, Russian Federation
, Russian Federation
, Russian Federation
The article highlights the significant increase of industrial capacities and automation of production, which requires taking effective management decisions by a responsible person. There have been outlined the important achievements of the scientists in application of the artificial neural networks in the various fields of activity and decision support systems involving the information analysis and processing with the results obtained. There has been proposed a review of publications on training artificial neural networks and on their efficient application in solving problems of classification, prediction and control. The most common structures of neural networks, their advantages and disadvantages, as well as the methods used to create training data arrays have been studied. A comparative analysis of using various structures of artificial neural networks and the effectiveness of existing teaching methods and the prospects for their use has been carried out. There has been defined the most preferred neural network topology for solving problems of fire safety management at the production facilities as an active decision support system. Using the analysis results, the most common and effective training methods have been identified, application of which is appropriate for developing and training various types of neural networks. The use of the technology is well grounded for reducing the errors in data processing, the financial costs for ensuring security, as well as for possible using the neural networks in the decision support systems to optimize these systems.
fire risk, artificial neural networks, decision support systems, fire safety
1. Ob utverzhdenii Metodiki opredeleniia raschetnykh velichin pozharnogo riska na proizvodstvennykh ob"ektakh. Prikaz MChS Rossii ot 10 iiulia 2009 g. № 404 [On approval of methods for determining calculated values of fire risk at production facilities: Order of the Ministry of Emergencies of Russia of July 10, 2009 No. 404]. Available at: https://base.garant.ru/196118/ (accessed: 16.07.19).
2. Rosenblatt F. The perceptron: a probabilistic model for information storage and organization in the brain. Psychological review, 1958, vol. 65, no. 6, p. 386.
3. Hopfield J. J. Neural networks and physical systems with emergent collective computational abilities. Proceedings of the national academy of sciences, 1982, vol. 79, no. 8, pp. 2554-2558.
4. Petrov E. P., Kharina N. L., Rzhanikova E. D. Matematicheskaia model' tsifrovykh polutonovykh izobrazhenii na osnove tsepei Markova s neskol'kimi sostoianiiami [Mathematical model of digital halftone images based on Markov chains with several states]. Nelineinyi mir, 2013, vol. 11, no. 7, pp. 487-492.
5. Hinton G. E. et al. Learning and relearning in Boltzmann machines. Parallel distributed processing: Explorations in the microstructure of cognition, 1986, vol. 1, no. 2, pp. 282-317.
6. Bourlard H., Kamp Y. Auto-association by multilayer perceptrons and singular value decomposition. Biological cybernetics, 1988, vol. 59, no. 4–5, pp. 291-294.
7. Bengio Y. et al. Greedy layer-wise training of deep networks. Advances in neural information processing systems, 2007, pp. 153-160.
8. LeCun Y. et al. Gradient-based learning applied to document recognition. Proceedings of the IEEE, 1998, vol. 86, no. 11, pp. 2278-2324.
9. Zeiler M. D., Krishnan D., Taylor G. W. et al. Deconvolutional networks. IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (San Francisco, USA, 13–18 June 2010), 2010, vol. 10, p. 7.
10. Voronov I. V., Politov E. A., Efremenko V. M. Obzor tipov iskusstvennykh neironnykh setei i metodov ikh obucheniia [Overview of artificial neural network types and their training methods]. Vestnik Kuzbasskogo gosudarstvennogo tekhnicheskogo universiteta, 2007, no. 3, p. 38-42.
11. Khaikin S. Neironnye seti [Neural networks]. Moscow, Vil'iams Publ., 2006. 1104 p.
12. Ostroukh E. N., Chegodar' M. Iu. Obuchenie neironnoi seti s ispol'zovaniem geneticheskikh algoritmov [Training neural network using genetic algorithms]. Izvestiia Iuzhnogo federal'nogo universiteta. Tekhnicheskie nauki, 2008, vol. 88, no. 11, pp. 86-87.
13. Tenenev V. A., Teneneva A. V. Obuchenie nechetkikh neironnykh setei geneticheskim algoritmom [Training fuzzy neural networks by using genetic algorithm]. Intellektual'nye sistemy v proizvodstve, 2010, no. 1, pp. 76-85.
14. Mishchenko V. A., Korobkin A. A. Ispol'zovanie geneticheskikh algoritmov v obuchenii neironnykh setei [Use of genetic algorithms in training neural networks]. Sovremennye problemy nauki i obrazovaniia, 2011, no. 6, pp. 116-119.
15. Gorbachevskaia E. N. Obuchenie iskusstvennoi neironnoi seti dlia zadach prognozirovaniia [Training artificial neural network for forecasting problems]. Vestnik Volzhskogo universiteta im. V. N. Tatishcheva, 2012, no. 2 (19), pp. 94-98.
