How to use neural network and web technologies in modeling complex technical systems

International Workshop "Advanced Technologies in Material Science, Mechanical and Automation Engineering – MIP: Engineering – 2019"

材料科学;机械制造;原子能学

Semenenko, M.G.^1 ; Kniazeva, I.V.^2 ; Beckel, L.S.^3 ; Rutskiy, V.N.^4 ; Tsarev, R. Yu.^4 ; Yamskikh, T.N.^4 ; Kartsan, I.N.^5

RANEPA, Kaluga Branch, Okruzhnaya street 4, Kaluga

248016, Russia^1

Financial University under the Government of the Russian Federation, Kaluga Branch, Chizhevsky street 17, Kaluga

248016, Russia^2

Bauman Moscow State Technical University, Kaluga Branch, Bazhenova street 2, Kaluga

248000, Russia^3

Siberian Federal University, Svobodny pr. 79, Krasnoyarsk

660041, Russia^4

Reshetnev Siberian State University of Science and Technology, Krasnoyarsky Rabochy Ave. 31, Krasnoyarsk

660037, Russia^5

关键词: Learning process;    Matlab simulations;    Model complexes;    Neural network method;    System of differential equations;    System of ordinary differential equations;    Technical objects;    Web technologies;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/537/3/032095/pdf
DOI  :  10.1088/1757-899X/537/3/032095

学科分类：材料科学（综合）

来源: IOP

【 摘 要 】

This paper discusses the problem of integrating modern methods of forecasting and modeling complex technical objects into the learning process. As an example, the problem of solving a system of ordinary differential equations is considered, which has significant practical application. In particular, solving a system of differential equations can be an essential part of patents. The neural network method to solve this problem by using Matlab simulation software and visual modeling tool Simulink is considered. Efficient cloud-based solution to ordinary differential equations is presented.