Dynamically engineered multi-modal feature learning for predictions of office building cooling loads

Yiren LIU; Xiangyu ZHAO; S. Joe QIN

doi:10.1016/j.apenergy.2023.122183

Dynamically engineered multi-modal feature learning for predictions of office building cooling loads

Yiren LIU, Xiangyu ZHAO, S. Joe QIN^*

^*Corresponding author for this work

Research output: Journal Publications › Journal Article (refereed) › peer-review

2 Citations (Scopus)

Abstract

This paper reports a new knowledge-driven engineered feature learning approach in response to the Global AI Challenge for Building E&M Facilities held by the Electrical and Mechanical Service Department (EMSD) of the Hong Kong SAR. The results were awarded with a Grand Prize by the competition organizer. A dynamically engineered multi-modal feature learning (DEMMFL) method is proposed for predicting the cooling load of two office buildings. The DEMMFL model is estimated with the Lasso-ridge regression and compared with other well-known methods such as the Lasso. The novel approach applies control system knowledge to engineer useful features and explore load patterns for multi-mode modeling. Deep learning methods including LSTM, GRU, and AutoGluon are implemented for automated machine learning and tested in parallel to compare the performance of the proposed model with existing methods. The proposed model is demonstrated to predict long-term cooling load most accurately using engineered features from weather information only.

Original language	English
Article number	122183
Number of pages	15
Journal	Applied Energy
Volume	355
Early online date	15 Nov 2023
DOIs	https://doi.org/10.1016/j.apenergy.2023.122183
Publication status	Published - 1 Feb 2024

Bibliographical note

The authors gratefully acknowledge the support of the EMSD E&M AI Lab of the Hong Kong SAR for providing the data. The authors acknowledge the assistance of Yixiao Huang, Shenglong Yao, and Guo Han on the LSTM, GRU, and AutoGluon work for comparison.

Publisher Copyright:
© 2023 The Author(s)

Funding

The work described in this paper was partially supported by a grant from a General Research Fund by the Research Grants Council (RGC) of Hong Kong SAR, China (Project No. 11303421), a Collaborative Research Fund by RGC of Hong Kong (Project No. C1143-20G), a grant from the Natural Science Foundation of China (U20A20189), a grant from ITF - Guangdong-Hong Kong Technology Cooperation Funding Scheme (Project Ref. No. GHP/145/20), a Math and Application Project (2021YFA1003504) under the National Key R and D Program, a Shenzhen-Hong Kong-Macau Science and Technology Project Category C (9240086), and an InnoHK initiative of The Government of the HKSAR for the Laboratory for AI-Powered Financial Technologies .

Keywords

Feature engineering
Building energy management
Cooling load prediction
Sparse statistical learning
Automated machine learning

UN SDGs

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

Access to Document

10.1016/j.apenergy.2023.122183Licence: CC BY-NC-ND

Cite this

@article{57c263fd60224fee8868f46290cfd89f,

title = "Dynamically engineered multi-modal feature learning for predictions of office building cooling loads",

abstract = "This paper reports a new knowledge-driven engineered feature learning approach in response to the Global AI Challenge for Building E&M Facilities held by the Electrical and Mechanical Service Department (EMSD) of the Hong Kong SAR. The results were awarded with a Grand Prize by the competition organizer. A dynamically engineered multi-modal feature learning (DEMMFL) method is proposed for predicting the cooling load of two office buildings. The DEMMFL model is estimated with the Lasso-ridge regression and compared with other well-known methods such as the Lasso. The novel approach applies control system knowledge to engineer useful features and explore load patterns for multi-mode modeling. Deep learning methods including LSTM, GRU, and AutoGluon are implemented for automated machine learning and tested in parallel to compare the performance of the proposed model with existing methods. The proposed model is demonstrated to predict long-term cooling load most accurately using engineered features from weather information only.",

keywords = "Feature engineering, Building energy management, Cooling load prediction, Sparse statistical learning, Automated machine learning",

author = "Yiren LIU and Xiangyu ZHAO and QIN, {S. Joe}",

note = "The authors gratefully acknowledge the support of the EMSD E&M AI Lab of the Hong Kong SAR for providing the data. The authors acknowledge the assistance of Yixiao Huang, Shenglong Yao, and Guo Han on the LSTM, GRU, and AutoGluon work for comparison. Publisher Copyright: {\textcopyright} 2023 The Author(s)",

year = "2024",

month = feb,

day = "1",

doi = "10.1016/j.apenergy.2023.122183",

language = "English",

volume = "355",

journal = "Applied Energy",

issn = "0306-2619",

publisher = "Elsevier B.V.",

}

TY - JOUR

T1 - Dynamically engineered multi-modal feature learning for predictions of office building cooling loads

