Artificial Neural Networks for the Spalling Classification & Failure Prediction Times of High Strength Concrete Columns

Abstract

This paper presents the results from two supervised Artificial Neural Networks (ANN) developed for the spalling classification and failure prediction of high strength concrete columns (HSCC) subjected to fire. The experimental test data used for the ANN are based on the HSCC tests undertaken at the Fire Research Laboratories at the University of Ulster. 80% of the chosen experimental test data was used to train the network with the remaining 20% used for testing. In the spalling classification example the key ANN input parameters were; furnace temperature, restraint, loading level, force, spalling degree, failure time and spalling type. This was also the case for the failure prediction example except for spalling type. The networks were trained using the resilient propagation algorithm. A 6-10-3 and 5-10-1 ANN architecture gave the best results for the classification and failure prediction times respectively. The results demonstrate that HSCC can be assessed using ANN.

Fire spalling behavior of high-strength concrete: A critical review

“…However, the available knowledge about the mechanisms, behavior, and parameters that affect the fire spalling of HSC is still infrequent. The readers may find parallel works using a much wider range of variables in prior investigations [284,285]. Many studies have been conducted to model HPC's spalling tendency and provide various solutions, such as active spalling prevention, materials that provide passive or alternate mixture proportions [286].…”

Review on the Use of Artificial Intelligence to Predict Fire Performance of Construction Materials and Their Flame Retardancy

“…An ANN also possesses power in predicting the performance of concrete columns exposed to high temperatures. McKinney and Ali [ 70 ] applied the supervised ANN method to classify the temperature-induced spalling phenomenon and predict the failure time of concrete columns. In this research, two neural network models with architectures of 6–10–3 and 5–10–1 (input nodes–hidden nodes–output nodes) were employed for spalling classification and failure time prediction, respectively.…”

Prediction of shear capacity of steel channel sections using machine learning algorithms

“…Erdem [44] applied the ANNs to predict the moment capacity of reinforced concrete slabs in fire. McKinney and Ali [45] have employed supervised ANNs in spalling classification and failure prediction of high strength concrete columns subjected to fire. Also, a genetic algorithm optimized back-propagation neural network was developed for the determination of flexural capacity of postfire reinforced concrete beams by Cai et al [46].…”