Location

San Diego, California

Session Start Date

5-24-2010

Session End Date

5-29-2010

Abstract

Potential functions and Fourier series method in the cylindrical coordinate system are employed to solve the problem of moving loads on the surface of a cylindrical bore in an infinite elastic and isotropic medium. The steady state dynamic equations of medium are uncoupled by applying potential functions. The medium responses are obtained by using an appropriate numerical method of Laplace transform inversion. The solution has an integral form; therefore, a feedforward backpropagation neural network is designed and trained using the response evaluated numerically in a finite set of random points to approximate stress and displacement components in the medium. It is shown that because of the super seismic nature of the problem, two mach cones are formed and opened toward the rear of the front in the medium.

Department(s)

Civil, Architectural and Environmental Engineering

Appears In

International Conferences on Recent Advances in Geotechnical Earthquake Engineering and Soil Dynamics

Meeting Name

Fifth Conference

Publisher

Missouri University of Science and Technology

Publication Date

5-24-2010

Document Version

Final Version

Rights

© 2010 Missouri University of Science and Technology, All rights reserved.

Document Type

Article - Conference proceedings

File Type

text

Language

English

Share

COinS
 
May 24th, 12:00 AM May 29th, 12:00 AM

The Effect of Step Load Moving on the Surface of a Cylindrical Cavity Using Neural Networks

San Diego, California

Potential functions and Fourier series method in the cylindrical coordinate system are employed to solve the problem of moving loads on the surface of a cylindrical bore in an infinite elastic and isotropic medium. The steady state dynamic equations of medium are uncoupled by applying potential functions. The medium responses are obtained by using an appropriate numerical method of Laplace transform inversion. The solution has an integral form; therefore, a feedforward backpropagation neural network is designed and trained using the response evaluated numerically in a finite set of random points to approximate stress and displacement components in the medium. It is shown that because of the super seismic nature of the problem, two mach cones are formed and opened toward the rear of the front in the medium.