Performance evaluation of automated data-driven feature extraction and selection methods for practical and scalable building energy consumption prediction models

Kim, Janghyun; Frank, Stephen; Buechler, Robert; Mishra, Sakshi; Petersen, Anya; Zhang, Liang; Eslinger, Hannah

doi:10.1016/j.jobe.2025.112045

Title: Performance evaluation of automated data-driven feature extraction and selection methods for practical and scalable building energy consumption prediction models

Journal Article · Fri Feb 14 19:00:00 EST 2025 · Journal of Building Engineering

DOI: https://doi.org/10.1016/j.jobe.2025.112045 · OSTI ID:2526219

^[1];

^[2];

^[3]; Petersen, Anya ^[1];

^[4]; Eslinger, Hannah ^[1]

National Renewable Energy Laboratory (NREL), Golden, CO (United States)
Stanford University, CA (United States)
Microsoft, Bellevue, WA (United States)
University of Arizona, Tucson, AZ (United States)

Here, this study quantifies the impact of automated feature engineering methods (feature extraction and selection) on the quality and accuracy of machine learning models that predict building energy consumption. The case study compares model performance for three main scenarios: baseline (no feature extraction and selection), feature extraction only, and feature extraction combined with feature selection (filter and/or wrapper methods) for fully trained machine learning models for 200 metered/sub-metered energy measurements across 118 real buildings. For consistency, the same machine learning model architecture (a black box deep learning neural network with probabilistic forecast output) was used for all scenarios. Based on results, all feature engineering methods provided noticeable prediction accuracy improvements (e.g., 29%-68% median prediction improvement) compared to baseline scenarios. However, in this application, feature selection methods provide little practical value due to their limited performance gains and high computational cost. Smarter algorithm development supported by better computational environments will be needed before feature selection methods can reliably and efficiently improve predictive model performance.

View Accepted Manuscript (DOE)

Cite

Export

Save

Research Organization:: National Renewable Energy Laboratory (NREL), Golden, CO (United States)

Sponsoring Organization:: USDOE Office of Energy Efficiency and Renewable Energy (EERE), Energy Efficiency Office. Building Technologies Office

Grant/Contract Number:: AC36-08GO28308

OSTI ID:: 2526219

Report Number(s):: NREL/JA--5500-89615; MainId:90394; UUID:e2c5cda2-f54c-463e-a86d-0ba264dbeed7; MainAdminId:76233

Journal Information:: Journal of Building Engineering, Journal Name: Journal of Building Engineering Vol. 103; ISSN 2352-7102

Publisher:: ElsevierCopyright Statement

Country of Publication:: United States

Language:: English

References (30)

