Record Details

Comparative analysis of intelligent algorithms to correlate strength and petrographic properties of some schistose rocks

DSpace at IIT Bombay

View Archive Info
 
 
Field Value
 
Title Comparative analysis of intelligent algorithms to correlate strength and petrographic properties of some schistose rocks
 
Creator SINGH, TN
VERMA, AK
 
Subject Deformational modulus
Texture coefficient
Resilient propagation
One step secant algorithm
t-statistic
MASS
 
Description Empirical models to correlate deformational modulus along with petrographic features which are intrinsic and inherent properties of rock with other basic mechanical and physical properties have earlier been proposed with experiential and assumed reasoning. However, in most cases, such empirical models make certain basic assumptions and hence bring in a degree of dispose and doubt. An attempt has been made in this paper to analyze and compare the efficiency and applicability of different cognitive algorithms for the prediction of deformational modulus and texture coefficient. The importance of knowledge of deformational modulus is unparalleled with the view to the operational difficulties in its determination. Rock samples were taken from a tectonically active and complex sequence from a large underground excavation in the Himalayan region and were tested in the laboratory to determine the different strength properties. One hundred and seventy six rock samples test results were used as part of the experiment. The uniaxial compressive strength, tensile strength, axial point load strength, porosity, and void ratio were taken as inputs to get deformational modulus and texture coefficient. Networks were trained to optimum number of epochs or iterations to make suitable prediction. The results of intelligent systems have been tested against that of statistical methods as a test of precision of the model in generalization principles.
 
Publisher SPRINGER
 
Date 2014-10-14T17:51:19Z
2014-10-14T17:51:19Z
2012
 
Type Article
 
Identifier ENGINEERING WITH COMPUTERS, 28(1)1-12
http://dx.doi.org/10.1007/s00366-011-0210-5
http://dspace.library.iitb.ac.in/jspui/handle/100/14613
 
Language en