16. Kotliarov E. V. Obuchenie neironnoi seti na osnove algoritma murav'inoi kolonii dlia zadachi klassifikatsii [Training neural network using ant colony algorithm for problem of classification]. Elektrotekhnіchnі ta komp’iuternіsistemi, 2012, no. 8, pp. 122-129.
17. Polupanov A. A., Garnaga V. V., Kol'tsov Iu. V. Evoliutsionnyi podkhod k obucheniiu iskusstvennoi neironnoi seti [Evolutionary approach to training artificial neural network]. Informatika, vychislitel'naia tekhnika i inzhenernoe obrazovanie, 2013, no. 3, pp. 10-15.
18. Lavrov E. A., Barchenko N. L. Neironnaia set' kak element sistemy upravleniia obucheniem [Neural network as element of training control system]. Teoriia i metodika obucheniia matematike, fizike, informatike, 2005, vol. 5, no. 3, pp. 148-152.
19. Fedosin S. A., Ladiaev D. A., Mar'ina O. A. Analiz i sravnenie metodov obucheniia neironnykh setei [Analysis and comparison of methods of training neural networks]. Vestnik Mordovskogo universiteta, 2010, no. 4, pp. 79-88.
20. Gritsai A. A. Opredelenie effektivnoi struktury vkhodnykh dannykh dlia obucheniia neironnoi seti resheniiu zadachi prognozirovaniia sprosa [Determination of effective input data structure for training neural network to solve demand forecasting problem]. Vestnik Tverskogo gosudarstvennogo universiteta. Seriia: Prikladnaia matematika, 2014, no. 2, pp. 95-106.
21. Andreeva E. A., Mazurova I. S. Chislennye metody obucheniia iskusstvennoi neironnoi seti [Numerical methods for training artificial neural network]. Uchenye zapiski Petrozavodskogo gosudarstvennogo universiteta, 2014, vol. 1, no. 8 (145), pp. 14-20.
22. Kotel'nikov V. V., Rykov A. N., Kozel'skaia S. O. Ispol'zovanie neironnykh setei s glubinnym obucheniem dlia prognozirovaniia i otsenki urovnia kritichnosti defektov konstruktsii [Using deep training neural networks to predict and evaluate level of criticality of structural defects]. Promyshlennye ASU i kontrollery, 2016, no. 12, pp. 39-45.
23. Sorokin A. S., Bondarev V. Iu., Krotova E. L. Sozdanie i obuchenie iskusstvennoi neironnoi seti dlia statisticheskogo otsenivaniia dannykh [Developing and training artificial neural network for statistical data estimation]. Vestnik UrFO. Bezopasnost' v informatsionnoi sfere, 2016, no. 2, pp. 29-32.
24. Ermakov B. S. Optimizatsiia roem chastits v obuchenii iskusstvennykh neironnykh setei [Particle swarm optimization in training artificial neural networks]. Sistemnyi analiz i logistika-2017, 2017, no. 1 (14), pp. 3-9.
25. Priezzhev I. I., Egorov S. V. Gibridnoe obuchenie neironnykh setei s tsel'iu prognoza parametrov neftegazovoi produktivnosti gornykh porod [Hybrid training of neural networks for predicting parameters of rock oil and gas productivity]. Seismicheskie tekhnologii-2017: materialy nauchno-prakticheskoi konferentsii (Moskva, 18–20 aprelia 2017 g.). Moscow, OOO «Izdatel'stvo Polipress», 2017. Pp. 205-208.
26. Piatakovich V. A., Vasilenko A. M., Mironenko M. V. Obuchenie neironnoi seti kak etap razrabotki ekspertnoi sistemy dlia klassifikatsii istochnikov fizicheskikh polei pri monitoringe akvatorii [Neural network training as stage in development of expert system for classifying sources of physical fields when monitoring water areas]. Vestnik Inzhenernoi shkoly Dal'nevostochnogo federal'nogo universiteta, 2017, no. 3 (32), pp. 138-149.
27. Chupakova A. O., Gudin S. V., Khabibulin R. Sh. Methods of using neural network technologies for modeling and conducting numerical experiments. Trudy 12-i Mul'tikonferentsii po problemam upravleniia (MKPU-2019) (Divnomorskoe, Krasnodarskii krai, 23–28 sentiabria 2019 g.). Rostov-na-Donu, Izd-vo Iuzhnogo federal'nogo universiteta, 2019. Pp. 144-146.
28. Chupakova A. O., Gudin S. V., Khabibulin R. Sh. Metodika sozdaniia neirosetevoi modeli dlia resheniia zadach klassifikatsii po optimizatsii pozharnykh riskov [Methods of developing neural network model for solving classification problems of fire risk optimization]. Problemy upravleniia i modelirovaniia v slozhnykh sistemakh: trudy XXI Mezhdunarodnoi konferentsii. Samara, OOO «Ofort», 2019. Pp. 551-554.