Journal Published Online: 05 November 2019
Volume 48, Issue 1

Data Mining Techniques for the Prediction of Bohme Surface Abrasion Rates from Rock Properties

CODEN: JTEVAB

Abstract

Abrasion refers to the wearing down of rock surfaces due to abrasive grains. Abrasion resistance refers to the ability of rocks to withstand wear. Abrasion resistance is used to determine the resistance of building materials produced for flooring, cladding, and pavements and to demonstrate suitability for higher movement areas. While it is, therefore, very necessary to determine the abrasion rate of building materials prior to construction, it is, however, highly demanding and time consuming to determine abrasion rates. Thus, the aim of this study is to use some rock properties to determine abrasion rates. The study samples, consisting of 32 different types of rocks (sedimentary, metamorphic, and igneous) collected from different regions in Turkey, were subjected to some physical and mechanical tests, namely the following: unit volume weight (UVW), apparent porosity (AP), modulus of elasticity (E), uniaxial compressive strength (UCS), tensile strength (TS), Shore hardness (SH), and point load strength (PL) and Bohme abrasion tests. To ascertain the abrasion rate from some physical and mechanical properties of rocks, the results of these tests were analyzed using data mining (DM) techniques. The results showed that there are high correlation coefficients between abrasion rate and the aforementioned rock properties with the support vector machines (SVM) and random forests (RF) models obtained as R = 0.882 and 0.881, respectively. This work has shown that the rock Bohme abrasion rate can be predicted from some of its physical and mechanical properties with significant level of confidence.

Author Information

Bayram, Fatih
Department of Mining Engineering, Afyon Kocatepe University, Afyonkarahisar, Turkey
Pages: 10
Price: $25.00
Related
Reprints and Permissions
Reprints and copyright permissions can be requested through the
Copyright Clearance Center
Details
Stock #: JTE20190130
ISSN: 0090-3973
DOI: 10.1520/JTE20190130