Perelygin V.N., Perelygina A.U. Prognozirovanie ekspluatatsionnykh parametrov raboty zheleznoy dorogi dlya prodvizheniya mobilnykh sredstv diagnostiki [Forecasting of operational operation parameters of the railroad for advance of mobile diagnostic aids]. Molodaya nauka Sibiri: ehlektronnyj nauchnyj zhurnal [Young science of Siberia: electronic scientific journal], 2019, no. 2(4). [Accessed 15/04/19]
The rationale for creating a predictive non-linear model of performance indicators of the railway provided. This will minimize losses when planning path validation by means of mobile diagnostic tools. The prediction deficiencies revealed when using a non-linear regression equation. The basic diagnostic tools which limit train speeds presented. Existing neural network models considered. The optimal network structure based on a multilayer perceptron determined. The sensitivity analysis of the trained neural networks performed. A more flexible structure of the multilayer network installed compared to the regression equation.
1. Safarbakov M. A. basics of technical diagnostics of parts and equipment: textbook / A. M. Safarbakov, A. V. Lukyanov, and S. V. Pakhomov. – Part 1 – Irkutsk: Irkutsk State University Of Communications, 2007. – 128 p.
2. Safarbakov M. A. basics of technical diagnostics of parts and equipment: textbook / A. M. Safarbakov, A. V. Lukyanov, and S. V. Pakhomov. – Part 2 – Irkutsk: Irkutsk State University Of Communications, 2007. – 128 p.
3. Gapanovich, VA. Mathematical and information support of the URRAN/VA system. Gapanovich, A.M. Zamyshlyaev, I.B. Shubinsky//Reliability.-2013. - No. 1. - Page 3-11.
4. Krakow, Yu.M. Forecasting of side wear of rails as assessment procedure of their residual resource / Yu.M. Krakovsky, V.A. Nachigin//Control. Diagnostics. - 2010. - No. 6. - Page 30-35.
5. Perelygin V. N. Forecasting of technical speed / V.N. Perelygin, A.A. Antonov, A.Yu. Perelygina//Railway transport – 2014. - No. 12. – Page 19-20.
6. Advanced Multivariable Control Systems of Aeroengines // Eds.: Sun Jianguo, V. I. Vasilyev, B.G. Ilyasov, Beijing, China, BUAA Press, 2005. – 621 p.
7. Isermann, R. Fault Diagnosis of Machines via Parameter Estimation and Knowledge Processing. – Automatica 29, 1993, p.815–836.
8. E.S. Borisov About methods of training of multilayer feedforward neural networks. h 1 General provisions - http://mechanoid.kiev.ua.
9. E.S. Borisov About methods of training of multilayer feedforward neural networks. Part 2 Gradient methods of first order - http://mechanoid.kiev.ua.
10. Narendra, K. S. Neural Networks for Control. Theory and Practice // Proceedings of the IEEE, Vol. 84, No. 10, 1996, p.1385 –1405.
11. GNU Octave - http://www.gnu.org/software/octave.
12. Artyomkin, D. E. Development mathematical and the software of the automated forecasting of time series on the basis of neurocomputer technologies: yew. edging. technical sciences / D.E. Artyomkin. – Ryazan, 2003. – 140 pages.
13. Boletuses, V. P. Neural networks. Statistica Neural Networks. Methodology and technologies of modern data analysis / Accusative. Borovikov. – the 2nd prod. – M.: A hot line – the Telecom, 2008. – 392 pages.
14. Bestens, D. E. Neural networks and financial markets. Decision-making in trade operations / D.-E. Бэстенс, B. M. Van Den Bergh, D. Wood. – M.: TVP, 1997. – 236 pages.
15. Khaykin, Page. Neural networks: full course, the 2nd prod., испр.: The lane with English / S. Khaykin. – M.:OOO I. D. Williams, 2006. – 1104 pages.