Abstract
Modification of the classical genetic algorithm is presented in the article and its application is shown for solving the problem of multicriteria optimization of the structure of charge materials for an electric arc furnace. The goal of the study is to determine effective algorithms to solve the optimization problem of the structure of charge materials for an electric arc furnace. The objective functions in the multicriteria optimization problem are the content of residual elements in steel such as chromium, nickel and copper. The required values are the tuple defining the share participation of the constituent parts of the electric arc furnace batch, such as the types of metal scrap. In the process of the investigation, the following tasks were performed: setting up a multicriteria optimization problem that is reduced to the one-criteria view using a modified genetic algorithm using the method of convolution of criteria with given weight coefficients; modification of the classical genetic algorithm based on the introduction of the stage of improvement of the individual; choice of fitness function for searching solutions in unacceptable areas; determination of the order of transition from a modified genetic algorithm to a classical one. In process of the research, the following methods were used: convolution of criteria for converting a multicriteria optimization problem to a one-criterion form, classical and modified genetic algorithms, and the method of penalty functions. The proposed modification of the classical genetic algorithm made it possible to achieve a stable process of convergence of the computational process at the first stages of the computational process. The composite combination of the classical and modified genetic algorithm allows the authors to achieve the solution of the optimization problem on the structure of the charge materials of an arc steelmaking furnace with the specified accuracy no more than 30 iterations.
Keywords
Genetic algorithm, improvement of the individual, structure of charge materials, electric arc furnace, optimization problem.
1. Logunova O.S. The technology of research of information flows at a metallurgical enterprise. Informacionnye tekhnologii v proektirovanii i proizvodstve [Information technology of CAD/CAM/CAE], 2008, no. 3, pp. 32-36. (In Russian)
2. Honjo N, Tsuno M, Shinkai M. Steelmaking. Development of an Optimized Burner Technology in Arc Furnace. 1999. 70(2). pp.163-171. https://doi.org/10.4262/denkiseiko.70.163.
3. Injection burner for auxiliary heating of scrap charge in electric arc furnace for steelmaking. https://docslide.com.br/ documents/9803257-injection-burner-for-auxiliary-heating-of-scrap-charge-in-electric-arc.html. Access 05.02.2018
4. Kornilov G.P., Nikolaev A.A., Anokhin V.V. Power Increase of Steelmaking Electric Arc Furnace. Metallurgist. 2016, no. 60, pp. 780-785. https://doi.org/10.1007/s11015-016-0367-7
5. Nikolaev A.A., Kornilov G.P., Anufriev A.V., Pekhterev S.V., Povelitsa E.V. Electrical optimization of superpowerful arc furnaces. Steel in Translation, 2014, no. 44(4), pp. 289-297.
6. Shevtsov A.Z., Yugov P.I., Okorokov G.N. et al. Efficiency of a combination «DC-ARC furnace-converter» steelmaking module. Metallurgist, 1998, no. 42, iss. 12, pp. 477-479. https://doi.org/10.1007/BF02511768
7. Bigeev V.A., Valiahmetov A.H., Burak Je, Fedyanin A.N. The experience of steel smelting in a super-powerful arc furnace with a high consumption of solid cast iron. Vestnik Magnitogorskogo gosudarstvennogo tekhnicheskogo universiteta im. G.I. Nosova [Vestnik of Nosov Magnitogorsk State Technical University], 2014, no. 1, pp. 15-18. (In Russian)
8. Logunova O.S., Filippov E.G., Pavlov I.V., Pavlov V.V. Multicriterial optimization of the batch composition for steel-smelting arc furnaces. Steel in Translation, 2013, no. 43(1), pp. 34-38. Doi: 10.3103/S0967091213010051
9. Logunova O.S., Sibileva N.S. Intelligent Support System of Steel Technical Preparation in an Arc Furnace: Functional Scheme of Interactive Builder of the Multi Objective Optimization Problem. IOP Conference Series: Materials Science and Engineering, 2018, no. 287, DOI: 10.1088/1757-899X/287/1/012009.
10. Logunova O.S., Devyatov D.H., Yachikov I.M., Kirpichev A.A. Mathematical modeling of macroscopic parameters of solidification of continuous ingots. Izvestiya vysshih uchebnyh zavedenij. Chernaya metallurgiya [Proceedings of Higher Educational Institutions. Ferrous Metallurgy], 1997, no. 2, pp. 49-51. (In Russian)
11. Kolokoltsev V.M., Sinickij E.V., Savinov A.S. Modeling of temperature fields in the production of castings. Vestnik Magnitogorskogo gosudarstvennogo tekhnicheskogo universiteta im. G.I. Nosova [Vestnik of Nosov Magnitogorsk State Technical University], 2015, no. 3 (51), pp. 39-43. (In Russian)
12. Konstantinov D.V., Korchunov A.G. Multiscale Computer Modeling of Processes of Metal Working with Pressure. Vestnik Magnitogorskogo gosudarstvennogo tekhnicheskogo universiteta im. G.I. Nosova [Vestnik of Nosov Magnitogorsk State Technical University], 2015, no. 1 (49), pp. 36-43. (In Russian)
13. Nazarov Sh.A., Ganiev I.N., Norova M.T., Ganieva N.I., Kalliari I. Potentiodynamic study of the alloy Al + 6% Li with yttrium in the NACL electrolyte medium. Vestnik Magnitogorskogo gosudarstvennogo tekhnicheskogo universiteta im. G.I. Nosova [Vestnik of Nosov Magnitogorsk State Technical University], 2016, vol. 14, no. 2, pp. 95-100. (In Russian)
14. Fogel L.J., Owens A.J., Walsh M.J. Artificial Intelligence through Simulated Evolution. New York: John Wiley & Sons. INc, 1966.
15. Rechenberg I. Evolutionstrategie: Optimierung technisher Systems nach Prinzipien der biologischen Evolution. Stuttgart:Fromman-Holzboog, 1973.
16. Koza J.R. Genetic Programming. Cambridge:MA:MIT Press, 1992.
17. Ivakhnenko A.G. Samoobuchayushchiesya sistemy raspoznavaniya i avtomaticheskogo upravleniya [Self-learning recognition and automatic control systems]. Kiev, Technology publ., 1969. (In Russian)
18. Bukatova I.L. Evolyucionnoe modelirovanie i ego prilozheniya [Evolutionary modeling and its applications]. Moscow, Science publ., 1979. (In Russian)
19. Banzhaf W., Brameier M. A comparison of linear genetic programming and neural networks in medical data mining. IEEE Transactions on Evolutionary Computation. 2001, no. 5, pp. 17–26.
20. Pavlov V.V., Ivin Yu.A., Pekhterev S.V., Macko I.I., Logunova O.S. Influence of the fractional composition of scrap metal on the performance of an arc steel furnace. Elektrometallurgiya [Electrometallurgy], 2011, no. 11, pp. 2-6. (In Russian)