
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/11121830/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849735776855654400
author Mohamed Mahmoud Hasan
Neamat El-Tazi
Ramadan Moawad
Amany H. B. Eissa
author_facet Mohamed Mahmoud Hasan
Neamat El-Tazi
Ramadan Moawad
Amany H. B. Eissa
author_sort Mohamed Mahmoud Hasan
collection DOAJ
description In smart cities, energy management systems are essential for efficient resource utilization, enhanced operational efficiency, and sustainability promotion. This work presents a novel load forecasting model, TSB-Forecast (Time Series BERT), a hybrid machine learning model aiming to improve short-term electrical demand forecasting by integrating structured and unstructured data. The model uses Sentence Bidirectional Encoder Representations from Transformers (SBERT) to extract semantic characteristics from textual news and Time to Vector (Time2Vec) to capture temporal patterns, acquiring cyclical behavior and context-sensitive impacts. TSB-Forecast shows significant improvements reducing MAE by 38.7%, RMSE by 18.2%, and SMAPE by 50.6% compared to several baselines, including the official ENTSO-E forecast, Extra Trees Regressor (ETR), a hybrid model with basic text processing (M6), and a deep learning model combining GRU and CNN (CNN-GRU). TSB-Forecast demonstrates a stronger ability to combine time-related and contextual information, which helps reduce errors. Our contributions in this paper are fourfold and can be summarized as follows: The model initially integrates multi-source inputs, encompassing load demand, BBC News, meteorological data, and holiday information. Secondly, it utilizes Time2Vec and SBERT for improved temporal and semantic feature encoding. Subsequently, Extreme Gradient Boosting (XGBoost) and Extra Trees Regressor (ETR) collectively enhance the robustness of a stacked ensemble. Finally, a thorough empirical analysis showing consistent outperformance of the proposed model over all the forecasting measures against baseline models.
format Article
id doaj-art-e038efe8754e416d90b5f6d5857f2b8b
institution DOAJ
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-e038efe8754e416d90b5f6d5857f2b8b2025-08-20T03:07:27ZengIEEEIEEE Access2169-35362025-01-011314169414171610.1109/ACCESS.2025.359742111121830TSB-Forecast: A Short-Term Load Forecasting Model in Smart Cities for Integrating Time Series Embeddings and Large Language ModelsMohamed Mahmoud Hasan0https://orcid.org/0000-0002-6350-4762Neamat El-Tazi1https://orcid.org/0000-0003-0690-4273Ramadan Moawad2https://orcid.org/0000-0003-2349-8203Amany H. B. Eissa3https://orcid.org/0000-0001-8663-2604Department of Information Systems, Faculty of Computers and Information Technology, Future University in Egypt, Cairo, EgyptDepartment of Information Systems, Faculty of Computers and Artificial Intelligence, Cairo University, Cairo, EgyptDepartment of Computer Science, Faculty of Computers and Information Technology, Future University in Egypt, Cairo, EgyptDepartment of Information Systems, Faculty of Computers and Artificial Intelligence, Cairo University, Cairo, EgyptIn smart cities, energy management systems are essential for efficient resource utilization, enhanced operational efficiency, and sustainability promotion. This work presents a novel load forecasting model, TSB-Forecast (Time Series BERT), a hybrid machine learning model aiming to improve short-term electrical demand forecasting by integrating structured and unstructured data. The model uses Sentence Bidirectional Encoder Representations from Transformers (SBERT) to extract semantic characteristics from textual news and Time to Vector (Time2Vec) to capture temporal patterns, acquiring cyclical behavior and context-sensitive impacts. TSB-Forecast shows significant improvements reducing MAE by 38.7%, RMSE by 18.2%, and SMAPE by 50.6% compared to several baselines, including the official ENTSO-E forecast, Extra Trees Regressor (ETR), a hybrid model with basic text processing (M6), and a deep learning model combining GRU and CNN (CNN-GRU). TSB-Forecast demonstrates a stronger ability to combine time-related and contextual information, which helps reduce errors. Our contributions in this paper are fourfold and can be summarized as follows: The model initially integrates multi-source inputs, encompassing load demand, BBC News, meteorological data, and holiday information. Secondly, it utilizes Time2Vec and SBERT for improved temporal and semantic feature encoding. Subsequently, Extreme Gradient Boosting (XGBoost) and Extra Trees Regressor (ETR) collectively enhance the robustness of a stacked ensemble. Finally, a thorough empirical analysis showing consistent outperformance of the proposed model over all the forecasting measures against baseline models.https://ieeexplore.ieee.org/document/11121830/Load forecastingtime series embeddingssemantic embeddings
spellingShingle Mohamed Mahmoud Hasan
Neamat El-Tazi
Ramadan Moawad
Amany H. B. Eissa
TSB-Forecast: A Short-Term Load Forecasting Model in Smart Cities for Integrating Time Series Embeddings and Large Language Models
IEEE Access
Load forecasting
time series embeddings
semantic embeddings
title TSB-Forecast: A Short-Term Load Forecasting Model in Smart Cities for Integrating Time Series Embeddings and Large Language Models
title_full TSB-Forecast: A Short-Term Load Forecasting Model in Smart Cities for Integrating Time Series Embeddings and Large Language Models
title_fullStr TSB-Forecast: A Short-Term Load Forecasting Model in Smart Cities for Integrating Time Series Embeddings and Large Language Models
title_full_unstemmed TSB-Forecast: A Short-Term Load Forecasting Model in Smart Cities for Integrating Time Series Embeddings and Large Language Models
title_short TSB-Forecast: A Short-Term Load Forecasting Model in Smart Cities for Integrating Time Series Embeddings and Large Language Models
title_sort tsb forecast a short term load forecasting model in smart cities for integrating time series embeddings and large language models
topic Load forecasting
time series embeddings
semantic embeddings
url https://ieeexplore.ieee.org/document/11121830/
work_keys_str_mv AT mohamedmahmoudhasan tsbforecastashorttermloadforecastingmodelinsmartcitiesforintegratingtimeseriesembeddingsandlargelanguagemodels
AT neamateltazi tsbforecastashorttermloadforecastingmodelinsmartcitiesforintegratingtimeseriesembeddingsandlargelanguagemodels
AT ramadanmoawad tsbforecastashorttermloadforecastingmodelinsmartcitiesforintegratingtimeseriesembeddingsandlargelanguagemodels
AT amanyhbeissa tsbforecastashorttermloadforecastingmodelinsmartcitiesforintegratingtimeseriesembeddingsandlargelanguagemodels