Potential of Surrogate Modelling for Probabilistic Fire Analysis of Structures

Abstract

The interest in probabilistic methodologies to demonstrate structural fire safety has increased significantly in recent times. However, the evaluation of the structural behavior under fire loading is computationally expensive even for simple structural models. In this regard, machine learning-based surrogate modeling provides an appealing way forward. Surrogate models trained to simulate the behavior of structural fire engineering (SFE) models predict the response at negligible computational expense, thereby allowing for rapid probabilistic analyses and design iterations. Herein, a framework is proposed for the probabilistic analysis of fire exposed structures leveraging surrogate modeling. As a proof-of-concept a simple (analytical) non-linear model for the capacity of a concrete slab and an advanced (numerical) model for the capacity of a concrete column are considered. First, the procedure for training surrogate models is elaborated. Subsequently, the surrogate models are developed, followed by a probabilistic analysis to evaluate the probability density functions for the capacity. The results show that fragility curves developed based on the surrogate model agree with those obtained through direct sampling of the computationally expensive model, with the 10 -2 capacity quantile predicted with an error of less than 5%. Moreover, the computational cost for the probabilistic studies is significantly reduced by the adoption of surrogate models.

Cost-optimization based target reliability for fire design of insulated steel columns

“…Considering 80 cases of fire scenarios and 54 cases of d i , the probabilistic analysis involving 10 5 Monte Carlo (MC) simulations each to determine the maximum steel temperatures is computationally demanding (approximately 120,000 core hours needed). To reduce the computational expense, a Neural Network (NN) based surrogate model is adopted [21]. The optimum NN model has 4 hidden layers with 300 neurons each and has a R 2 of 0.99 (3000 samples used for training and 500 as test data set).…”

Regression-Based Surrogate Models for the Probabilistic Study of Fire Exposed Composite Structures Considering Tensile Membrane Action

“…Fig. 2 illustrates the framework applied to develop the regression-based surrogate model [17]. First, the model variables need to be selected.…”

Special Issue on “Smart Systems in Fire Engineering”

“…Brown et al [17] leveraged data obtained from wireless sensor networks embedded in fire hoses to enable smart firefighting activities. Chaudhary et al [18] combined probabilistic and surrogate modeling techniques to evaluate the fire response of concrete slabs. Finally, one letter to the editor by Gomaa et al [19] was also published.…”