The trade secret to improving the loading of Blast-Hole
Drones as a standard tool in blast design
Drone acquired data in the beginning of the blast design is rapidly being adopted throughout the mine and quarries blasting arena. Some mines and quarries are looking at smart drills to get the complete information loop, to better enhance production and management in the complete blasting cycle.
Limited amount of data increases standard deviation
Fragmentation is a function of the amount of explosives, the placement of the borehole, type of explosives, bench height burden, spacing, and the geology of the rock itself. Previous research is typically done in a controlled environment. These studies have limited amount of sample data, and sometimes do not take into account the different aspects of the variables that are acquitted in the blasting field.
The issue is the standard deviations between actually by hand sorting, using a sieve and screen, samples of the muck pile and using a drone’s photo measurement. The deviation needs to be less than 10%.
The future of loading blast-hole
Drone cameras have some inherent accuracy issues compare to hand sampling do to resolution of the camera. We feel that this can be overcome by using the following path:
Measuring the pre-blast volume accurately, then flies over muck pile and compares the blast site hole with the muck pile. This shows expansion.
The drone will fly the muck pile and use sizing software to measure the volume of the rock:
- 24 inches or greater
- 12 inches or greater
- 8 inches or greater
- 6 inches or greater
- 4 inches or greater
- 2 inches or greater
- 2 inches and less
- 1 inch or less
The camera cannot see below the slice of the rock on top. The way to get greater accuracy is to allow more samples of the muck. Different avenues of sampling that do not interfere with the muck pile removal can be achieved by the following:
- Fly multi-missions over the muck pile as it is getting moved.
- Sample-photo- trucks hauling the muck pile.
- Sample the conveyor or dump pit.
Presently, we only have small data sampling and is problematic in prediction of fragmentation size as a function of Powder factor.
The better way to calculate the powder factors through a yield muck pile is to keep a ongoing record and build data points into a solvable equation.
Machine learning for accuracy
Machine learning will help in dissolving all of the variables that are inherent in particle mine or quarry site.
With the usage of aerial drones and cloud based software relating back to the original actual data will solve for a more accurate prediction.
Image courtesy : http://arizonaexperience.org/themes/mining-minerals
Check out our 2 Free E-books on AI applications for the drilling, blasting, and mining industries to see all the amazing advances that are available.
AI Guide for Drilling and Blasting
AI Guide for Mining
TLDR? Watch our videos instead:
Get the latest posts delivered right to your inbox