GET THE APP

Artificial neural network for soil cohesion and soil interna | 15801
International Research Journals

International Research Journal of Agricultural Science and Soil Science

All submissions of the EM system will be redirected to Online Manuscript Submission System. Authors are requested to submit articles directly to Online Manuscript Submission System of respective journal.

Artificial neural network for soil cohesion and soil internal friction angle prediction from soil physical properties data

Abstract

Saad Abdulrahman Al-Hamed, Mohamed Fouad Wahby and Abdulwahed Mohamed Aboukarima

An artificial neural network (ANN) model was employed to predict the soil cohesion and soil internal friction angle. The soil samples were collected from different cultivated sites in seven regions in Saudi Arabia. Direct shear box method was used to determine soil cohesion and soil internal friction angle. The input factors to ANN model were soil dry density, soil moisture content and soil texture index. The best 3-layer ANN model produced correlation coefficients of 0.9328 and 0.9485 between the observed and predicted soil cohesion and soil internal friction angle, respectively during training phase. Results of using testing data showed that the ANN model gave RMSE values of 4.826 kPa and 0.928 degree for soil cohesion and soil internal friction angle, respectively indicating that ANN-based model had good accuracy in predicting soil cohesion and soil internal friction angle.

Share this article