Neural network estimation of duration of strong ground motion using Japanese earthquake records |
| |
Authors: | CR Arjun Ashok Kumar |
| |
Institution: | Department of Earthquake Engineering, IIT Roorkee, India |
| |
Abstract: | The duration of strong motion has a significant influence on the severity of ground shaking. In this work, a combination of average values of four geophysical properties of site (Standard Penetration Test (SPT) blow count, primary wave velocity, shear wave velocity, and density of soil) including hypocentral distance of less than 50 km and magnitudes more than 5.0 from Japanese ground motion records were used for development of neural network model, to estimate duration of strong ground motion. Since majority of strong motion databases provide only average shear wave velocity for site characterization, an attempt has also been made to train the neural network with magnitude, hypocentral distance and average shear wave velocity as three input variables. Results obtained from this study show that the duration of strong motion is mostly dependent on average shear wave velocity rather than other geophysical properties of site. |
| |
Keywords: | Duration of strong motion Geophysical properties Neural network Hypocentral distance Shear wave velocity |
本文献已被 ScienceDirect 等数据库收录! |
|