IMPLEMENTATION OF A NEURO-FUZZY MODEL FOR EVALUATING THE EFFICIENCY OF THE TRANSPORT SYSTEM
Abstract
Introduction. Reliable and efficient transport systems provide a target level of quality of life. The indicators of the operation of its elements are the criteria for the optimization of the transport system. When designing modules of intelligent transport systems, it is advisable to lay down mathematical models that can be adapted to the changing operating conditions of the specified systems. The analysis of recent studies has shown that it is appropriate to use neural technologies and fuzzy inference systems to evaluate the efficiency of the transport system. Purpose. This article clarifies the mathematical models developed based on the results of previous studies for evaluating the energy efficiency of given configurations of the transport system by applying neurofuzzy technologies. Results. Based on the results of traffic monitoring under the specified conditions, 3 arrays of statistical data were formed: training, test, and control samples for input into the ANFIS system of the Matlab environment. Taking into account the number of input parameters, the generation of a neural network is performed using the method of subtractive clustering in the neuroFuzzyDesigner GUI. The network was trained using two methods: the error backpropagation method and the hybrid method. The hybrid method turned out to be more productive. The RMSE root mean square error on the training sample is 2.1153×10-8. The developed neural network made it possible to investigate the total influence of vehicle parameters on the energy efficiency of vehicles of the studied categories and transport technologies. According to the research results, it can be stated that the parameters of the vehicle are more sensitive to changes in their values for passenger transport than for cargo. Conclusions. A higher accuracy of the nonlinear model for evaluating the energy efficiency of vehicles was achieved compared to previous studies. The accuracy of the model has significantly increased due to the combination of fuzzy inference systems with neural network technologies. It is advisable to use this model in the subsystem for evaluating the performance of the functional elements of local transport systems as part of the ITS at the regional and state levels.
Downloads
References
2. Mateichyk V., Kostian N., Smieszek M., Mosciszewski J., Tarandushka L. Evaluating Vehicle Energy Efficiency in Urban Transport Systems Based on Fuzzy Logic Models. Energies. 2023. 16(2). 734. http://doi.org/10.3390/en16020734
3. Mateichyk V., Kostian N., Smieszek M., Gritsuk I., Verbovskyi V. Review of Methods for Evaluating the Energy Efficiency of Vehicles with Conventional and Alternative Power Plants. Energies. 2023. 16(17). 6331. http://doi.org/10.3390/en16176331
4. Zhou W., Wu N., Liu Q., Pan C., Chen L. Research on Ecological Driving Following Strategy Based on Deep Reinforcement Learning. Sustainability. 2023. 15. 13325. https://doi.org/10.3390/su151813325
5. Кашканова А.А. Застосування нечіткої нейронної мережі для визначення інформативності факторів впливу на реалізацію зчіпної здатності дороги і шини. Вісник машинобудування та транспорту. 2022. Вип. 15. № 1. С. 88–99. https://doi.org/10.31649/2413-4503-2022-15-1-88-99
6. Снитюк В.Є. Прогнозування. Моделі. Методи. Алгоритми. Київ : Маклаут, 2008. 364 с.
7. Пахомова В.М., Мандибура Є.С. Визначення оптимального маршруту в інформаційній мережі залізничного транспорту з використанням нейро-нечітких моделей. Наука та прогрес транспорту. 2019. № 5(83). С. 81–98. https://doi.org/10.15802/stp2019/184385
8. Zegai M.L., Bendjebbar M., Belhadri K., Lakhdari F. Adaptive Neuro-Fuzzy Speed Regulator Applied in Direct Torque Control for Induction Motor Drive Using Multilevel Inverter. International Review of Automatic Control (IREACO). 2016. 9(4). 182–191. https://doi.org/10.15866/ireaco.v9i4.9149
9. Пахомова В.М., Дмітрієв С.Ю. Розробка підсистеми оперативного прогнозування простоїв прибуваючих поїздів на основі ANFIS-системи. Інформаційно-керуючі системи на залізничному транспорті. 2013. № 4. С. 46–55. http://nbuv.gov.ua/UJRN/Ikszt_2013_4_10
10. Шевченко А.П. Моделі та методи прогнозування технічного стану засобів водного транспорту на основі м’яких обчислень : дис. ... доктора філософії : 271. Київ, 2020. 181 с.
