Abstract
This article discusses a method for the blast furnace regimes determination using a data-mining algorithm based on Kohonen self-organizing maps. The aim of this work is to develop software for the blast furnace regimes con- trol by the multidimensional decision area construction for the main indicators of blast furnace smelting, including the quality of coke, iron ore and hot blast. The article presents the results of the ratio of natural gas and technological oxy- gen influence on productivity and coke rate in each cluster based on the blast furnaces technological statistics of iron and steel plant.
Keywords
blast furnace, clustering, neural network, regimes, Kohonen self-organizing map.
1. Parsunkin B.N. [i dr.] (2011) Vestnik MGTU im. G.I. Nosova,4 : 69-73.
2. Spirin N.A. [i dr.] (2006) Optimizacziya i identifikacziya tekhnologicheskikh proczessov v metallurgii.UGTU-UPY, Ekaterinburg.
3. Suchkov A.V., Lisienko V.G., Suchkov V.A. (2012) Sovershenstvovanie upravleniya mnogomernym tekhnologicheskim obektom na primere domennoj pechi. UrFU, Ekaterinburg.
4. Tovarovskij I.G. (2014) Stal, 5 : 4-11.
5. Tovarovskij I.G. (2009) Domennaya plavka. Porogi, Dnepropetrovsk.
6. Kohonen T. (1990) Proceedings of the IEEE, 78 : 1464-1480.
Barbasova T.A., Bauman E.V., Samoletova P.A. and Cherepanova S.A. (2021) Neural network application for blast furnace regimes determination. Software of systems in the industrial and social fields. 9 (2): 17-20. DOI: 10.18503/2306-2053-2021-9-2-17-20.