Optimization of the Continuous Casting Process of Hypoeutectoid Steel Grades Using Multiple Linear Regression and Genetic Programming—An Industrial Study

Štore Steel Ltd. is one of the major flat spring steel producers in Europe. Until 2016 company used a three-strand continuous casting machine with 6 m radius, when it was replaced by completely new two-strand caster 9 radius. For comparison tensile strength 41 hypoeutectoid grades, we conducted 1847 tests during first period testing using old caster, and 713 second caster. It found that for 11 grades rolled material statistically significantly lower (t-test method) whereas all other remained same. To improve process, decided to study process more detail Multiple Linear Regression method Genetic Programming approach based on items empirical data obtained machine. Based models determined most influential parameters product. According model’s analysis, secondary cooling at improved installation self-cleaning filter 2019. After implementing this modification, performed an additional 794 third testing. out that, after filter, 19, condition significantly,