An artificial intelligence approach for the prediction of maximum dry density of stabilized soil

Category Other
Group GSI.IR
Location International Geological Congress,oslo 2008
Author Alavi, Amir Hossein۱; Heshmati, Ali Akbar۱; Gandomi, Amir Hossein۲
Holding Date 16 September 2008

Maximum dry density (MDD) as a quality of stabilized soil after compaction and before curing can be used as performance criteria to evaluate soil suitability for stabilization. The value of MDD in chemical stabilization is influenced by many interrelated parameters. Due to a need to avoid extensive and cumbersome MDD experimental tests on soils on every new construction situation and also presence of complex relationships between the MDD and the influencing factors, it is idealistic to develop models to be able to simulate the behavior of the density improvement in chemical stabilization process. Hence, in the present study, we aim at using two of artificial neural network (ANN) models namely, multilayer perception (MLP) and radial basis function (RBF) in order to develop the mathematical models to be capable of predicting the MDD as one of the most important outcomes of stabilization. A large database obtained from published stabilization test results was used to develop the ANN models. The performance of these models has been subsequently analyzed and compared. Natural soil properties such as textural properties, plasticity, and stabilizer quantities and types for a wide range of soil types were used in order to generate the mathematical models.