Prediction system of strip chew in hot rolling mill

A prediction system of strip chew in a hot rolling mill in which a strip-like metal material is heated to a high temperature and rolled through a plurality of rolling passes, the strip chew being a phenomenon occurring at a head end or tail end of a strip due to steering of the strip or bending of the strip to width direction,

• the system comprising one or more computers,
• the one or more computers programmed to execute:
• a process of collecting and storing adaptive model construction data used to construct an adaptive model for predicting occurrence of the strip chew;
• a process of constructing the adaptive model using the adaptive model construction data;
• a process of storing an adapted model, which is the adaptive model that has been constructed;
• a process of collecting prediction data used to predict the occurrence of the strip chew;
• and
• a process of predicting the occurrence of the strip chew by inputting the prediction data to the adapted model,
• wherein, in the process of collecting and storing the adaptive model construction data, the one or more computers collect a plurality of sets of first data and second data as the adaptive model construction data, the first data being data which indicates the occurrence or non-occurrence of the strip chew in an object rolling path which is an object of strip chew occurrence prediction and an occurrence point of the strip chew, and the second data being data which includes information on a preceding rolling path preceding the object rolling path in rolling order, which is obtained when a strip linked to the first data is rolled in the preceding rolling path, and attribute information on the same strip,
• in the process of collecting the prediction data, the one or more computers collect, as the prediction data, data which includes information on a preceding rolling path preceding the object rolling path in rolling order, which is obtained when a prediction object strip is rolled in the preceding rolling path, and attribute information on the same strip, and
• in the process of predicting the occurrence of the strip chew, the one or more computers predict the occurrence or non-occurrence of the strip chew in the object rolling path and all or some of the occurrence points of the strip chew before the prediction object strip reaches the object rolling path.