Search

The properties and behavior of soils depend on many factors. The interaction of individual factors is difficult to determine by traditional statistical methods due to their interdependence. The paper presents a procedure of creating an artificial neural network (ANN) model to determine the value of the damping ratio (D) of clay soils. The main purpose of this paper is to compare the appropriateness of ANN model application with empirical formulas described in the literature. The ANN model was developed using a series of laboratory tests of the damping ratio performed in the Resonance Column. Predicted values of the damping ratio of clay soils obtained from the ANN model are characterized by high convergence (coefficient of determination R2 = 0.976). In comparison with other published empirical formulas, the ANN model showed an improvement in the prediction accuracy. What is more, ANN models proved to be more flexible compared to formulas and relationships with a predetermined structure, and they were well suited to modeling the complex behavior of most geotechnical engineering materials, which, by their very nature, exhibit extreme variability. In conclusion, ANNs have the potential to predict the damping ratio (D) of clay soils and can do much better than traditional statistical techniques. [ABSTRACT FROM AUTHOR]

This paper presents a method describing the application of artificial neural networks to evaluate the change in undrained shear strength in cohesive soils due to principal stress rotation. For analysis, the results of torsional shear hollow cylinder (TSHC) tests were used. An artificial neural network with an architecture of 7–6–1 was able to predict the real value of normalized undrained shear strength, τfu/σ’v, based on soil type, over-consolidation ratio (OCR), plasticity index, IP, and the angle of principal stress rotation, α, with an average relative error of around ±3%, and a single maximum value of relative error around 6%. [ABSTRACT FROM AUTHOR]