The BPS-K600 continuous casting mould breakout prediction system is developed by Wuhan CenterRise M&C Measuring Co., Ltd. It adopts artificial neural network and combines with traditional artificial intelligence and information processing technology to overcome the defects of traditional logic-based prediction models. It has the functions of self-adaptation, self-organization and real-time learning. The breakout prediction system has an optimized industrial network structure. According to the distribution and change rules of the temperature field of the mold, it can track and respond to the breakout process in real time. It can learn and judge by itself, and provide early warning information for on-site operations. It can effectively prevent the occurrence of breakout accidents, greatly reducing the cost and improving the safety factor of continuous casting. The system is suitable for all kinds of continuous casting machines such as square billet, round billet, rectangular billet and slab.