سال انتشار: ۱۳۹۱

محل انتشار: نهمین کنگره بین المللی مهندسی عمران

تعداد صفحات: ۷

نویسنده(ها):

Abdolhossein Haddad – Department of Civil Engineering, Semnan University, Iran
Marziyeh Hassan Abadi –

چکیده:

There are several methods to investigate the bearing capacity of shallow foundations. Majority of these methods are only compatible with homogeneous or up to two layer foundation soils. In reality, footingsare most likely to be founded on multi-layered soils. The most common approach to predict the bearing capacity of shallow foundations in these cases is numerical methods. As an alternative approach, an artificial neural network trained with datasets are derived from numerical methods can be used. The most advantage of this approach is that it is simpler and faster than the other methods, moreover there is noneed to have any knowledge about software and numerical method. So there is a model based on multilayer perceptrons (MLPs) which can predict the bearing capacity of foundations. Bearing capacity results obtained by MLP are compared with the predicted values of traditional methods. The results indicate that ANNs are able to predict the bearing capacity of strip footings and outperform the existing methods