Determination of coefficient of lateral pressure of sandy soil at rest using results of calibration of cone penetration test and artificial neural network

The estimation of soil parameters in geotechnical practice is always an important step. Accurate prediction of sands parameters from insitu tests such as CPT is one of the most challenging problems in geotechnical engineering. In this study, using a series of reliable CPT calibration chamber test data and a system consisting of three types of neural networks, the coefficient of lateral pressure of sandy soil at rest (K0) is predicted while it has good agreement with measured data gathered in database. In this system, a series of neural networks perform some tasks and finally by strategically combining of networks, the system will be able to predict parameter (K0) with reasonable accuracy. The proposed system uses Self Organizing Map (SOM) for clustering data into training, testing and validating sets and probabilistic neural networks for classifying of sands and back propagation neural networks for conclusive function approximation. Details on the development of such a system are described in the present paper and finally results obtained by this system are compared to the available relations suggested by other researchers.