Open Access Open Access  Restricted Access Subscription Access
Open Access Open Access Open Access  Restricted Access Restricted Access Subscription Access

Ann Prediction for Shear Strength Parameters of Soils


Affiliations
1 Deptt. of Struct. Engg. & App. Mech., Samrat Ashok Technological Institute, Vidisha, M.P., India
2 Deptt. of Civil Engg., MANIT, Bhopal, India
3 Deptt. of Civil Engg., UIT, RGPV, Bhopal, India
     

   Subscribe/Renew Journal


Prediction has a long and renowned history. Throughout the historical time, the demand for prediction has not diminished. Several empirical models and procedures have been developed to predicting the Shear strength of soil. The objective of this study is to predict unconsolidated undrained shear strength parameters of sandy clay soil. Triaxial shear tests were conduct to obtain these parameters at different water content and density. The results were used to develop the artificial neural network model to predict the strength parameters. To train the neural network followings parameters were considered as input data-the compaction energy, degree of saturation, dry density and C & ∅ were output parameter. To train the network two hidden layers of twenty five and fifteen neurons with learning rate moment 0.01 and 5000 iteration. Compare the neural network method results with the results obtained by the conventional method. The neural network predictions were found to be more consistent and reliable than conventional methods.

Keywords

Angle of Internal Friction, Cohesion, Dry Density, Neural Network.
User
Subscription Login to verify subscription
Notifications
Font Size

Abstract Views: 226

PDF Views: 4




  • Ann Prediction for Shear Strength Parameters of Soils

Abstract Views: 226  |  PDF Views: 4

Authors

Rajeev Jain
Deptt. of Struct. Engg. & App. Mech., Samrat Ashok Technological Institute, Vidisha, M.P., India
Pradeep Kumar Jain
Deptt. of Civil Engg., MANIT, Bhopal, India
Sudhir Singh Bhadauria
Deptt. of Civil Engg., UIT, RGPV, Bhopal, India

Abstract


Prediction has a long and renowned history. Throughout the historical time, the demand for prediction has not diminished. Several empirical models and procedures have been developed to predicting the Shear strength of soil. The objective of this study is to predict unconsolidated undrained shear strength parameters of sandy clay soil. Triaxial shear tests were conduct to obtain these parameters at different water content and density. The results were used to develop the artificial neural network model to predict the strength parameters. To train the neural network followings parameters were considered as input data-the compaction energy, degree of saturation, dry density and C & ∅ were output parameter. To train the network two hidden layers of twenty five and fifteen neurons with learning rate moment 0.01 and 5000 iteration. Compare the neural network method results with the results obtained by the conventional method. The neural network predictions were found to be more consistent and reliable than conventional methods.

Keywords


Angle of Internal Friction, Cohesion, Dry Density, Neural Network.