This research aims to accurately predict andcontrol temperature in real time during the continuouscasting of large, semi-finished steel shapes. Theapproach is to create a fast, accurate transient computermodel of heat transfer during the solidification processthat serves as a “software sensor”, calibrated in real timethrough online temperature measurements to providefeedback to a control system, based on algorithmswhich will be designed specifically for this class ofproblem. The new software system will continuouslyread in operating conditions and mold temperatures andcontinuously adjust the spray-water flow rates in thesecondary cooling zone of the caster, in order tomaintain the desired temperature profile throughout thesteel. This profile will be set by steel plant engineers, inorder to minimize the formation of cracks and otherdefects. The system will be calibrated usingthermocouple and optical temperature sensors, testedand implemented at an operating U.S. thin slab caster.This project is important because 96% of the 100million tons of steel produced in the U.S. each year iscontinuously cast, and the fraction produced by the newhigh-speed thin-slab casting process grows every year. This process experiences many defects caused byundesired temperature variations during spray cooling,which are unavoidable using current control systems.Conventional feedback control cannot be used becausetemperature sensors are too inaccurate and expensive.The model-based predictive control system proposedhere must overcome many challenges, including thehigh speed of the process and increased relativeimportance of mold solidification.In the first year of this large project, progress has beenon several different subprojects. A finite differencemodel, CON1D, has been optimized to run in an onlineenvironment under fully transient conditions. It hasbeen integrated into a control system and found tooutperform the existing control system used at the steelplant. It is currently being tested. Spray coolingexperiments have begun, new control algorithms havebeen developed, and new insights into defect formationhave been found. Future improvements to this novelmodel-based control system should revolutionize thecontrol of continuous casting spray systems, withimproved steel quality, and will have beneficial impacton related scientific fields and commercial processes.
展开▼