AU - LIU, Yiren

AU - ZHAO, Xiangyu

AU - QIN, S. Joe

N1 - The authors gratefully acknowledge the support of the EMSD E&M AI Lab of the Hong Kong SAR for providing the data. The authors acknowledge the assistance of Yixiao Huang, Shenglong Yao, and Guo Han on the LSTM, GRU, and AutoGluon work for comparison. Publisher Copyright: © 2023 The Author(s)

PY - 2024/2/1

Y1 - 2024/2/1

N2 - This paper reports a new knowledge-driven engineered feature learning approach in response to the Global AI Challenge for Building E&M Facilities held by the Electrical and Mechanical Service Department (EMSD) of the Hong Kong SAR. The results were awarded with a Grand Prize by the competition organizer. A dynamically engineered multi-modal feature learning (DEMMFL) method is proposed for predicting the cooling load of two office buildings. The DEMMFL model is estimated with the Lasso-ridge regression and compared with other well-known methods such as the Lasso. The novel approach applies control system knowledge to engineer useful features and explore load patterns for multi-mode modeling. Deep learning methods including LSTM, GRU, and AutoGluon are implemented for automated machine learning and tested in parallel to compare the performance of the proposed model with existing methods. The proposed model is demonstrated to predict long-term cooling load most accurately using engineered features from weather information only.

AB - This paper reports a new knowledge-driven engineered feature learning approach in response to the Global AI Challenge for Building E&M Facilities held by the Electrical and Mechanical Service Department (EMSD) of the Hong Kong SAR. The results were awarded with a Grand Prize by the competition organizer. A dynamically engineered multi-modal feature learning (DEMMFL) method is proposed for predicting the cooling load of two office buildings. The DEMMFL model is estimated with the Lasso-ridge regression and compared with other well-known methods such as the Lasso. The novel approach applies control system knowledge to engineer useful features and explore load patterns for multi-mode modeling. Deep learning methods including LSTM, GRU, and AutoGluon are implemented for automated machine learning and tested in parallel to compare the performance of the proposed model with existing methods. The proposed model is demonstrated to predict long-term cooling load most accurately using engineered features from weather information only.

KW - Feature engineering

KW - Building energy management

KW - Cooling load prediction

KW - Sparse statistical learning

KW - Automated machine learning

UR - http://www.scopus.com/inward/record.url?scp=85176964550&partnerID=8YFLogxK

U2 - 10.1016/j.apenergy.2023.122183

DO - 10.1016/j.apenergy.2023.122183

M3 - Journal Article (refereed)

SN - 0306-2619

VL - 355

JO - Applied Energy

JF - Applied Energy

M1 - 122183

ER -

Dynamically engineered multi-modal feature learning for predictions of office building cooling loads

Abstract

Bibliographical note

Funding

Keywords

UN SDGs

Access to Document

Other files and links

Fingerprint

Dimension reduction modeling methods for high dimensional dynamic data in smart manufacturing and operations (智能製造與運營系統中高維動態數據的降維建模方法)

Resilient PPE Supply Chains for Hong Kong Health Systems: Current and Post Covid-19 Pandemic (LU-Part)

Cite this

Dynamically engineered multi-modal feature learning for predictions of office building cooling loads

Abstract

Bibliographical note

Funding

Keywords

UN SDGs

Access to Document

Other files and links

Fingerprint

Projects

Dimension reduction modeling methods for high dimensional dynamic data in smart manufacturing and operations (智能製造與運營系統中高維動態數據的降維建模方法)

Resilient PPE Supply Chains for Hong Kong Health Systems: Current and Post Covid-19 Pandemic (LU-Part)

Cite this