Menu
Publications
2024
2023
2022
2021
2020
2019
2018
2017
2016
2015
2014
2013
2012
2011
2010
2009
2008
2007
2006
2005
2004
2003
2002
2001
Editor-in-Chief
Nikiforov
Vladimir O.
D.Sc., Prof.
Partners
doi: 10.17586/2226-1494-2023-23-6-1162-1170
Implementation of neural networks in the method of multilevel component circuits
Read the full article
';
Article in Russian
For citation:
Abstract
For citation:
Kochergin M.I. Implementation of neural networks in the method of multilevel component circuits. Scientific and Technical Journal of Information Technologies, Mechanics and Optics, 2023, vol. 23, no. 6, pp. 1162–1170 (in Russian). doi: 10.17586/2226-1494-2023-23-6-1162-1170
Abstract
The paper analyzes the features of representing artificial neural networks in Simulink and SimInTech. Examples of visual schemes (models) built in these modeling environments using neural network blocks are given. The following shortcomings of such representations are the lack of mechanisms: for carrying out structural optimization of neural networks, for combining them into ensembles, for training them synchronously with the simulation of the object model. It was noted that there are difficulties in using other tools, such as specialized Python libraries (Keras, PyTorch, etc.), the NeuroGenetic Optimizer (BioCompSystems) for building neural network control models. A method is shown to implement the representation of neural networks in the formalism of the method of multilevel component circuits, according to which the construction of models of an object and a control system is carried out in a visual language from ready-made blocks (components) with directional and non-directional connections. A technique has been developed for multilevel representation of neural network control models, which allows them to be combined with other tools of the component circuit method. Two options for representing neural networks are proposed: with an encapsulated structure and with a component structure. The first version of the representation is characterized by the compactness of the representation of the control model, the possibility of automated variation and optimization of the structure of the neural network, and the possibility of changing the structure of the network during the executing of the model within a computational experiment (scenario). The second option has the ability to perform detailed debugging and research of the network learning process, and the ability to construct a network of any structural complexity. The paper describes the main developed components with their connections: a neural network, a training block, an ensemble unit (bagging), a block for reading data from a file, a sampling block, a neural network layer (input, hidden, output). A multilevel computer model of the uncontrolled flight of a body (target) and the controlled flight of a projectile is presented as an example to illustrate the operation of the developed components to solve the problem of controlling a projectile to hit the target. The developed component libraries can be used as part of the MARS modeling environment to build multilevel control systems for objects of a multiphysics nature.
Keywords: neural networks, modeling, component circuits method, machine learning, Simulink, SimInTech, simulation environment MARS
References
References
- Korikov A.M. Intellectual management in technical systems. Scientific Bulletin of NSTU, 2014, no. 1(54), pp. 18–26. (in Russian)
- Onufrey A.Yu., Razumov A.V., Kakaev V.V. A method of optimizing the structure of hierarchical distributed control systems. Scientific and Technical Journal of Information Technologies, Mechanics and Optics, 2023, vol. 23, no. 1, pp. 44–53. (in Russian). https://doi.org/10.17586/2226-1494-2023-23-1-44-53
- Kolesnikov A.A., Kolesnikov Al.A., Kuzmenko A.A. ADAR method and theory of adaptive control in the tasks of synthesis of the nonlinear control systems. Mekhatronika, Avtomatizatsiya, Upravlenie, 2017, vol. 18, no. 9, pp. 579–589. (in Russian). https://doi.org/10.17587/mau.18.579-589
- Dayev Z.A., Kairakbayev A.K., Sultanov N.Z. Development of humidity of natural gas control system based on linear programming methods.Energy Systems, 2021,vol. 12, no. 1, pp. 233–245. https://doi.org/10.1007/s12667-019-00353-z
- Matveew M., Kabulova E., Kanisheva O. Fuzzy modeling for steel making processes. ARPN Journal of Engineering and Applied Sciences, 2015, vol. 10, no. 18, pp. 8271–8277.
