Machine learning–assisted prediction of heat fluxes through thermally anisotropic building envelopes

Shen, Zhenglai; Shrestha, Som; Howard, Daniel; Feng, Tianli; Hun, Diana E.; She, Buxin

doi:10.1016/j.buildenv.2023.110157

Machine learning–assisted prediction of heat fluxes through thermally anisotropic building envelopes

Journal Article · Wed Mar 01 00:00:00 EST 2023 · Building and Environment

DOI:https://doi.org/10.1016/j.buildenv.2023.110157· OSTI ID:1961960

Shen, Zhenglai ^[1]; ^[1]; Howard, Daniel ^[1]; Feng, Tianli ^[2]; ^[1]; ^[3]

Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)
Univ. of Utah, Salt Lake City, UT (United States)
Univ. of Tennessee, Knoxville, TN (United States)

Thermally anisotropic building envelope (TABE) is a novel active building envelope that can save energy use to maintain thermal comfort in buildings by redirecting heat and coolness from building envelopes to thermal loops. Finite element models (FEMs) can be used to compute the heat fluxes through TABEs, but the high computational cost of finite element simulations has prevented parametric studies and design optimizations. This paper proposes a domain knowledge–informed, finite element–based machine learning framework to reduce the computation cost for the energy management of buildings installed with TABE that uses a ground thermal loop. First, the training heat flux data set was generated by FEM simulations with different thermal loop schedules. Then, both shallow learning models (i.e., multivariate linear regression and eXtreme Gradient Boost, or XGBoost) and a deep learning model (i.e., deep neural network, or DNN) were trained to predict the heat fluxes. Domain knowledge was used for data preprocessing and feature selection. Finally, the suitability of the selected machine learning model was tested under different thermal loop schedules. Herein, the case study results showed that: (1) XGBoost can be as accurate as DNN (coefficient of determination equal to 0.81) with much less training time; (2) the annual energy cost savings for different thermal loop schedules obtained by the XGBoost-predicted and FEM-calculated heat fluxes are consistent, having a difference of only 4%; and (3) XGBoost can reduce the computation time for the annual energy analysis of the case study building with a given thermal loop schedule from around 12 h by using FEM to less than 1 min.

View Accepted Manuscript (DOE)

Research Organization:: Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)

Sponsoring Organization:: USDOE Office of Science (SC)

Grant/Contract Number:: AC05-00OR22725

OSTI ID:: 1961960

Journal Information:: Building and Environment, Journal Name: Building and Environment Journal Issue: 1 Vol. 234; ISSN 0360-1323

Publisher:: ElsevierCopyright Statement

Country of Publication:: United States

Language:: English

References (36)

