Analyzing Shovel Tooth Wear Patterns with an Automated Machine Vision System

Tooth-wear on mining shovels requires continuous monitoring. Worn teeth impede excavation performance, increase energy usage, and increase the risk of lost or broken teeth and adapters. A machine-vision-based automated tooth-wear monitoring system is presented. The system employs a rugged boom or stick-mounted camera to capture images of the shovel teeth. An onboard computer processes the images in real-time to determine the length of each tooth. The current wear state is available via a network-accessible log for on-demand status updates. By continuously recording tooth length changes over time, the logs can be used to estimate wear rates and patterns, permitting comparisons between different G.E.T. configurations, operator behaviour patterns, or material types. This detailed, quantitative data permits site-specific optimizations to wear packages and maintenance scheduling. Automating wear monitoring offers the crucial wear information necessary to minimize maintenance costs and reduce unscheduled downtime. A 37-day field trial in a copper-gold mine demonstrated the ability of the system to measure tooth lengths, compute wear rates, and detect differences in tooth wear based on their position on the bucket.