A machine learning approach to predict explosive spalling of heated concrete

  • PDF / 1,425,359 Bytes
  • 25 Pages / 595.276 x 790.866 pts Page_size
  • 3 Downloads / 279 Views

DOWNLOAD

REPORT


(2020) 20:134

ORIGINAL ARTICLE

A machine learning approach to predict explosive spalling of heated concrete Jin‑Cheng Liu1 · Zhigang Zhang2  Received: 18 January 2020 / Revised: 16 September 2020 / Accepted: 5 October 2020 © Wroclaw University of Science and Technology 2020

Abstract Explosive spalling is an unfavorable phenomenon observed in concrete when exposed to heating load. It is a great potential threat to safety of concrete structures subjected to accidental thermal loads. Therefore, assessing explosive spalling risk of concrete is important for fire safety design of concrete structures. This paper proposed a popular machine learning approach, i.e., artificial neural network (ANN), to assess explosive spalling risk of concrete. Besides, the decision tree method was also used to execute the same mission for a comparison purpose. Twenty-eight groups of heating tests were conducted to validate the proposed ANN model. The ANN model behaved well in assessing explosive spalling of concrete, with a prediction accuracy of 82.1%. This study shows that ANN is a promising method for adequate classification of concrete as material resistant or not resistant to thermal explosive spalling. Keywords  Concrete · Explosive spalling · High temperature · Machine learning · Artificial neural network

1 Introduction When concrete is under heating (for example, a fire), there is a risk of thermal instability that occurs in the form of spalling. Fire spalling of concrete reduces section of concrete members, exposes embedded rebar directly to fire load, and thereby accelerates failure of concrete members. The local failure of concrete structural members tends to trigger a progressive collapse of the whole structure. Therefore, understanding why fire spalling occurs in concrete and furthermore assessing fire spalling risk of concrete are two important research topics. Regarding the first topic, many studies have contributed to understanding of the underlying driving mechanism to initiate fire spalling of concrete [1–3]. There are three main types of mechanisms on fire spalling of concrete: pore * Zhigang Zhang [email protected] Jin‑Cheng Liu [email protected] 1



Department of Civil Engineering, The University of Hong Kong, Pokfulam, Hong Kong, China



Key Laboratory of New Technology for Construction of Cities in Mountain Area, Chongqing University, Ministry of Education, Chongqing 400045, China

2

pressure mechanism, thermal stress mechanism, and combined pore pressure and thermal stress mechanism. Pore pressure mechanism claims that fire spalling of concrete is caused by vapor pressure in concrete pores. Thermal stress mechanism assumes that fire spalling of concrete is induced by restraint-induced thermal stress. Combined pore pressure and thermal stress mechanism states that fire spalling of concrete is a result of combined action of vapor pressure and restraint-induced thermal stress. Despite many implemented studies, no consensus agreement on the relative contribution of the influencing factors on the fire spalling pheno