Automated Control of Complex Metallurgical Units Based on the CBR Method.

Autor: Kulakov, S. M., Koinov, R. S., Lyakhovets, M. V., Taraborina, E. N.
Zdroj: Steel in Translation; Jun2022, Vol. 52 Issue 6, p586-593, 8p
Abstrakt: The paper considers the vital problem of human-machine control of complex process units and complexes with a large variety of states, multidimensionality, variability, and uncertainty. In ferrous metallurgy, these units include coke batteries, blast furnaces, steelmaking outfits (arc furnaces, oxygen converters), foundry and rolling complexes, rolling mills, main shops and production facilities. It is shown that, in the context of the twenty-first century, the model approach to creating control systems for these objects does not exhibit sufficient efficacy. Alternative approaches based on case-based reasoning (CBR) are considered. In particular, they include the full-scale model of and the full-scale approach to developing support systems and management decision-making. The well-known full-scale model procedures for applying the best practices (methods of standard representative situations and exemplary process cycles) are presented. A new CBR method of automated selection and implementation of control actions with the involvement of process operators is proposed for process control systems. A modified CBR cycle of command selection and the corresponding flowchart of the software control system for a cyclic process unit are developed. The improved CBR-cycle includes several additional operations such as the correction of control decisions for selected cases; retrospective optimization of implemented control decisions; preservation of not only the best and optimized, but also erroneous decisions; case base updating; generation of solutions in unique or previously unreported situations. The structure of the case information model is formed by the example of the software control of steel melting at an oxygen converter shop. This structure includes the data on a specific situation in the control system, parameters of selected control actions, and steel melting results. An example of the control program formation for preparing and conducting the upcoming steel heat is developed on the basis of the data about a pre-selected melting case at a modern oxygen converter shop. [ABSTRACT FROM AUTHOR]
Databáze: Complementary Index