- Evsyutkin I.V., Markov N.G. Deep artificial neural networks for forecasting debit values for production wells. Bulletin of the Tomsk Polytechnic University. Geo Assets Engineering, 2020, vol. 331, no. 11, pp. 88–95. (inRussian). https://doi.org/10.18799/24131830/2020/11/2888
- Dmitriev V.M.,Gandzha T.V.,Kochergin M.I.Matrix topological analysis of component circuits. Tomsk State University Journal of Control and Computer Science, 2023, no. 62, pp. 25–35. (in Russian). https://doi.org/10.17223/19988605/62/3
- Pantelic V., Postma S., Lawford M., Jaskolka M., Mackenzie B., Korobkine A., Bender M., Ong J., Marks G., Wassyng A. Software engineering practices and simulink: bridging the gap. International Journal on Software Tools for Technology Transfer, 2018, vol. 20, no. 1, pp. 95–117. https://doi.org/10.1007/s10009-017-0450-9
- Myznikova V.A., Ustimenko V.V., Chubar A.V. Fuzzy controllers construction in the simintech environment. Spacecrafts & Technologies, 2019, vol. 3, no. 1(27), pp. 22–27. (in Russian). https://doi.org/10.26732/2618-7957-2019-1-22-27
- Dmitriev V.M., Gandzha T.V. The method of constructing multilevel component chains for modeling chemical-technological systems. Proceedings of TUSUR University, 2017, vol. 20, no. 3, pp. 98–102. (in Russian). https://doi.org/10.21293/1818-0442-2017-20-3-82-86
- Dmitriev V.M., Gandzha T.V., Gandzha V.V., Maltcev Iu.I. SVID - System of Virtual Instruments and Devices. Tomsk, V-Spektr, 2014, 216 p. (in Russian)
- Gandzha T.V., Dmitriev V.M., Zaichenko T.N., Aksenova N.V. Computer modeling of chemical process systems with distributed parameters by means of component circuit method with non-uniform vector coupling. Petroleum and Coal, 2021, vol. 63, no. 1, pp. 68–77.
- Karelin A.E., Maystrenko A.V., Svetlakov A.A., Dmitriev V.M., Gandzha T.V., Aksenova N.V. Synthesis of an automatic control method for major oil pipelines based on inverse dynamics problem concept. Petroleum and Coal, 2018, vol. 60, no. 1, pp. 152–156.
- Tcvenger I.G., Nizamov I.R. Application of neural network controllers in the electric drive control systems. Bulletin of the Technological University, 2017, vol. 20, no. 8, pp. 111–114. (in Russian)
- Sukhanova N. Developing and applying neural network models in automating equipment control and technological processes. Automation and Modelling in Design and Management, 2022, no. 1(15), pp. 24–32. (in Russian). https://doi.org/10.30987/2658-6436-2022-1-24-32
- Kochergin M.I. Interpretation of the statechart diagram into a multilevel simulation language. Proceedings of TUSUR University, 2017, vol. 20, no. 4, pp. 122–125. https://doi.org/10.21293/1818-0442-2017-20-4-122-125
- Ezhov A.A., Cherepnev A.S. The uplearning of the Hopfield neural network: the search for a global minimum of a unctional and the model of rapid eye movement. Matematicheskoe modelirovanie, 2009, vol. 21, no. 5, pp. 10–20. (in Russian)
- Dmitrienko V.D., Zakovorotnyi A.Y. Solving the problem of additional training of classical neural networks. Automation of technologies and production, 2015, no. 4 (10), pp. 32–40. (in Russian).
- Fedotov D.V., Popov E.A., Okhorzin V.A. Optimisation of convolutional neural network structure with self-configuring evolutionary algorithm in one identification problem. Vestnik SibGAU, 2015, vol. 16, no. 4, pp. 857–863. (in Russian)
- Rybin P.K., Gorin R.V. Method of optimizing of the neural network structure for planning train arrival at pre-port stations. Vestnik transporta Povolzhya, 2017, no. 5 (65), pp. 55–63. (in Russian)
