Estimation of Soil Shear Strength Parameters from Index Properties Using ANN the Case of Addis Ababa
No Thumbnail Available
Date
2022-05
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Addis Ababa University
Abstract
Shear strength of a soil is perhaps the most important of its engineering properties, as stability
analyses in the field of geotechnical engineering are dependent on it. This research work seeks to
develop models for predicting the shear strength parameters (cohesion and angle of friction) of
soils in Addis Ababa city using artificial neural network modeling technique; with a view to
reducing time, effort and cost usually incurred in determining these shear strength parameters in
the laboratory for future planning, design and construction projects in the study area. An attempt
has been made to develop separate neural network models for c and ϕ from the index properties of
soil consisting of Sand % (SP), Fines % (FP), Liquid limit (LL), Plasticity Index (PI), water content
(ω), and Bulk density (BD) as input parameters. A multi-layer perceptron network with feed
forward back propagation is used to model varying the number of hidden layers. For this purpose,
284 soil test result data was used. The geotechnical soil properties were determined in accordance
with ASTM Standards. Direct shear box method was used to determine soil cohesion and soil
internal friction angle. The developed models were found to be quite satisfactory in predicting
shear strength parameters with correlation coefficients of about 0.98 and 0.92 for cohesion and
angle of internal friction, respectively during the testing phase. The models are validated by
primary soil test data and compared with some existing correlation methods. The result showed
that the artificial neural network method gave better fit and accuracy than the selected empirical
formulae in the prediction of shear strength parameters.
Description
Keywords
ANN, Shear Strength, Cohesion, Friction Angle, Prediction, Index Property