Differential Blast Design Using Neural Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In mining operations, existing blast design methods fail to efficiently target areas with varying concentrations of valuable minerals, leading to suboptimal energy distribution and fragmentation, resulting in reduced mineral recovery and increased costs.
Innovation Solution
A differential blast design system that uses neural networks and borehole imaging to identify areas with high mineral concentrations, adjusting explosive energy distribution by inserting interstitial blast holes or using stronger explosives, and implementing a data-driven approach to optimize blast energy density and fragmentation patterns based on seismic and drilling data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If uniform blast design is used across the entire mine bench, then the blasting process is simple and cost-effective, but mineral recovery is reduced due to inability to target high-concentration areas
Solution Approach 1:
The patent divides the mine bench into multiple zones based on mineral concentration levels identified through borehole imaging and neural network analysis. Each zone is assigned different blast design parameters (explosive type, charge weight, hole spacing) to optimize mineral recovery in high-concentration areas while maintaining efficiency in lower-concentration areas.
Solution Approach 2:
The patent implements location-specific blast design where high-concentration mineral zones receive intensified blasting (stronger explosives, higher energy density) while low-concentration zones use standard or reduced blasting. This local differentiation maximizes mineral recovery where it matters most while avoiding unnecessary energy expenditure in low-value areas.
2Productivity
If additional interstitial blast holes are inserted to target high mineral concentration areas, then mineral recovery increases, but the complexity and cost of blast hole drilling increases
Solution Approach 1:
The patent performs preliminary borehole imaging and neural network analysis before finalizing the blast design to identify high-concentration mineral zones. This advance knowledge allows precise placement of interstitial blast holes exactly where needed, avoiding unnecessary drilling in low-concentration areas and optimizing the return on drilling investment.
Solution Approach 2:
The patent uses borehole imaging to create detailed digital models of the subsurface mineral distribution, which are then processed by neural networks to generate optimized blast hole patterns. These digital models serve as templates for planning the precise location and configuration of interstitial blast holes.
3Productivity
If higher energy density explosives are used in high mineral concentration areas, then fragmentation and mineral recovery improve, but energy consumption and environmental impact increase
Solution Approach 1:
The patent applies different explosive energy densities to different zones based on mineral concentration. High-energy-density explosives are concentrated only in high-value mineral zones where they provide maximum recovery benefit, while standard energy-density explosives are used in lower-concentration zones, thereby reducing overall energy consumption and environmental impact.
Solution Approach 2:
The patent dynamically adjusts explosive parameters (energy density, charge weight, detonation velocity) based on the spatial distribution of mineral concentrations identified through borehole imaging and neural network analysis. This parameter optimization ensures high energy expenditure only where it generates proportional value in terms of mineral recovery.
4Manufacturing precision
If data-driven approach with neural networks is implemented to optimize blast design, then blast energy distribution and fragmentation are optimized, but the system complexity and data processing requirements increase
Solution Approach 1:
The patent replaces traditional mechanical and empirical blast design methods with neural network-based computational models that process borehole imaging data to automatically generate optimized blast patterns. This substitution enables precise energy distribution and fragmentation control through data-driven insights rather than conventional trial-and-error approaches.
Solution Approach 2:
The patent introduces neural networks as an intermediary between raw borehole imaging data and blast design decisions. The neural networks process complex geological and mineralization patterns to generate actionable blast design parameters, serving as an intelligent mediator that translates subsurface data into optimized blasting strategies.
Data Source
AI summary
Respective embodiments disclosed herein include methods and apparatuses (1) for surveying a mine bench or other material body using at least seismic data obtained via geophone and measurement module data synchronized via a wireless link; (2) for generating hyperspectral panoramic imaging data of a blast hole or other borehole; or (3) for allowing a neural network to facilitate a differential blast design that targets a first bench part more weakly than the differential blast design targets a second bench part (along the same mine bench) at least partly based on data indicative of a much higher concentration of a valuable material in the second bench part than in the first.