Adaptive Blast Pattern Optimization via Machine Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current blast mining techniques lack efficiency in fragmenting ore due to variable geological conditions, leading to suboptimal blast patterns and reduced productivity.
Innovation Solution
A system utilizing high-speed optical video recordings and machine learning models to analyze blast patterns, correlate geological characteristics, and determine improved blast patterns, including charge size, spacing, depth, and detonation timing, for more uniform fragmentation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If traditional blast patterns are used without adaptation, then the blasting process is simple and fast, but the fragmentation uniformity and productivity are reduced due to variable geological conditions
Solution Approach 1:
The blast pattern is transformed from a static, fixed design to a dynamic, adaptive pattern that changes based on real-time geological conditions. The system continuously monitors geological parameters and adjusts charge sizes, spacing, and detonation sequences to optimize fragmentation while accounting for variations in rock hardness, density, and fracture locations.
Solution Approach 2:
The system implements a feedback loop where geological data from monitoring sensors is fed into the machine learning model, which then generates optimized blast patterns. The results of each blast are analyzed and used to refine subsequent blast patterns, creating a continuous improvement cycle that enhances fragmentation uniformity over time.
2Productivity
If blast patterns are optimized for each specific geological condition, then fragmentation effectiveness is improved, but the time and resources required for analysis and planning increase
Solution Approach 1:
The system employs machine learning models that automatically analyze geological data and generate optimized blast patterns without requiring extensive manual intervention. The model learns from historical blast data and continuously improves its predictions, enabling rapid determination of optimal patterns while reducing the need for expert manual analysis.
Solution Approach 2:
The system performs preliminary analysis of geological conditions before each blast event, using the machine learning model to predict optimal blast patterns in advance. This allows充分 time for planning and preparation while maintaining efficiency, as the model quickly processes geological data and generates recommendations before the actual blasting operation begins.
3Manufacturing precision
If more variables in the blast pattern are adjusted to improve fragmentation, then the blast effectiveness increases, but the complexity of controlling and coordinating the blast increases
Solution Approach 1:
The system uses a unified machine learning model that simultaneously optimizes multiple blast pattern variables including charge sizes, spacing, depths, and detonation timing. This multi-functional approach allows the system to coordinate all these variables through a single integrated platform, simplifying control while achieving superior fragmentation results.
Solution Approach 2:
The system dynamically adjusts multiple parameters of the blast pattern based on real-time geological conditions. The machine learning model modifies charge sizes, spacing intervals, depths, and detonation sequences as needed, allowing flexible adaptation to varying rock conditions while maintaining ease of operation through automated parameter optimization.
Data Source
AI summary
Techniques for improving a blast pattern at a mining site include conducting an initial blast and recording the initial blast as a high speed optical video. The high speed optical video, and the blast pattern used in the initial blast are sent as inputs to a machine learning model, which correlates one or more characteristics of the region being blasted with measurements associated with characteristics of the region being blasted obtained from the high speed optical video. The machine learning model can then determine an improved blast pattern based on the correlation made. This improved blast pattern can be displayed on a user computing device, or transmitted to a drilling system to automatically drill the improved blast pattern for subsequent blasts.