Open AccessMultiple Regression and ANN ( MLP) Model for Predicting Swelling index of Ramadi Cohesive Soil
Author(s) -
Ahmed H. Abdulkareem,
Dalal O. Aziz
Publication year - 2020
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/737/1/012116
Subject(s) - oedometer test , consolidation (business) , artificial neural network , swelling , geotechnical engineering , expansive clay , soil water , regression analysis , shrinkage , soil science , environmental science , statistics , mathematics , computer science , geology , artificial intelligence , engineering , accounting , chemical engineering , business
The response of expansive soils in the form of swelling and shrinkage due to changes in water content is frequently expressed as heaving and settling of lightly loaded structures such as roads, buildings, and canal linings. Compression and/or swelling indices are used for the calculation of the consolidation settlement of fine grained soils. They are conventionally determined by laboratory oedometer tests. Because of the time and expense involved in performing consolidation tests, it is often desirable to obtain approximate values of swelling index, C s by using other soil properties which are more easily determined. A database consisting of 102 consolidation tests from different parts of Ramadi city were compiled, identified, and used to conduct and utilized a statistical study to estimate suitable relations to determine the swelling index. this study develops the artificial neural networks based multi-layer perceptron, ANN-MLP and multiple regression, MR models. Neural model offers significant improvements in prediction accuracy of the statistical models.