Estimation of spectral acceleration based on neural networks

This study presents an effective method based on artificial intelligence to predict spectral acceleration from the data of the ‘Next generation attenuation' project. The proposed method uses the learning abilities of artificial neural networks to expand the knowledge of the mapping from earthquake parameters to spectral accelerations, which results in spectral accelerations for special frequencies. In this paper, the Levenberg–Marquardt algorithm is applied for training neural networks. For each type of faults, considered earthquake parameters included in the input of the neural network are moment magnitude, distance from the recording site to epicentre, hypocentre depth and average shear-wave velocity in the top 30 m. To test the adequacy of the trained neural networks, they were examined through their training set and new data. The obtained results demonstrate that neural networks can be effectively and reliably employed in the estimation of spectral acceleration.

[1]  Pierre-Yves Bard,et al.  Adapting the Neural Network Approach to PGA Prediction: An Example Based on the KiK-net Data , 2012 .

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

[3]  I. Ahmad,et al.  Neural Network Based Attenuation of Strong Motion Peaks in Europe , 2008 .

[4]  K. Campbell,et al.  NGA Ground Motion Model for the Geometric Mean Horizontal Component of PGA, PGV, PGD and 5% Damped Linear Elastic Response Spectra for Periods Ranging from 0.01 to 10 s , 2008 .

[5]  Hamza Güllü,et al.  A neural network approach for attenuation relationships: An application using strong ground motion data from Turkey , 2007 .

[6]  BrianS-J. Chiou,et al.  An NGA Model for the Average Horizontal Component of Peak Ground Motion and Response Spectra , 2008 .

[7]  Tienfuan Kerh,et al.  Neural network estimation of ground peak acceleration at stations along Taiwan high-speed rail system , 2005, Eng. Appl. Artif. Intell..

[8]  N. Abrahamson,et al.  Summary of the Abrahamson & Silva NGA Ground-Motion Relations , 2008 .

[9]  Mohammad Bagher Menhaj,et al.  Training feedforward networks with the Marquardt algorithm , 1994, IEEE Trans. Neural Networks.

[10]  G. Atkinson,et al.  Ground-Motion Prediction Equations for the Average Horizontal Component of PGA, PGV, and 5%-Damped PSA at Spectral Periods between 0.01 s and 10.0 s , 2008 .

[11]  Miguel P. Romo,et al.  Estimation of peak ground accelerations for Mexican subduction zone earthquakes using neural networks , 2007 .

[12]  W. B. Joyner,et al.  Methods for regression analysis of strong-motion data , 1993, Bulletin of the Seismological Society of America.

[13]  Jamshid Ghaboussi,et al.  New method of generating spectrum compatible accelerograms using neural networks , 1998 .

[14]  Abdollah Bagheri,et al.  Application of wavelet multiresolution analysis and artificial intelligence for generation of artificial earthquake accelerograms , 2008 .

[15]  G. Ghodrati Amiri,et al.  Generation of Multiple Earthquake Accelerograms Compatible with Spectrum Via the Wavelet Packet Transform and Stochastic Neural Networks , 2009 .

[16]  Abdollah Bagheri,et al.  Application of neural networks and an adapted wavelet packet for generating artificial ground motion , 2011 .

[17]  Nick Gregor,et al.  NGA Project Strong-Motion Database , 2008 .

[18]  Maurice S. Power,et al.  An Overview of the NGA Project , 2008 .

[19]  J. Douglas.  Earthquake ground motion estimation using strong-motion records: a review of equations for the estimation of peak ground acceleration and response spectral ordinates , 2003 .