Method for improving convexity hit rate of hot-rolled strip steel

The invention provides a method for improving the convexity hit rate of hot-rolled strip steel. The method comprises the specific steps that data actual measurement and simulation analysis are conducted on the process before rough rolling; s2, completing the construction of a mechanism model according to the basic data obtained in S1; relevant parameters of each production process window and a continuous casting billet before rough rolling of each production line are determined; performing batch calculation on continuous casting and rough rolling data in the process window by combining the mechanism model calculated in the step S2 with the relevant parameters obtained in the step S1 and the step S3; real-time data of the continuous casting process, the rough rolling process and the steel grade components and mechanism model calculation data are trained in a machine learning mode, and a prediction model of the convexity of the intermediate billet at the inlet of the finishing mill frame is trained; and the pre