Non-crossing nonlinear regression quantiles by monotone composite quantile regression neural network, with application to rainfall extremes Cannon, Alex J. Stochastic Environmental Research and Risk Assessment, Vol. 32, Issue 11 https://doi.org/10.1007/s00477-018-1573-6	journal	June 2018
Accuracy of automated measurement and verification (M&V) techniques for energy savings in commercial buildings Granderson, Jessica; Touzani, Samir; Custodio, Claudine Applied Energy, Vol. 173 https://doi.org/10.1016/j.apenergy.2016.04.049	journal	July 2016
Probabilistic individual load forecasting using pinball loss guided LSTM Wang, Yi; Gan, Dahua; Sun, Mingyang Applied Energy, Vol. 235 https://doi.org/10.1016/j.apenergy.2018.10.078	journal	February 2019
A review of machine learning in building load prediction Zhang, Liang; Wen, Jin; Li, Yanfei Applied Energy, Vol. 285 https://doi.org/10.1016/j.apenergy.2021.116452	journal	March 2021
A review of data-driven fault detection and diagnostics for building HVAC systems Chen, Zhelun; O’Neill, Zheng; Wen, Jin Applied Energy, Vol. 339 https://doi.org/10.1016/j.apenergy.2023.121030	journal	June 2023
All you need to know about model predictive control for buildings Drgoňa, Ján; Arroyo, Javier; Cupeiro Figueroa, Iago Annual Reviews in Control, Vol. 50 https://doi.org/10.1016/j.arcontrol.2020.09.001	journal	January 2020
Feature selection strategy for machine learning methods in building energy consumption prediction Qiao, Qingyao; Yunusa-Kaltungo, Akilu; Edwards, Rodger E. Energy Reports, Vol. 8 https://doi.org/10.1016/j.egyr.2022.10.125	journal	November 2022
A comprehensive overview on the data driven and large scale based approaches for forecasting of building energy demand: A review Ahmad, Tanveer; Chen, Huanxin; Guo, Yabin Energy and Buildings, Vol. 165 https://doi.org/10.1016/j.enbuild.2018.01.017	journal	April 2018
Research on systematic optimization methods for chilled water systems in a high-rise office building Deng, Jiewen; He, Shi; Wei, Qingpeng Energy and Buildings, Vol. 209 https://doi.org/10.1016/j.enbuild.2019.109695	journal	February 2020
Accuracy of different machine learning algorithms and added-value of predicting aggregated-level energy performance of commercial buildings Walker, Shalika; Khan, Waqas; Katic, Katarina Energy and Buildings, Vol. 209 https://doi.org/10.1016/j.enbuild.2019.109705	journal	February 2020
A review of the-state-of-the-art in data-driven approaches for building energy prediction Sun, Ying; Haghighat, Fariborz; Fung, Benjamin C. M. Energy and Buildings, Vol. 221 https://doi.org/10.1016/j.enbuild.2020.110022	journal	August 2020
Assessment of Model-Based peak electric consumption prediction for commercial buildings Granderson, Jessica; Sharma, Mrinalini; Crowe, Eliot Energy and Buildings, Vol. 245 https://doi.org/10.1016/j.enbuild.2021.111031	journal	August 2021
Investigating the performance of machine learning models combined with different feature selection methods to estimate the energy consumption of buildings Liu, Xue; Tang, Hao; Ding, Yong Energy and Buildings, Vol. 273 https://doi.org/10.1016/j.enbuild.2022.112408	journal	October 2022
Research on systematic analysis and optimization method for chillers based on model predictive control: A case study Deng, Jiewen; Qiang, Wenbo; Peng, Chenwei Energy and Buildings, Vol. 285 https://doi.org/10.1016/j.enbuild.2023.112916	journal	April 2023
Flexible demand-side management program in accordance with the consumers’ requested constraints Atefi, Ahmad; Gholaminia, Vahid Energy and Buildings, Vol. 309 https://doi.org/10.1016/j.enbuild.2024.114013	journal	April 2024
Optimal management of microgrid energy by considering demand side management plan and maintenance cost with developed particle swarm algorithm Sun, Hua; Cui, Xuan; Latifi, Hassan Electric Power Systems Research, Vol. 231 https://doi.org/10.1016/j.epsr.2024.110312	journal	June 2024
Data fusion strategies for energy efficiency in buildings: Overview, challenges and novel orientations Himeur, Yassine; Alsalemi, Abdullah; Al-Kababji, Ayman Information Fusion, Vol. 64 https://doi.org/10.1016/j.inffus.2020.07.003	journal	December 2020
