📋 Case Study

Iron Ore Mine Crushing Plant Equipment Thermal Inspection

Unexpected bearing failures in primary gyratory crushers causing 8–12 hr unplanned downtime; IR thermography traditionally performed during shutdowns only

🏗️ Project Overview

Roy Hill Operations, Pilbara, Australia

🎯 Challenge

Unexpected bearing failures in primary gyratory crushers causing 8–12 hr unplanned downtime; IR thermography traditionally performed during shutdowns only

🔧 Design Approach

Autonomous thermal drone patrols during live operation using pre-programmed waypoints; FLIR Boson 640 × 512 core; anomaly detection via temperature gradient clustering (ΔT/Δx > 12°C/m)

📐 Design Diagram

Gyratory Crusher Bearing Autonomous Patrol Path Boson AI Anomaly Engine ΔT/Δx > 12°C/m Alarm: T > 93°C FP Rate: 2.1% T_baseline = 68°C Alarm: 93°C Challenge: 8–12 hr unplanned downtime from bearing failures Crusher Drone Path Alarm Threshold Challenge

AI-generated project design illustration

📐 Key Calculations

Critical Bearing Temp Rise

T_surface > T_baseline + 25°C
Result: T_baseline = 68°C → alarm at 93°C
Validated against OEM failure thresholds

False Positive Rate

FP / (FP + TN)
Result: 2.1%
Kept operator trust in alert system

📊 Results

Detected 3 incipient bearing faults 72–96 hrs pre-failure; extended mean time between failures (MTBF) by 44%; eliminated 100% of unplanned crusher stoppages in Q3 2023

💡 Lessons Learned

  • Solar loading on steel surfaces required dynamic baseline modeling
  • Drone IR calibration must be performed at operational ambient (42°C+) to avoid bias

Key Takeaways

  • 1Solar loading on steel surfaces required dynamic baseline modeling
  • 2Drone IR calibration must be performed at operational ambient (42°C+) to avoid bias