CONTROL STRUCTURE FOR THE REDUCTION OF DEFECTS IN CONTINUOUS CASTING
F.R. Camisani-Calzolari* I.K. Craig** P.C. Pistorius***
* Department of Electrical, Electronic and Computer Engineering, University of Pretoria, Pretoria, 0002, South Africa. Tel. +27 14 420 2166 Fax. +27 12 362 5000 email: cami-fr@suntiger.ee.up.ac.za
** Corresponding Author. Department of Electrical, Electronic and Computer Engineering, University of Pretoria, Petoria, 0002, South Africa. Tel +27 12 420 2172 Fax. +27 12 362 5000 email: icraig@postino.up.ac.za
*** Department of Materials Science and Metallurgical Engineering, University of Pretoria, Pretoria, 0002, South Aferica. Tel. +27 12 420 3922 Fax. +27 12 362 5304 email: pistori@postino.up.ac.za
The causes for surface defects in the continuous casting process must be understood so that surface defects can be eradicated. The defects which were considered are presented and an approach to eliminate many of the influencing variables using statistical hypothesis testing is presented. These methods show that only the thermocouple temperature readings within the mould are necessary to accurately predict the occurance of defects. The use of ARX modeling is then presented for a predictor which uses thermocouple temperatures as inputs and defects as outputs. A mould model can be determined in the same way using controlled variables as inputs and the thermocouples as outputs. Optimal control and feedback control can be used to diminish the defects.
Keywords: steel industry, continuous casting, defect, hot charging, direct rolling, CAQC, control system, statistical tests, system identification
Session slot T-Th-E12: Control Applications in Casting Plants/Area code 7b : Mining, Mineral and Metal Processing

|