Research at ³ÉÈË´óƬAn onboard sensor and machine learning system for real-time, continuous fragmentation analysis during excavation
Background
Efficient fragmentation measurement is a critical aspect of modern mining operations, influencing both upstream and downstream processes. Accurately understanding the size distribution of fragmented rock after blasting helps mining operations fine-tune future blasting parameters, improve equipment efficiency, and optimize the delivery of materials to comminution circuits — processes that collectively consume a significant portion of energy use.
Current fragmentation measurement techniques, such as image-based systems and sieve analysis, suffer from several limitations. These methods often provide only periodic sampling rather than continuous data, leaving significant gaps in understanding the full fragmentation profile. Furthermore, image-based technologies are limited to surface-level analysis, require optimal lighting and calibration, and are impractical for underground applications where it is dark. 3D scanning systems also present challenges due to their stationary nature, shadow interference, and sensitivity to harsh mining environments. These conventional approaches are intrusive, requiring operators to acquire data manually at the excavation face, disrupting the production process.
Technology Overview
This novel vision-less technology improves fragmentation measurement by embedding proprioceptive sensors directly onto mining equipment such as excavators. These sensors continuously capture force and position data during excavation, enabling real-time estimation of rock fragmentation across the full volume of extracted material
The system employs machine learning algorithms trained to correlate sensor data patterns with fragmentation characteristics, delivering ongoing insights without interrupting operations. This approach captures fragmentation data as part of the normal production cycle, offering a comprehensive and dynamic view of fragmentation conditions at every stage of the excavation process.
While initially developed for excavators, the technology can be extended to other mining equipment that interacts with rock media, such as loaders, haul trucks, and drilling rigs, further enhancing the data ecosystem for integrated mine management.
Benefits
- Continuous Data Acquisition: Provides real-time fragmentation estimates across the full volume of material during excavation.
- Non-Intrusive: Fully integrated with existing production processes — no operator intervention required.
- Complete Fragmentation Data: Captures data throughout the excavation face, rather than just surface or sample-based data.
- Operational Flexibility: Effective in both surface and underground mining environments, without dependence on lighting.
- Rugged & Reliable: Sensor-based system is more durable than fragile camera and 3D scanning systems, reducing maintenance and replacement costs.
- Optimization Insights: Data supports dynamic updates to blast design, processing strategies, and material handling workflows.
- Reduced Environmental Impact: Enables optimization of blasting and comminution, potentially reducing energy consumption and waste.
Applications
- Mining Operations (Surface & Underground)
- Blast Design Optimization
- Comminution Circuit Optimization
- Fleet Management System Integration
- Autonomous Mining Systems
- Real-Time Mine Planning & Operations Control
Opportunity
Queen’s University is seeking partners for further development, field validation, and commercialization of this novel fragmentation technology.
Collaborators will benefit from access to a novel technology that delivers enhanced operational efficiency, improved safety, and reduced environmental footprint through smarter blasting and processing decisions based on real-time, comprehensive fragmentation data.
Patents
- CA3214713A1
- AU2022254939A1
IP Status
Patent application submitted
Seeking
- Development partner
- Commercial partner
- Licensing
Posted
March 24, 2025