Predictive maintenance in a rolling mill
Health indicators and anomaly detection models help detect abnormal behavior in rolling equipment before it turns into downtime.
Health indicators and anomaly detection models help detect abnormal behavior in rolling equipment before it turns into downtime.
Challenge
Rolling mills generate many process signals across campaigns, products and routes. Maintenance teams need to know when equipment behavior deviates from normal operation.
Data
Historical campaigns were grouped by product and route, using process variables such as speed, torque, tension, RPM and load indicators.
Solution
Artelnics combined statistical ranges with auto-associative neural networks to calculate health indicators and detect deviations in real time.
Auto-associative neural networks and statistical baselines.
Deviation indicators and alarms by equipment context.
Results
The solution supports proactive maintenance by highlighting when process behavior moves outside expected operating patterns.
Illustrations
