Mechanism and forecast model of sticker breakout in continuous casting

Continuous casting is one of the most important processes in the steel making process. There are many steel leakage accidents in the continuous casting process, which have many influencing factors and complex mechanisms. Among them, sticker breakout is the most common accident, accounting for about 70% of the total steel leakage. Continuous casting leakage can easily cause steel leakage, burns, fires and even explosions, resulting in casualties and huge property losses. In view of the above difficulties, the expansion of sticker fracture and the formation mechanism of sticker breakout were analyzed, the principle of forecasting sticker breakout was proposed based on thermocouple temperature measurement method and a sticker breakout forecasting model was established using neural network. The correct reporting rate of the test samples for sticker breakout forecasting model reached 100%, the forecasting rate was 97.56%. The spatial network model was verified, the output of the A-type spatial network met the expectation and could achieve the prediction of bonding in the spatial fracture expansion. The model has good application value and can provide support for the safe production of continuous casting.