سال انتشار: ۱۳۹۰

محل انتشار: ششمین کنفرانس بین المللی زلزله شناسی و مهندسی زلزله

تعداد صفحات: ۷

نویسنده(ها):

a Heidari – Department of Civil Engineering, Shahrekord University, Shahrekord, Iran

چکیده:

Estimation of the seismic earth pressure is an important topic of research for the safe design of retaining wall in the seismic zone. Design of retaining wall needs the complete knowledge of earth pressures for both active and passive conditions. In the present work, the seismic earthpressure on retaining wall is calculated by artificial neural network (ANN). It is a common practice to consider the seismic accelerations in both horizontal and vertical directions in terms of equivalent static forces. Force-based analysis is used to compute seismic earth pressure. Then the ANN is created and the seismic earth pressures are evaluated at any conditions. A significant benefit of the ANN is its ability to learn relationships between variables with repeated exposure to those variables. Therefore, instead of deriving an analytical relationship from mathematical formulations, the ANN learns the relationship through an adaptive training process. Numerical example shows the merit of the ANN.