SOFTWARE COMPONENTS OF THE ON-BOARD COMPUTER SYSTEM OF A DIESEL TRAIN
DOI:
https://doi.org/10.26906/SUNZ.2019.5.011Keywords:
software, Brunovsky's form, mathematical model, neural networks, geometrical control theoryAbstract
When solving problems within the framework of the geometric theory of control, problems arise related to the difficulty of calculating the Lie derivatives, checking distributions for involutivity, searching for transformation functions that relate variables and equations of linear and nonlinear models. When performing these operations, a person has a need to perform too voluminous analytical calculations, which may cause the refusal to apply the geometric theory of control. This problem can be solved by using specialized software. We consider software for the on-board computer system of a diesel train, which is able to automate the necessary calculations, thereby significantly reducing the time it takes to linearize and search for the conversion functions for mathematical models by using the power of computer technology and neural networks. The aim of the work is the development of specialized software to perform linearization of mathematical models and search for conversion functions through the use of neural networks and the capabilities of a programming language, has a graphical interface for user interaction. Results. Using the capabilities of modern programming languages based on the proposed data processing algorithms and neural networks of the proposed structure, specialized software has been developed for converting non-linear mathematical models into the linear Brunovsky form and searching for conversion functions. When using the developed software, the speed of the linearization process, the search for conversion functions increases, and the graphical interface and comments that the software illuminates during operation make it possible to operate users who do not have special experience. A comparison of the results of modeling a nonlinear mathematical model with a linear mathematical model in the form of Brunovsky showed complete coincidence and confirmed the correctness of the theoretical positions and equivalence of the nonlinear and linear mathematical models. Conclusions. Specialized software has been developed for automating analytical transformations of the geometric control theory, for solving systems of partial differential equations, for defining transformation functions connecting variables of linear and nonlinear models. A number of objects were modeled that showed the operability of the software.Downloads
References
Заковоротный А. Ю. Синтез автоматизированной системы управления подвижным составом на основе геометрической теории управления и нейронных сетей [Электронный ресурс]: дис. ... д-ра техн. наук: спец. 05.13.07 / Александр Юрьевич Заковоротный; науч. консультант Дмитриенко В. Д. ; Нац. техн. ун-т “Харьков. политехн. ин-т”. – Харьков, 2017. – 433 с. – Библиогр.: с. 326 – 358.
Liu, B. & Gao, C. / Key Technologies of the Independent and Innovative Communications-Based Train Control Systems in China / Urban Rail Transit (2016) 2: 28. doi:10.1007/s40864-016-0035-z
Дмитриенко В. Д. Моделирование и оптимизация процессов управления движением дизель–поездов / В. Д. Дмитриенко, А. Ю. Заковоротный. – Х.: Изд. центр «HTMT», 2013. – 248 с.
Verma, A. & Pattanaik, K.K. / Multi-agent communication-based train control system for Indian railways: the behavioural analysis / Journal of Modern Transportation (2015) 23: 272. doi:10.1007/s40534-015-0083-1
Albrecht, T. (2013), “Human factor challenges in the development of a driver advisory system for regional passenger trains”, Rail human factors: supporting reliability, safety and cost reduction, Taylor & Francis, London, pp 129–138.
Программная компонента для поиска решений системы уравнений в частных производных в ГТУ методом группового учёта аргументов / В.Д. Дмитрієнко, А.Ю. Заковоротный, С.Ю. Леонов, Д.М. Главчев // Вісник Національного технічного університету “ХПІ”. Серія: Інформатика та моделювання. – Х.: НТУ “ХПІ”, 2019. – №13 (1338). – С. 61 – 72.
Аграчев А.А., Сачков Ю.Л. Геометрическая теория управления: моногр. – М.: ФИЗМАТЛИТ, 2015. – 392 c.
Исследование метода поиска функций преобразования нелинейных систем к эквивалентным линейным в геометрической теории управления / В.Д. Дмитрієнко, А.Ю. Заковоротный, Н.В. Мезенцев, Д.М. Главчев // Вісник Національного технічного університету “ХПІ”. Серія: Інформатика та моделювання. – Х.: НТУ “ХПІ”, 2018. – № 42 (1318). – С. 20 – 35.
Charu C. Aggarwal Neural Networks and Deep Learning: A Textbook.– IMB T.J. Watson Research Center, 2018. – 94 с.
Главчев Д.М. Нейронная сеть, распознающая группы изображений / В.Д. Дмитриенко, С.Ю. Леонов, Д.М. Главчев / Вісник НТУ “ХПІ”. – Харків: НТУ “ХПІ”, 2015, № 32 (1141). – С. 42 – 50.
ДСТУ ISO/IEC 25010:2016 Інженерія систем і програмних засобів. Вимоги до якості систем і програмних засобів та її оцінювання. Моделі якості системи та програмних засобів (ISO/IEC 25010:2011, IDT).
