Journals →  Tsvetnye Metally →  2017 →  #7 →  Back

AUTOMATION
ArticleName Intelligent system for automatic control of powerful rotating furnaces, designed for sintering of free-flowing metallurgical materials using associative knowledge bases
DOI 10.17580/tsm.2017.07.15
ArticleAuthor Salikhov M. Z., Salikhov Z. G.
ArticleAuthorData

Institute of Control Sciences RAS, Moscow, Russia:

M. Z. Salikhov, Senior Researcher
Z. G. Salikhov, Professor, Chief Researcher, e-mail: zuf1940@yandex.ru

Abstract

Rotary furnaces have been used for a long time in the production of nonferrous and ferrous metals and processing of technology-related waste, e.g. making cement clinker from smelter slag. However, a full automation of manufacturing processes inside such furnaces still poses a challenge. The accuracy and reliability of temperature control during material burning processes remains unsatisfactory due to the failure of conventionally used thermometers, thermocouples and other measuring equipment. The use of mathematical models brought no success either since working parameters of the process vary in time and depend on the charge mixture, furnace length and furnace operation modes. For the time being, the processes in most furnaces are controlled intuitively by the process operator, which reduces the quality of the burnt stuff as well as the service life of the furnace lining and gas-cleaning filters due to a “blow-by” of high temperature waste furnace gases, whilst also creating environmentally hazardous situations inside the workshop. The authors propose for such systems to be designed based on a cumulative use of fuzzy function theory and a new method for selection of experts from among the currently working process operators in order to extend the associative data knowledge base. Such systems are able to generate control actions by taking into account the distributed non-stationary, poorly observed process parameters, and by using situational data from the associative knowledge base. The results of industrial testing of this system demonstrated a 5–6% increase of the static and dynamic relative control accuracy and a possible improvement in the furnace throughput by 20–30% thanks to extending the continuous operation time by 30–40 days. The authors also confirmed the possibility of a timely identification of defects which may result in destruction of lining and furnace during its operation, as well as concretization of the lining work’s quality at low furnace heating temperatures. The proposed approach for building intelligent automatic process control systems for burning of free-flowing materials in powerful rotary furnaces thus makes it possible to achieve the following goals: enhanced control accuracy; extended lining service life; increased equipment capacity; early identification of defects in lining before they take place.

keywords Rotating furnaces, intelligent automatic control systems, random and deterministic disturbances, associative knowledge base, theory of fuzzy functions, evaluation of process engineers' quality indicators, objective method for the identification of uncontrolled distributed parameters with actual parameters
References

1. Available at: http://www.consultant.ru/document/cons_doc_LAW_207967/
2. Salikhov Z. G., Salikhov K. Z. Method of control of rotary kiln materials roasting process. Patent RF, No. 2249775. Applied: 09.10.2001. Published: 10.04.2005. Bulletin No. 10.
3. Salikhov Z. G., Bekarevich A. A. Automatic control systems for calcination processes in rotary kilns. Avtomatizatsiya v promyshlennosti. 2003. No. 3. pp. 15–17.
4. Yudin D. A., Magergut V. Z. Vision systems for monitoring of roasting process in rotary kilns. Belgorod : BGTU, 2014. 108 p.
5. Burgin Mark. Theory of Knowledge : Structures and Processes. World Scientific, 2016. 957 p.
6. Tao F., Zhang L., Laili Y. Configurable intelligent optimization algorithm: design and practice in manufacturing. Springer, 2014. 364 p.
7. Kozin R. G. Expert systems. Moscow : MIFI, 2008. 87 p.
8. Hudson D. Statistics for physicists. Moscow, 1970. 296 p.
9. Karabutov N. N. Structural identification of statistical objects: fields, structures, methods. Moscow : URSS Knizhnyy dom “Librokom”, 2011. 152 p.
10. Lapack — Linear Algebra PACKage. Available at : http://www.netlib.org/lapack/
11. Wang J.-S., Zhang L., Gao X.-W., Sun S.-F. Intelligent control metod of rotary kiln process based on image processing technology : asurvey. 29th Chinese Conference (CCC). 2010. pp. 2930–2935.
12. Gavrilova T. A., Khoroshevskiy V. F. Knowledge bases of intellectual systems. Saint Petersburg : Piter, 2000. 384 p.
13. Salikhov Z. G., Zykov I. E. et al. Quantitative assessment of quality of metallurgical unit control. Method of a choice of experts. Tsvetnye Metally. 2002. No. 10. pp. 88–92.
14. Klumberg O. A., Fedorov I. P., Zmanovskiy T. P. Methods of organization of production representation of knowledge. Riga : Rizhskiy politekhnicheskiy institut. 1989. 128 p.

Language of full-text russian
Full content Buy
Back