Energy Load Forecast in Green Buildings Based on LSTM Deep Learning Model

Guofeng Qiang; Shu Tang

doi:10.1007/978-981-96-1965-8_37

Energy Load Forecast in Green Buildings Based on LSTM Deep Learning Model

Guofeng Qiang
, Shu Tang^*

^*Corresponding author for this work

Research output: Chapter in Book or Report/Conference proceeding › Conference Proceeding › peer-review

Abstract

Accurately predicting electrical load is crucial for effective energy management in green buildings (GB). However, the task of forecasting electricity consumption is inherently challenging due to the dynamic nature of indoor environmental changes. This study addresses this issue by employing a Long Short-Term Memory (LSTM) deep learning model to predict energy load in green buildings. By utilizing a month’s historical data encompassing temperature, humidity, and energy consumption, the LSTM model is trained to forecast energy load. The results demonstrated the effectiveness of the LSTM model in predicting energy load, with impressive performance metrics of R2 Score of 0.992 when forecasting energy load for a week. This research contributes to the field of energy management in green buildings by providing a reliable and efficient method for predicting electrical load, ultimately aiding in integrating GB into smart grids.

Original language	English
Title of host publication	Proceedings of 2024 International Conference on Smart Electrical Grid and Renewable Energy, SEGRE 2024 - Volume 2
Editors	Fushuan Wen, Haoming Liu, Huiqing Wen, Shunli Wang
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	408-413
Number of pages	6
ISBN (Print)	9789819619641
DOIs	https://doi.org/10.1007/978-981-96-1965-8_37
Publication status	Published - Mar 2025
Event	2nd International Conference on Smart Electrical Grid and Renewable Energy, SEGRE 2024 - Suzhou, China Duration: 9 Aug 2024 → 12 Aug 2024

Publication series

Name	Lecture Notes in Electrical Engineering
Volume	1336 LNEE
ISSN (Print)	1876-1100
ISSN (Electronic)	1876-1119

Conference

Conference	2nd International Conference on Smart Electrical Grid and Renewable Energy, SEGRE 2024
Country/Territory	China
City	Suzhou
Period	9/08/24 → 12/08/24

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 7 Affordable and Clean Energy

Keywords

Deep Learning
Energy Load Forecast
Green Buildings
LSTM
Time-series Prediction

Access to Document

10.1007/978-981-96-1965-8_37

Cite this

Qiang, G., & Tang, S. (2025). Energy Load Forecast in Green Buildings Based on LSTM Deep Learning Model. In F. Wen, H. Liu, H. Wen, & S. Wang (Eds.), Proceedings of 2024 International Conference on Smart Electrical Grid and Renewable Energy, SEGRE 2024 - Volume 2 (pp. 408-413). (Lecture Notes in Electrical Engineering; Vol. 1336 LNEE). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-981-96-1965-8_37

Qiang, Guofeng ; Tang, Shu. / Energy Load Forecast in Green Buildings Based on LSTM Deep Learning Model. Proceedings of 2024 International Conference on Smart Electrical Grid and Renewable Energy, SEGRE 2024 - Volume 2. editor / Fushuan Wen ; Haoming Liu ; Huiqing Wen ; Shunli Wang. Springer Science and Business Media Deutschland GmbH, 2025. pp. 408-413 (Lecture Notes in Electrical Engineering).

@inproceedings{de8fa785a590455797b63771a8a0ca99,

title = "Energy Load Forecast in Green Buildings Based on LSTM Deep Learning Model",

abstract = "Accurately predicting electrical load is crucial for effective energy management in green buildings (GB). However, the task of forecasting electricity consumption is inherently challenging due to the dynamic nature of indoor environmental changes. This study addresses this issue by employing a Long Short-Term Memory (LSTM) deep learning model to predict energy load in green buildings. By utilizing a month{\textquoteright}s historical data encompassing temperature, humidity, and energy consumption, the LSTM model is trained to forecast energy load. The results demonstrated the effectiveness of the LSTM model in predicting energy load, with impressive performance metrics of R2 Score of 0.992 when forecasting energy load for a week. This research contributes to the field of energy management in green buildings by providing a reliable and efficient method for predicting electrical load, ultimately aiding in integrating GB into smart grids.",

keywords = "Deep Learning, Energy Load Forecast, Green Buildings, LSTM, Time-series Prediction",

author = "Guofeng Qiang and Shu Tang",

note = "Publisher Copyright: {\textcopyright} The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.; 2nd International Conference on Smart Electrical Grid and Renewable Energy, SEGRE 2024 ; Conference date: 09-08-2024 Through 12-08-2024",

year = "2025",

month = mar,

doi = "10.1007/978-981-96-1965-8\_37",

language = "English",

isbn = "9789819619641",

series = "Lecture Notes in Electrical Engineering",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "408--413",

editor = "Fushuan Wen and Haoming Liu and Huiqing Wen and Shunli Wang",

booktitle = "Proceedings of 2024 International Conference on Smart Electrical Grid and Renewable Energy, SEGRE 2024 - Volume 2",