Universal Approximation Using Feedforward Neural Networks: A Survey of Some Existing Methods, and Some New Results Scarselli, Franco; Chung Tsoi, Ah Neural Networks, Vol. 11, Issue 1 https://doi.org/10.1016/S0893-6080(97)00097-X	journal	January 1998
Deep solar radiation forecasting with convolutional neural network and long short-term memory network algorithms Ghimire, Sujan; Deo, Ravinesh C.; Raj, Nawin Applied Energy, Vol. 253 https://doi.org/10.1016/j.apenergy.2019.113541	journal	November 2019
Building thermal load prediction through shallow machine learning and deep learning Wang, Zhe; Hong, Tianzhen; Piette, Mary Ann Applied Energy, Vol. 263 https://doi.org/10.1016/j.apenergy.2020.114683	journal	April 2020
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
Evaluation of a prototype active solar thermoelectric radiant wall system in winter conditions Liu, Zhongbing; Zhang, Ling; Gong, Guangcai Applied Thermal Engineering, Vol. 89 https://doi.org/10.1016/j.applthermaleng.2015.05.076	journal	October 2015
Abilities and limitations of thermal mass activation for thermal comfort, peak shifting and shaving: A review Olsthoorn, Dave; Haghighat, Fariborz; Moreau, Alain Building and Environment, Vol. 118 https://doi.org/10.1016/j.buildenv.2017.03.029	journal	June 2017
Can vacuum insulation panels be cost-effective when applied in building façades? Simões, Nuno; Gonçalves, Márcio; Serra, Catarina Building and Environment, Vol. 191 https://doi.org/10.1016/j.buildenv.2021.107602	journal	March 2021
Theoretical Design of an Active Façade System with Peltier Cells Ibáñez-Puy, María; Martín-Gómez, César; Vidaurre-Arbizu, Marina Energy Procedia, Vol. 61 https://doi.org/10.1016/j.egypro.2014.11.946	journal	January 2014
Physical energy and data-driven models in building energy prediction: A review Chen, Yongbao; Guo, Mingyue; Chen, Zhisen Energy Reports, Vol. 8 https://doi.org/10.1016/j.egyr.2022.01.162	journal	November 2022
Vacuum insulation panels for building applications: A review and beyond Baetens, Ruben; Jelle, Bjørn Petter; Thue, Jan Vincent Energy and Buildings, Vol. 42, Issue 2 https://doi.org/10.1016/j.enbuild.2009.09.005	journal	February 2010
Active pipe-embedded structures in buildings for utilizing low-grade energy sources: A review Xu, Xinhua; Wang, Shengwei; Wang, Jinbo Energy and Buildings, Vol. 42, Issue 10 https://doi.org/10.1016/j.enbuild.2010.05.002	journal	October 2010
Aerogel insulation for building applications: A state-of-the-art review Baetens, Ruben; Jelle, Bjørn Petter; Gustavsen, Arild Energy and Buildings, Vol. 43, Issue 4 https://doi.org/10.1016/j.enbuild.2010.12.012	journal	April 2011
Trees vs Neurons: Comparison between random forest and ANN for high-resolution prediction of building energy consumption Ahmad, Muhammad Waseem; Mourshed, Monjur; Rezgui, Yacine Energy and Buildings, Vol. 147 https://doi.org/10.1016/j.enbuild.2017.04.038	journal	July 2017
Modelling the relationship between heating energy use and indoor temperatures in residential buildings through Artificial Neural Networks considering occupant behavior Magalhães, Sara M. C.; Leal, Vítor M. S.; Horta, Isabel M. Energy and Buildings, Vol. 151 https://doi.org/10.1016/j.enbuild.2017.06.076	journal	September 2017
Towards adoption of building energy simulation and optimization for passive building design: A survey and a review Tian, Zhichao; Zhang, Xinkai; Jin, Xing Energy and Buildings, Vol. 158 https://doi.org/10.1016/j.enbuild.2017.11.022	journal	January 2018
Evaluation of the performance for a dynamic insulation system suitable for switchable building envelope Dabbagh, M.; Krarti, M. Energy and Buildings, Vol. 222 https://doi.org/10.1016/j.enbuild.2020.110025	journal	September 2020
Trombe wall thermal performance: Data mining techniques for indoor temperatures and heat flux forecasting Briga-Sá, Ana; Leitão, Dinis; Boaventura-Cunha, José Energy and Buildings, Vol. 252 https://doi.org/10.1016/j.enbuild.2021.111407	journal	December 2021
Building energy prediction using artificial neural networks: A literature survey Lu, Chujie; Li, Sihui; Lu, Zhengjun Energy and Buildings, Vol. 262 https://doi.org/10.1016/j.enbuild.2021.111718	journal	May 2022
Experimental evaluation of an active solar thermoelectric radiant wall system Liu, ZhongBing; Zhang, Ling; Gong, GuangCai Energy Conversion and Management, Vol. 94 https://doi.org/10.1016/j.enconman.2015.01.077	journal	April 2015
Incorporating deep learning of load predictions to enhance the optimal active energy management of combined cooling, heating and power system Zhou, Yuan; Wang, Jiangjiang; Liu, Yi Energy, Vol. 233 https://doi.org/10.1016/j.energy.2021.121134	journal	October 2021
