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.
Developed for studying long sequences of regularly sampled data, time series analysis methods are being increasingly investigated for the use of Structural Health Monitoring (SHM). In this research, Autoregressive (AR) models were used to fit the acceleration time histories obtained from two experimental structures: a 3-storey bookshelf structure and the ASCE Phase II Experimental SHM Benchmark Structure, in undamaged and limited number of damaged states. The coefficients of the AR models were considered to be damage sensitive features and used as input into an Artificial Neural Network (ANN). The ANN was trained to classify damage cases or estimate remaining structural stiffness. The results showed that the combination of AR models and ANNs are efficient tools for damage classification and estimation, and perform well using small number of damage sensitive features and limited sensors.
The application of time series analysis methods to structural health monitoring (SHM) is a relatively new but promising approach. This study focuses on the use of statistical pattern recognition techniques to classify damage based on analysis of the time series model coefficients. Autoregressive (AR) models were used to fit the acceleration time histories obtained from two experimental structures; a three-storey laboratory bookshelf structure and the ASCE Phase II experimental SHM benchmark structure in undamaged and various damaged states. The coefficients of the AR models were used as damage sensitive features. Principal component analysis and Sammon mapping were used to firstly obtain two-dimensional projections for quick visualization of clusters among the AR coefficients corresponding to the various damage states, and later for dimensionality reduction of data for automatic damage classifications. Data reduction based on the selection of sensors and AR coefficients was also studied. Two supervised learning algorithms, nearest neighbor classification and learning vector quantization were applied in order to systematically classify damage into states. The results showed both classification techniques were able to successfully classify damage. Nearest neighbor classificationNNC is a simple supervised pattern recognition technique [32]. Given a set of pre-selected and fixed reference or codebook vectors m i (i 5 1,y,k) corresponding to known classes, an unknown input vector x is assigned to the class which the nearest m i belongs. Several distance measures
Developed for studying long, periodic records of various measured quantities, time series analysis methods are inherently suited and offer interesting possibilities for Structural Health Monitoring (SHM) applications. However, their use in SHM can still be regarded as an emerging application and deserves more studies. In this research, Autoregressive (AR) models were used to fit experimental acceleration time histories from two experimental structural systems, a 3-storey bookshelf-type laboratory structure and the ASCE Phase II SHM Benchmark Structure, in healthy and several damaged states. The coefficients of the AR models were chosen as damage sensitive features. Preliminary visual inspection of the large, multidimensional sets of AR coefficients to check the presence of clusters corresponding to different damage severities was achieved using Sammon mapping -an efficient nonlinear data compression technique. Systematic classification of damage into states based on the analysis of the AR coefficients was achieved using two supervised classification techniques: Nearest Neighbor Classification (NNC) and Learning Vector Quantization (LVQ), and one unsupervised technique: Self-organizing Maps (SOM). This paper discusses the performance of AR coefficients as damage sensitive features and compares the efficiency of the three classification techniques using experimental data.
Time series based Structural Health Monitoring (SHM) methods are being increasingly explored. In this study, Autoregressive (AR) models were used to fit the acceleration time histories of a 3-storey laboratory structure under excitation by earthquake records in several damaged and undamaged states. The coefficients of the AR models were used as inputs into an Artificial Neural Network (ANN) and the ANN was trained to relate the AR coefficients to the damage at each storey. The results showed that proposed method was able to detect, locate and quantify the damage in the structure with a very high accuracy.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.