Seismic design of earth-retaining by artificial neural network

In this paper the seismic earth pressures is calculated by artificial neural network (ANN). The most commonly used methods to design the retaining wall under seismic conditions are the force equilibrium based pseudo- static analysis. Pseudo-dynamic analysis and displacement based sliding block methods. In the design of these retaining structures, knowledge of earth pressures under both active and passive conditions should be cleared to compute the earth forces and their point of applications. In this paper, the force based analysis method is used to compute the active and passive earth pressures under seismic conditions for retaining wall to the earthquake hazard. 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 or experimental results, the ANN learns the relationship through an adaptive training process. In this method, the time and cost of analysis is reduced substantially.

کلیدواژه ها:

Seismic design ، force based analysis ، active and passive pressure ، earth-retaining wall ، artificial neural network