Principles, research status, and prospects of feature engineering for data-driven building energy prediction: A comprehensive review Wang, Zeyu; Xia, Lisha; Yuan, Hongping Journal of Building Engineering, Vol. 58 https://doi.org/10.1016/j.jobe.2022.105028	journal	October 2022
A review on the prediction of building energy consumption Zhao, Hai-xiang; Magoulès, Frédéric Renewable and Sustainable Energy Reviews, Vol. 16, Issue 6 https://doi.org/10.1016/j.rser.2012.02.049	journal	August 2012
A review of artificial intelligence based building energy use prediction: Contrasting the capabilities of single and ensemble prediction models Wang, Zeyu; Srinivasan, Ravi S. Renewable and Sustainable Energy Reviews, Vol. 75 https://doi.org/10.1016/j.rser.2016.10.079	journal	August 2017
Building electrical energy consumption forecasting analysis using conventional and artificial intelligence methods: A review Mat Daut, Mohammad Azhar; Hassan, Mohammad Yusri; Abdullah, Hayati Renewable and Sustainable Energy Reviews, Vol. 70 https://doi.org/10.1016/j.rser.2016.12.015	journal	April 2017
A review and analysis of regression and machine learning models on commercial building electricity load forecasting Yildiz, B.; Bilbao, J. I.; Sproul, A. B. Renewable and Sustainable Energy Reviews, Vol. 73 https://doi.org/10.1016/j.rser.2017.02.023	journal	June 2017
A review on time series forecasting techniques for building energy consumption Deb, Chirag; Zhang, Fan; Yang, Junjing Renewable and Sustainable Energy Reviews, Vol. 74 https://doi.org/10.1016/j.rser.2017.02.085	journal	July 2017
A review of data-driven building energy consumption prediction studies Amasyali, Kadir; El-Gohary, Nora M. Renewable and Sustainable Energy Reviews, Vol. 81 https://doi.org/10.1016/j.rser.2017.04.095	journal	January 2018
A review of data-driven approaches for prediction and classification of building energy consumption Wei, Yixuan; Zhang, Xingxing; Shi, Yong Renewable and Sustainable Energy Reviews, Vol. 82 https://doi.org/10.1016/j.rser.2017.09.108	journal	February 2018
Modeling and forecasting building energy consumption: A review of data-driven techniques Bourdeau, Mathieu; Zhai, Xiao qiang; Nefzaoui, Elyes Sustainable Cities and Society, Vol. 48 https://doi.org/10.1016/j.scs.2019.101533	journal	July 2019
The Building Data Genome Project 2, energy meter data from the ASHRAE Great Energy Predictor III competition Miller, Clayton; Kathirgamanathan, Anjukan; Picchetti, Bianca Scientific Data, Vol. 7, Issue 1 https://doi.org/10.1038/s41597-020-00712-x	journal	October 2020
Machine learning for estimation of building energy consumption and performance: a review Seyedzadeh, Saleh; Rahimian, Farzad Pour; Glesk, Ivan Visualization in Engineering, Vol. 6, Issue 1 https://doi.org/10.1186/s40327-018-0064-7	journal	October 2018
Forecasting Energy Use in Buildings Using Artificial Neural Networks: A Review Runge, Jason; Zmeureanu, Radu Energies, Vol. 12, Issue 17 https://doi.org/10.3390/en12173254	journal	August 2019
Feature Selection: A literature Review Kumar, Vipin The Smart Computing Review, Vol. 4, Issue 3 https://doi.org/10.6029/smartcr.2014.03.007	journal	June 2014

Similar Records

Gradient boosting machine for modeling the energy consumption of commercial buildings

Journal Article · Sat Nov 25 19:00:00 EST 2017 · Energy and Buildings · OSTI ID:1439233

Touzani, Samir; Granderson, Jessica; Fernandes, Samuel

A systematic feature extraction and selection framework for data-driven whole-building automated fault detection and diagnostics in commercial buildings

Journal Article · Mon Sep 28 20:00:00 EDT 2020 · Building and Environment · OSTI ID:1710174

Zhang, Liang; Frank, Stephen; Kim, Janghyun; +2 more

A review of machine learning in building load prediction

Journal Article · Sun Feb 28 23:00:00 EST 2021 · Applied Energy · OSTI ID:1762408

Liang, Zhang; Wen, Jin; Li, Yanfei; +4 more

Related Subjects

97 MATHEMATICS AND COMPUTING
Building
Feature engineering
Feature extraction
Feature selection
Load prediction
Machine learning

Title: Performance evaluation of automated data-driven feature extraction and selection methods for practical and scalable building energy consumption prediction models

Citation Formats

References (30)

Similar Records

Related Subjects