Abstract: As a high-temperature, high-pressure, multi-phase reaction vessel, the converter is prone to splashing or slag overflow. Good molten pool surge can increase slag-gold reaction area and improve steelmaking efficiency; Abnormal molten pool surge will cause metal loss, damage the furnace body and its auxiliary equipment, and even threaten the personal safety of workers in front of the furnace. This paper summarizes the previous research results on splashing mechanism and influencing factors. According to the occurrence principle, converter splashing can be divided into explosive splashing, foam splashing, metallic splashing and other splashes, among which explosive splashing is the most harmful and foam splashing occurs most frequently. The occurrence of splashing accidents can be generally attributed to the high-temperature melt splashing driven by bubbles generated by the intense chemical reaction in the furnace and the splashing generated by the flow energy provided by the top and bottom combined blowing for the molten pool. The influencing factors of splashing are summarized from six aspects: charging system, slag making system, oxygen supply system, bottom blowing system, temperature system and safety system, and the foam of slag, oxygen lance blowing parameters and bottom blowing parameters are emphatically analyzed. It is found that the occurrence of a splashing accident is often caused by the coupling of multiple factors. It is too one-sided to analyze the cause of the splashing accident unilaterally, and there is no effective method to quantify the impact of each factor on the splashing at present. Therefore, it is urgent to develop a set of safety evaluation models suitable for converter splashing. In addition, the author summarizes the existing splash prediction models, analyzes the advantages and disadvantages of some splash prediction models, and summarizes the prediction principles and some application results of furnace gas analysis method, audio analysis method and image analysis method. Although preliminary progress has been made in the research of prediction models, there are still more or less problems. It is pointed out that the reason why the existing prediction models have not been widely used is that the prediction accuracy is low, the prediction time is short, the cost is high, and the practicability is low. Some researchers use several models in combination. The results show that different models can learn from each other and the prediction accuracy of the comprehensive model is higher than that of the single model. In the future, the spray prediction model will be more intelligent and refined.