•  

Reconstruction of the initial profile of the shock wave by Neural Network method

TitleReconstruction of the initial profile of the shock wave by Neural Network method
AuthorsD. A. Tarkhov1, T. A. Shemyakina1, A. R. Belyaewa1, T. T. Kaverzneva1, I. U. Zulkarnay2
1Peter the Great St. Petersburg Polytechnic University
2Bashkir State Universit
AnnotationRecovery process and identifying the causes of the accident or disaster is one of the important problems of technogenic safety. Blast waves are one of the objects of study such processes. The consequence of an explosion could be the formation of surface discontinuities currents, which are referred to as shock waves. In this work neural network modeling is considered in order to clarify the processes of occurrence and propagation of the shock wave. The object of study is the primary incident shock wave in an atmospheric shock tube. The pressure distribution profile at the initial time is based on the use of differential equations and experimental data.
Keywordsneural networks, nonlinear optimization, shock wave, differential equations, inverse problem
CitationTarkhov D. A., Shemyakina T. A., Belyaewa A. R., Kaverzneva T. T., Zulkarnay I. U. ''Reconstruction of the initial profile of the shock wave by Neural Network method'' [Electronic resource]. Proceedings of the XIII International scientific conference ''Differential equations and their applications in mathematical modeling''. (Saransk, July 12-16, 2017). Saransk: SVMO Publ, 2017. - pp. 205-209. Available at: http://conf.svmo.ru/files/deamm2017/papers/paper29.pdf. - Date of access: 10.12.2019.