State-of-the-art in artificial neural network applications: A survey Abiodun, Oludare Isaac; Jantan, Aman; Omolara, Abiodun Esther Heliyon, Vol. 4, Issue 11 https://doi.org/10.1016/j.heliyon.2018.e00938	journal	November 2018
Development of dynamic simplified thermal models of active pipe-embedded building envelopes using genetic algorithm Zhu, Qiuyuan; Xu, Xinhua; Wang, Jinbo International Journal of Thermal Sciences, Vol. 76 https://doi.org/10.1016/j.ijthermalsci.2013.09.008	journal	February 2014
Feature selection in machine learning: A new perspective Cai, Jie; Luo, Jiawei; Wang, Shulin Neurocomputing, Vol. 300 https://doi.org/10.1016/j.neucom.2017.11.077	journal	July 2018
Passive building energy savings: A review of building envelope components Sadineni, Suresh B.; Madala, Srikanth; Boehm, Robert F. Renewable and Sustainable Energy Reviews, Vol. 15, Issue 8 https://doi.org/10.1016/j.rser.2011.07.014	journal	October 2011
State of the art in building modelling and energy performances prediction: A review Foucquier, Aurélie; Robert, Sylvain; Suard, Frédéric Renewable and Sustainable Energy Reviews, Vol. 23 https://doi.org/10.1016/j.rser.2013.03.004	journal	July 2013
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
Active building envelope systems toward renewable and sustainable energy Luo, Yongqiang; Zhang, Ling; Bozlar, Michael Renewable and Sustainable Energy Reviews, Vol. 104 https://doi.org/10.1016/j.rser.2019.01.005	journal	April 2019
Machine learning applications in urban building energy performance forecasting: A systematic review Fathi, Soheil; Srinivasan, Ravi; Fenner, Andriel Renewable and Sustainable Energy Reviews, Vol. 133 https://doi.org/10.1016/j.rser.2020.110287	journal	November 2020
Thermal performance of the building envelope integrated with phase change material for thermal energy storage: an updated review. Rathore, Pushpendra Kumar Singh; Gupta, Naveen Kumar; Yadav, Devanand Sustainable Cities and Society, Vol. 79 https://doi.org/10.1016/j.scs.2022.103690	journal	April 2022
Prediction of undisturbed ground temperature using analytical and numerical modeling. Part I: Model development and experimental validation Xing, Lu; Spitler, Jeffrey D. Science and Technology for the Built Environment, Vol. 23, Issue 5 https://doi.org/10.1080/23744731.2016.1258371	journal	December 2016
Review of Deep Learning Algorithms and Architectures Shrestha, Ajay; Mahmood, Ausif IEEE Access, Vol. 7 https://doi.org/10.1109/ACCESS.2019.2912200	journal	January 2019
Hyperopt: A Python Library for Optimizing the Hyperparameters of Machine Learning Algorithms Bergstra, James; Yamins, Dan; Cox, David Proceedings of the Python in Science Conference https://doi.org/10.25080/Majora-8b375195-003	conference	January 2013
Heat Loss Coefficient Estimation Applied to Existing Buildings through Machine Learning Models Martínez-Comesaña, Miguel; Febrero-Garrido, Lara; Granada-Álvarez, Enrique Applied Sciences, Vol. 10, Issue 24 https://doi.org/10.3390/app10248968	journal	December 2020
Development and Validation of Numerical Models for Evaluation of Foam-Vacuum Insulation Panel Composite Boards, Including Edge Effects Biswas, Kaushik Energies, Vol. 11, Issue 9 https://doi.org/10.3390/en11092228	journal	August 2018
Thermally Anisotropic Composites for Improving the Energy Efficiency of Building Envelopes Biswas, Kaushik; Shrestha, Som; Hun, Diana Energies, Vol. 12, Issue 19 https://doi.org/10.3390/en12193783	journal	October 2019
Root mean square error (RMSE) or mean absolute error (MAE)? – Arguments against avoiding RMSE in the literature Chai, T.; Draxler, R. R. Geoscientific Model Development, Vol. 7, Issue 3 https://doi.org/10.5194/gmd-7-1247-2014	journal	January 2014

Similar Records

A Machine Learning-Assisted Framework to Control Thermally Anisotropic Building Envelopes in Residential Buildings

Conference · Wed Nov 30 23:00:00 EST 2022 · OSTI ID:1923208

Coupling thermal energy storage with a thermally anisotropic building envelope for building demand-side management across various US climate conditions

Journal Article · Mon Dec 16 19:00:00 EST 2024 · Energy and Buildings · OSTI ID:2483954

Thermally anisotropic building envelope for thermal management: finite element model calibration using field evaluation data

Journal Article · Thu Sep 19 20:00:00 EDT 2024 · Journal of Building Performance Simulation · OSTI ID:2474773

Related Subjects

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATION
Building energy management
Heat flux prediction
Machine learning
Thermally anisotropic building envelope (TABE)

Machine learning–assisted prediction of heat fluxes through thermally anisotropic building envelopes

Citation Formats

References (36)

Similar Records

Related Subjects