SOIL TYPE PREDICTION USING KERAS FRAMEWORK

SOIL TYPE PREDICTION USING KERAS FRAMEWORK

MRS VIDHYA. A

RAHUL RAJBHAR.S

KAMLANATHAN.R

KARAN GANESH.G

Department Of Computer Science and Engineering

Jeppiaar Engineering College , Chennai , India

---------------------------------------------------------------------***---------------------------------------------------------------------

Abstract - The soil characterization is one of the major  prerequisites in grounding system studies. Along with an  adequate measurement procedure and a reliable field data  processing, they are responsible for the success in the design  and the installation of grounding grids in electrical power  systems. In this frame of reference, the purpose of this work is  to present an embracing investigation combining in-situ measuring data from soil resistivity and geological survey,  supported by the Wenner Method and Standard Penetration  Test (SPT). The study is focused on investigating the  differences in the apparent soil resistivity interpretation, provided by distinct approaches, identifying the one that better  correlates with SPT.

Keywords: Soil type, deep learning, TensorFlow, Keras, CNN