TY - JOUR
T1 - Explainable machine learning using real, synthetic and augmented fire tests to predict fire resistance and spalling of RC columns
AU - Naser, M. Z.
AU - Kodur, V. K.
N1 - Publisher Copyright:
© 2022 Elsevier Ltd
PY - 2022/2/15
Y1 - 2022/2/15
N2 - This paper presents the development of systematic machine learning (ML) approach to enable explainable and rapid assessment of fire resistance and fire-induced spalling of reinforced concrete (RC) columns. The developed approach comprises an ensemble of three novel ML algorithms namely; random forest (RF), extreme gradient boosted trees (ExGBT), and deep learning (DL). These algorithms are trained to account for a wide collection of geometric characteristics and material properties, as well as loading conditions to examine fire performance of normal and high strength RC columns by analyzing a comprehensive database of fire tests comprising of over 494 observations. The developed ensemble is also capable of presenting quantifiable insights to ML predictions; thus, breaking free from the notion of “black-box” ML and establishing a solid step towards transparent and explainable ML. Most importantly, this work tackles the scarcity of available fire tests by proposing new techniques to leverage the use of real, synthetic, and augmented fire test observations. The developed ML ensemble has been calibrated and validated for standard and design fire exposures and one-, two-, three- and four-sided fire exposures thus; covering a wide range of practical scenarios present during fire incidents. When fully deployed, the developed ensemble can analyze over 5,000 RC columns in under 60 s; thus, providing an attractive solution for researchers and practitioners. The presented approach can also be easily extended for evaluating fire resistance and spalling of other structural members under varying fire scenarios and loading conditions and hence paves the way to modernize the state of this research area and practice.
AB - This paper presents the development of systematic machine learning (ML) approach to enable explainable and rapid assessment of fire resistance and fire-induced spalling of reinforced concrete (RC) columns. The developed approach comprises an ensemble of three novel ML algorithms namely; random forest (RF), extreme gradient boosted trees (ExGBT), and deep learning (DL). These algorithms are trained to account for a wide collection of geometric characteristics and material properties, as well as loading conditions to examine fire performance of normal and high strength RC columns by analyzing a comprehensive database of fire tests comprising of over 494 observations. The developed ensemble is also capable of presenting quantifiable insights to ML predictions; thus, breaking free from the notion of “black-box” ML and establishing a solid step towards transparent and explainable ML. Most importantly, this work tackles the scarcity of available fire tests by proposing new techniques to leverage the use of real, synthetic, and augmented fire test observations. The developed ML ensemble has been calibrated and validated for standard and design fire exposures and one-, two-, three- and four-sided fire exposures thus; covering a wide range of practical scenarios present during fire incidents. When fully deployed, the developed ensemble can analyze over 5,000 RC columns in under 60 s; thus, providing an attractive solution for researchers and practitioners. The presented approach can also be easily extended for evaluating fire resistance and spalling of other structural members under varying fire scenarios and loading conditions and hence paves the way to modernize the state of this research area and practice.
KW - Columns
KW - Explainability, Concrete
KW - Fire resistance
KW - Machine learning
KW - Spalling
UR - http://www.scopus.com/inward/record.url?scp=85122261194&partnerID=8YFLogxK
U2 - 10.1016/j.engstruct.2021.113824
DO - 10.1016/j.engstruct.2021.113824
M3 - Article
AN - SCOPUS:85122261194
SN - 0141-0296
VL - 253
JO - Engineering Structures
JF - Engineering Structures
M1 - 113824
ER -