The overload prediction model is a real-time model that, every minute, outputs the estimated probability of an imminent or active overload happening in the SAG mill.
The minute-wise prediction means that the model is not identifying isolated overload events, but rather gives a constant stream of estimates for the classification that it was trained on.

Its output provides valuable information about current mill condition and efficiency and can be further used to make decisions such as the feed reduction before an overload occurs and thus prevent them.

Did this answer your question?