%0 Journal Article
%T Artificial Neural Networks (ANNS) For Prediction of Engineering Properties of Soils
%A Dr.CH. Sudha Rani
%A Phani Kumar Vaddi
%A N.V.Vamsi Krishna Togati
%J International Journal of Innovative Technology and Exploring Engineering
%D 2013
%I IJITEE
%X The behaviour of soil at the location of the project and interactions of the earth materials during and after construction has a major influence on the success, economy and safety of the work. Another complexity associated with some geotechnical engineering materials, such as sand and gravel, is the difficulty in obtaining undisturbed samples and time consuming involving skilled technician. Shear strength of a soil is perhaps the most important of its Engineering properties, as all stability analyses in the field of Geotechnical Engineering are dependent on Shear strength of soil. Permeability is very important engineering property of soils. Knowledge of permeability is essential in settlement of buildings, yield of wells, seepage trough and below the earth structures. The compression of a saturated soil under a steady static pressure is known as consolidation. It is entirely due to expulsion of water from the voids. To cope up with the difficulties involved, an attempt has been made to model Engineering properties of soil i.e. Shear Strength parameters, permeability and compression index in terms of Fine Fraction (FF), Liquid Limit (WL), Plasticity Index (IP), Maximum Dry density(MDD), and Optimum Moisture content(OMC). A multi-layer perceptron network with feed forward back propagation is used to model varying the number of hidden layers. For this purposes 68 soils test data was collected from the laboratory test results. Among the test data 47 soils data is used for training and remaining 27 soils for testing using 60-40 distribution. The architectures developed are 5-5-4(inputs-hidden layers-outputs), 5-6-4, 5-7-4, and 5-8-4. Model with 5-8-4 architecture is found to be quite satisfactory in predicting Engineering properties of soil i.e. Shear Strength parameters, permeability and compression index. Pictorial presentation of results gives a better idea than quantative assessment. A graph is plotted between the predicted values and observed values of outputs for training and testing process, from the graph it is found that all the points are close to equality line, indicating predicted values are close to observed values.
%K Artificial Neural Networks
%K Shear Strength
%K permeability
%K Compression Index
%K Fine fraction
%K Liquid limit
%K Optimum Moisture content
%K Maximum Dry density and plasticity index.
%U http://www.ijitee.org/attachments/File/v3i1/A0909063113.pdf