}

Qiang, G & Tang, S 2025, Energy Load Forecast in Green Buildings Based on LSTM Deep Learning Model. in F Wen, H Liu, H Wen & S Wang (eds), Proceedings of 2024 International Conference on Smart Electrical Grid and Renewable Energy, SEGRE 2024 - Volume 2. Lecture Notes in Electrical Engineering, vol. 1336 LNEE, Springer Science and Business Media Deutschland GmbH, pp. 408-413, 2nd International Conference on Smart Electrical Grid and Renewable Energy, SEGRE 2024, Suzhou, China, 9/08/24. https://doi.org/10.1007/978-981-96-1965-8_37

Energy Load Forecast in Green Buildings Based on LSTM Deep Learning Model. / Qiang, Guofeng; Tang, Shu.
Proceedings of 2024 International Conference on Smart Electrical Grid and Renewable Energy, SEGRE 2024 - Volume 2. ed. / Fushuan Wen; Haoming Liu; Huiqing Wen; Shunli Wang. Springer Science and Business Media Deutschland GmbH, 2025. p. 408-413 (Lecture Notes in Electrical Engineering; Vol. 1336 LNEE).

Research output: Chapter in Book or Report/Conference proceeding › Conference Proceeding › peer-review

TY - GEN

T1 - Energy Load Forecast in Green Buildings Based on LSTM Deep Learning Model

AU - Qiang, Guofeng

AU - Tang, Shu

N1 - Publisher Copyright: © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

PY - 2025/3

Y1 - 2025/3

N2 - Accurately predicting electrical load is crucial for effective energy management in green buildings (GB). However, the task of forecasting electricity consumption is inherently challenging due to the dynamic nature of indoor environmental changes. This study addresses this issue by employing a Long Short-Term Memory (LSTM) deep learning model to predict energy load in green buildings. By utilizing a month’s historical data encompassing temperature, humidity, and energy consumption, the LSTM model is trained to forecast energy load. The results demonstrated the effectiveness of the LSTM model in predicting energy load, with impressive performance metrics of R2 Score of 0.992 when forecasting energy load for a week. This research contributes to the field of energy management in green buildings by providing a reliable and efficient method for predicting electrical load, ultimately aiding in integrating GB into smart grids.

AB - Accurately predicting electrical load is crucial for effective energy management in green buildings (GB). However, the task of forecasting electricity consumption is inherently challenging due to the dynamic nature of indoor environmental changes. This study addresses this issue by employing a Long Short-Term Memory (LSTM) deep learning model to predict energy load in green buildings. By utilizing a month’s historical data encompassing temperature, humidity, and energy consumption, the LSTM model is trained to forecast energy load. The results demonstrated the effectiveness of the LSTM model in predicting energy load, with impressive performance metrics of R2 Score of 0.992 when forecasting energy load for a week. This research contributes to the field of energy management in green buildings by providing a reliable and efficient method for predicting electrical load, ultimately aiding in integrating GB into smart grids.

KW - Deep Learning

KW - Energy Load Forecast

KW - Green Buildings

KW - LSTM

KW - Time-series Prediction

UR - https://www.scopus.com/pages/publications/105000239559

U2 - 10.1007/978-981-96-1965-8_37

DO - 10.1007/978-981-96-1965-8_37

M3 - Conference Proceeding

AN - SCOPUS:105000239559

SN - 9789819619641

T3 - Lecture Notes in Electrical Engineering

SP - 408

EP - 413

BT - Proceedings of 2024 International Conference on Smart Electrical Grid and Renewable Energy, SEGRE 2024 - Volume 2

A2 - Wen, Fushuan

A2 - Liu, Haoming

A2 - Wen, Huiqing

A2 - Wang, Shunli

PB - Springer Science and Business Media Deutschland GmbH

T2 - 2nd International Conference on Smart Electrical Grid and Renewable Energy, SEGRE 2024

Y2 - 9 August 2024 through 12 August 2024

ER -

Qiang G, Tang S. Energy Load Forecast in Green Buildings Based on LSTM Deep Learning Model. In Wen F, Liu H, Wen H, Wang S, editors, Proceedings of 2024 International Conference on Smart Electrical Grid and Renewable Energy, SEGRE 2024 - Volume 2. Springer Science and Business Media Deutschland GmbH. 2025. p. 408-413. (Lecture Notes in Electrical Engineering). doi: 10.1007/978-981-96-1965-8_37

Energy Load Forecast in Green Buildings Based on LSTM Deep Learning Model

Abstract

Publication series

Conference

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this