Visual quality control of the vacuum tank degassing

Abstract In this paper, a machine vision system for the quality control of the vacuum tank degassing in steel plants is introduced. The system can classify the degassing quality during the degassing process and predict the final hydrogen content of the liquid steel after the degassing. The introduced system can be integrated into an implementation of an automatic process control system for the vacuum tank degassing station. It could be possible for the control system to automatically complete the degassing process and also to control the stirring gas flow in the vacuum tank during the process. Visual inspection of the machine vision system is based on image analysis where digital images are classified using neural network methods. The results are promising, as compared to exact measurements and to manual inspection by vacuum tank degassing experts.

[1]  Balázs E. Pataki,et al.  Black-box modeling of a complex industrial process , 1999, Proceedings ECBS'99. IEEE Conference and Workshop on Engineering of Computer-Based Systems.

[2]  Anil K. Jain,et al.  A Survey of Automated Visual Inspection , 1995, Comput. Vis. Image Underst..

[3]  Robert J. Schalkoff,et al.  Pattern recognition - statistical, structural and neural approaches , 1991 .

[4]  P. Marino,et al.  Inspection of steel sheets based on CCD image sensors , 1999, Proceedings. 1999 IEEE/SICE/RSJ. International Conference on Multisensor Fusion and Integration for Intelligent Systems. MFI'99 (Cat. No.99TH8480).

[5]  Rajkumar Roy,et al.  Fuzzy process modelling for secondary steelmaking , 1999, 18th International Conference of the North American Fuzzy Information Processing Society - NAFIPS (Cat. No.99TH8397).