Estimation of Structural Damage Due To Seismic Soil Parameters Following the Northridge Earthquake

Abstract eng:
This study is based on determining critical inputs to investigate the effect of structural components and seismic parameters on the damage of existing buildings, following 1994 Northridge earthquake for the forthcoming earthquakes. Data derived from ATC 38 report - Database on the Performance of Structures Near Strong-Motion Recordings: 1994 Northridge, California Earthquake (ATC, 2000) is used in the analysis. After the 1994 Northridge earthquake, engineers inspected more than 500 buildings located in a range within 1000 feet of 30 strong motion recording stations. An artificial intelligence tool, the Artificial Neural Network (ANN) method is used to determine the damage potential of the buildings. ANN can be used to build a complex relationship between input and output, or to find patterns in data. Neural networks have the ability to model complex nonlinear relationships between input and output. Once trained, neurons are able to make new decisions, classifications, and forecasts. The complexity of building a relationship between input and output data was well handled with neural network methodology in civil engineering related problems such as slope stability, settlement analysis, prediction of pullout capacity of ground anchors, pile capacity determination, predicting of foundation settlement, evaluation of liquefaction. The input parameters utilized include plan irregularities, peak ground acceleration, epicentral distance, modified Mercalli intensity scale, number of basement floors and structural damage is the output data. Based on this artificial intelligence analysis, the important parameters and corresponding sensitivities affecting the structural damage which occurred during the Northridge earthquake is presented.

• Export as: BibTeX | MARC | MARCXML | DC | EndNote | NLM | RefWorks
• Add to your basket