Speaker: Dr. Anton Kepic, Curtin University, Perth, Western Australia

Title:
Creating Volumetric Models of RQD for mining from Geophysical Measurements

Abstract:

An example using data from the Kevitsa Ni-Cu-PGE mine is provided to
demonstrate how geophysical log data from many boreholes may be combined
with 3D seismic data to provide a 3D model of RQD. A fuzzy cluster
algorithm (FCM) is used to sort the log data into clusters, rock units
with characteristic properties. Then borehole cluster data is
interpolated between the boreholes by a special FCM version of seismic
impedance inversion applied on 3D seismic reflection data set to create 3D
volume of sonic velocity. Sonic velocity may be related to other
engineering properties. Using the cluster model plus sonic velocity
allows prediction of RQD with good correlation in blind tests (>90%).