Modeling Blast Movement for Grade Control - · PDF fileModeling Blast Movement for Grade Control Edward Isaaks Robert Barr . Onie Handayani

Modeling Blast Movement for Grade Control

Edward Isaaks Robert Barr Onie Handayani

Ore Control at Boddington

Post Blast Dig Lines?

Proposed Solution Simulate the blast movement of each sub-block conditional to blast movement data.

Available Data

The Ore Control Block Model

Blast Movement Monitoring (BMM) Data • 1,515 BMM observations • 35 variables per observation

Classification Analysis

1515

South Pit North Pit

An, Di An, Di, Do An, Di, Do Di, Do

Echelon Center Lift

Group Statistics

Group 3 - horizontal movement versus depth from top

Post Blast Surface Topography

Blast Initiation Sequence

Modeling Blast Movement

Ore Control Model (OCM) Voxels Pre Blast OCM cross section

Post Blast OCM cross section

Assign a Distribution of Horizontal Movements to each Pre-Blast Voxel

N(6,3)

N(16,4)

N(8,3)

Assign Movement Direction to each Pre-Blast Voxel and Post-Blast Grid Node

Add a random component to each pre-blast voxel direction at time of simulation

Simulate Blast Movement 1. Randomly select a pre-blast voxel. 2. Draw a random distance from the distribution of

voxel horizontal movements. 3. Add a random component to the voxel movement

direction. 4. Calculate the destination grid node in the post-

blast enevelope. 5. Move the pre-blast voxel to the post-blast grid

node if vacant. 6. Continue steps 1 – 5 until all post blast grid nodes

are occupied.

Simulate blast movement

Results

Post-Blast Tonnes = Pre-Blast Tonnes

Post Blast-Volume = 1.38 x Pre-Blast Volume

Post-Blast Grade = Pre-Blast Grade

Post-Blast Model Topo = Surveyed Post-Blast Topo

Histogram of 257,100 Simulated Horizontal Movement Vectors

Simulated Horizontal Movement Vectors Versus Depth-From-Top

Application to Grade Control

Pre-Blast Dig Lines Post-Blast Dig Lines

Tonnes

Grade

Ounces Au

Recoverable Ounces Au

Recoverable pre-blast oz Au = 14,241

Recoverable post-blast oz Au = 14,169

Loss = 72 oz Au

Future Work

• Validation of the model. • Add control parameters to the simulation algorithm. • Additional work on characterizing the random directional component. • Development of a software product.

Thank you