Prediction Of Longitudinal Cracks On Slab Surface During Continuous Casting Production

With the continuous improvement of the quality of the slab,the research on the quality control of the slab has become the main research topic.In the actual production,Timely and accurate online prediction of continuous casting slab quality is helpful to improve the quality of the slab and reduce the cost.Longitudinal surface crack is one of the common surface quality defects of continuous casting slabs.The slight longitudinal surface cracks have less influence on the next processing by final finishing,but the severe longitudinal surface cracks would cause serious quality problem,even leading to the production emergency.The factors leading to the surface longitudinal crack mainly include the steel chemicals,operation process and equipment factors.The production data analysis and research.The main factors affecting the longitudinal surface crack are the composition of molten steel([C],[Mn],[P],[S],[Si]),casting speed,superheat,fluxes,sectional dimensions of the billet.In order to improve the accuracy of the network and adapt to the field production,the regression analysis of the data,according to the absolute value of the factors in the regression formula,the impact of various factors on the crack,and determine the final input.The main conclusions are as follows:(1)The main factors affecting the longitudinal surface crack are the composition of molten steel([C],[Cu],[P],[S],[Si]),casting speed,Left heat flux distribution、broadface heat flux distribution,right heat flux distribution.、taper of mold.(2)A prediction model for longitudinal surface crack of slab based on artificial neural network technology has been developed,The error ratio is less than 10%,and the hit rate is more than 90%.(3)The input data is processed in a custom way.The model accuracy rate is higher than the traditional model.