Prediction of seismic-induced structural damage using artificial neural networks

Abstract

a b s t r a c tContemporary methods for estimating the extent of seismic-induced damage to structures include the use of nonlinear finite element method (FEM) and seismic vulnerability curves. FEM is applicable when a small number of predetermined structures is to be assessed, but becomes inefficient for larger stocks. Seismic vulnerability curves enable damage estimation for classes of similar structures characterised by a small number of parameters, and typically use only one parameter to describe ground motion. Hence, they are unable to extend damage prognosis to wider classes of structures, e.g. buildings with a different number of storeys and/or bays, or capture the full complexity of the relationship between damage and seismic excitation parameters. Motivated by these shortcomings, this study presents a general method for predicting seismic-induced damage using Artificial Neural Networks (ANNs). The approach was to describe both the structure and ground motion using a large number of structural and ground motion properties. The class of structures analysed were 2D reinforced concrete (RC) frames that varied in topology, stiffness, strength and damping, and were subjected to a suite of ground motions. Dynamic structural responses were simulated using nonlinear FEM analysis and damage indices describing the extent of damage calculated. Using the results of the numerical simulations, a mapping between the structural and ground motion properties and the damage indices was than established using an ANN. The performance of the ANN was assessed using several examples and the ANN was found to be capable of successfully predicting damage.

Approach to Prediction of R/C Buildings’ Seismic Damage as Pattern Recognition Problem Using Artificial Neural Networks

“…Since then, this approach has been the subject of numerous research studies (e.g. [10,11]), which led to highly important and interesting results. These results designate the ability of ANNs to predict the potential seismic damage of buildings in an approximate but generally reliable way.…”

“…This is due to the fact that differences in the performance of the networks are caused by the initial values of the synaptic weights and biases (see e.g. [11]), and also due to the random composition of the three sub-sets of the data set (training, validation and testing data sub-sets). The best trained network (optimum trained network) for each one of the different ANN architectures was finally adopted in the procedure.…”

“…This choice was based on the fact that this type of ANNs was successfully used in many published investigations which are related to the subject of the present investigation (see e.g. [10,11,16,17,18]). …”

Neurocomputing in Civil Infrastructure

“…de Lautour and Omenzetter [110] used ANN to predict seismic-induced structural damage in 2D reinforced concrete frames based on the variation of structural properties such as sti ness, strength, and damping. Zhang et al [111] used ANN and cellular automata for predicting the cracking pattern of masonry walls.…”

Quantifying the value of SHM for emergency management of bridges at-risk from seismic damage based on their performance indicators

“…Based on such analyses during the decision-making stage, relationships, e.g. utilising artificial neural networks, can be built between the identified stiffness loss, or even just recorded ground and response intensity measures like PGA and peak structural response acceleration, and failure probabilities for quick, near real-time estimation of the associated risks (de Lautour & Omenzetter, 2009). …”