Predictive Maintenance for Steel Mill Roughing Stands
In steel rolling mills, the rollers are housed in stands through which the steel billets pass to achieve a desired shape. Stable operating conditions for these stands are vital to achieving a good dimensional quality of the final product.
Typically, steel rolling is done in three phases:
- The roughing stage
- The intermediate stage
- The finishing stage.
This case study was done on a roughing stand within a line of 8 total stands in which the material was alternated between being rolled vertically and horizontally.
SORBA.ai offers predictive maintenance capabilities by using machine learning models to detect anomalies in real time data and predict when failures will occur so that they can be mitigated or fixed prior to a failure occurring.
Predictive maintenance offers cost saving advantages over conducting maintenance on a set schedule by allowing for maintenance to only be conducted only when it is required. The use of machine learning methods to create these models offers even further advantages over other predictive methods; the incorporation of artificial intelligence allows the predictive maintenance model to change over time, reacting to changes to the machine that may occur over time and by finding patterns in the data that cannot be identified by human intelligence alone.