A hybrid neural network method for simulating spatial variation in earthquake ground motion

Abstract Different excitations for supports should be considered for the analysis of long-span structures. The excitation of each support has time delay and spatial variation relative to other support excitations. The present study aims to propose a new method for simulating accelerograms for various distances considering spatial variation of earthquake records. The accelerograms are simulated based on response or design spectra using the learning capabilities of neural networks. In this method, the response spectrum, and the distance parameter (distance from fault rupture) are the input, and the corresponding accelerograms are the output of the network. There are three stages involved in this study. In the first stage, a replicator neural network is used as a data compressor to increase capability of the simulation. In the second stage, a radial basis function neural network is employed to generate a compressed accelerogram for a certain distance and a response spectrum. In the third stage, the compressed acceleration data is decompressed to resemble real earthquake records. Recorded accelerograms of the strong motion array in Taiwan are used to train the artificial neural network. The obtained results show the robustness of the applied method in producing spatially varying accelerograms. Finally, compatible accelerograms of the design spectrum, suggested in the Taiwan building seismic design code, are simulated for different distances with the proposed method.

[1]  Mihailo D. Trifunac,et al.  Synthetic earthquake ground motions on an array , 2013 .

[2]  Pierfrancesco Cacciola,et al.  Generation of response-spectrum-compatible artificial earthquake accelerograms with random joint time-frequency distributions , 2012 .

[4]  Jooyoung Park,et al.  Approximation and Radial-Basis-Function Networks , 1993, Neural Computation.

[5]  R. Harichandran.  Estimating the spatial variation of earthquake ground motion from dense array recordings , 1991 .

[6]  Y. Yeh,et al.  Spatial variation and stochastic modelling of seismic differential ground movement , 1988 .

[7]  H. Hao,et al.  Influence of irregular topography and random soil properties on coherency loss of spatial seismic ground motions , 2011 .

[8]  Victor Ciesielski,et al.  Anomaly Detection Using Replicator Neural Networks Trained on Examples of One Class , 2014, SEAL.

[9]  Terence D. Sanger,et al.  Optimal unsupervised learning in a single-layer linear feedforward neural network , 1989, Neural Networks.

[10]  André Preumont,et al.  The generation of non-separable artificial earthquake accelerograms for the design of nuclear power plants , 1985 .

[11]  Wen-Yeau Chang,et al.  An RBF Neural Network Combined with OLS Algorithm and Genetic Algorithm for Short-Term Wind Power Forecasting , 2013, J. Appl. Math..

[12]  Hilmi Berk Celikoglu,et al.  Application of radial basis function and generalized regression neural networks in non-linear utility function specification for travel mode choice modelling , 2006, Math. Comput. Model..

[13]  Irmela Zentner,et al.  Simulation of non-stationary conditional ground motion fields in the time domain , 2013 .

[14]  Sukhan Lee,et al.  A Gaussian potential function network with hierarchically self-organizing learning , 1991, Neural Networks.

[15]  John F. Schneider,et al.  Empirical Spatial Coherency Functions for Application to Soil-Structure Interaction Analyses , 1991 .

[16]  Jamshid Ghaboussi,et al.  Generating multiple spectrum compatible accelerograms using stochastic neural networks , 2001 .

[17]  Frederic Ward Williams,et al.  Random vibration analysis of long-span structures subjected to spatially varying ground motions , 2009 .

[18]  Michael D. Shields,et al.  Simulation of Spatially Correlated Nonstationary Response Spectrum–Compatible Ground Motion Time Histories , 2015 .

[19]  Masanobu Shinozuka,et al.  Simulation of Nonstationary Stochastic Processes by Spectral Representation , 2007 .

[20]  John G. Anderson,et al.  QUANTITATIVE MEASURE OF THE GOODNESS-OFFIT OF SYNTHETIC SEISMOGRAMS , 2002 .

[21]  Jooyoung Park,et al.  Universal Approximation Using Radial-Basis-Function Networks , 1991, Neural Computation.

[22]  Shang-Hong Lai,et al.  A hierarchical neural network algorithm for robust and automatic windowing of MR images , 2000, Artif. Intell. Medicine.

[23]  D. Wisell,et al.  Wide-band dynamic modeling of power amplifiers using radial-basis function neural networks , 2005, IEEE Transactions on Microwave Theory and Techniques.

[24]  Aspasia Zerva,et al.  Spatial Variation of Seismic Ground Motions: Modeling and Engineering Applications , 2009 .

[25]  George Deodatis,et al.  Non-stationary stochastic vector processes: seismic ground motion applications , 1996 .

[26]  Jian Wang,et al.  Short, medium and long term load forecasting model and virtual load forecaster based on radial basis function neural networks , 2010 .

[27]  Keh-Chyuan Tsai,et al.  Development of seismic force requirements for buildings in Taiwan , 2009 .

[28]  H. B. Mitchell,et al.  New simple three-layer neural network for image compression , 1997 .

[29]  Armen Der Kiureghian,et al.  Extended MSRS rule for seismic analysis of bridges subjected to differential support motions , 2011 .