Title: Prediction of temperature distribution and fire resistance of RC slab using artificial neural networks

Authors: Mariam Farouk Ghazy; Metwally Abd Allah Abd Elaty; Nagat Mahmoud Zalhaf

Addresses: Department of Structural Engineering, Faculty of Engineering, Tanta University, Tanta, Egypt ' Department of Structural Engineering, Faculty of Engineering, Tanta University, Tanta, Egypt ' Department of Structural Engineering, Faculty of Engineering, Kafrelsheikh University, Kafrelsheikh, Egypt

Abstract: This paper present the use of artificial neural networks as alternative approach for predicting the temperature distribution of RC slab and its fire resistance according thermal criteria (the temperature of reinforcement criteria and the temperature of unexposed surface criteria). The predicted fire resistance compared with target fire resistance in codes of EN 1992-1-2 and ACI 216.1. Two set of data are used in training and testing two different ANNs. Dataset 1 which consists of temperature profile for siliceous aggregate concrete slab presented in EN 1992-1-2. Dataset 2 represents the temperature profile for carbonate concrete slab presented in ACI 216.1. Two ANNs models have been constructed. The first ANN is used to predict the temperature distribution in RC section. Second ANNs have been used to predict the fire resistance of RC slab. These models are used to study the effect of different parameters include aggregate type, slab thickness, concrete cover and reinforcement type on the fire resistance of RC slab. The results showed that ANNs can predict the temperature in RC slab section and its fire resistance with a good accuracy. Also, the ANNs models are succeed in predicting the effect of different parameters in the fire resistance of RC slab.

Keywords: prediction; temperature distribution; fire resistance; RC slab; artificial neural network; ANN.

DOI: 10.1504/IJSTRUCTE.2021.112084

International Journal of Structural Engineering, 2021 Vol.11 No.1, pp.1 - 18

Received: 26 Nov 2019
Accepted: 14 Mar 2020
Published online: 04 Jan 2021 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article