YTY. A. Tsygankov

Y. A. Tsygankov

*This is an algorithmically generated author record based on publicly available information.

Y. A. Tsygankov has a total of 4 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Development of a mathematical model for assessing traffic accident risk on urban road networks

    by A. A. Jung, A. G. Shevtsova, D. А. Poleshchenko, Y. A. Tsygankov

    Introduction. Nowadays, personal mobility devices (PMDs), a new and rapidly developing mode of transport, are increasingly integrated into the urban road network (URN) of large cities and administrative centers. Numerous advantages are associated with this practice, including environmental friendliness, maneuverability, cost-effectiveness, social distancing, route flexibility, etc. However, due to the lack of infrastructure adapted for PMD movement, a sharp increase in road traffic accidents (RTAs) involving PMDs has been observed. To mitigate this negative effect, a mathematical model for assessing the probability of accident occurrence at each identified type of road network objects has been proposed in this research. Materials and Methods. This article presents the results of mathematical modeling based on the analysis of accident rates at various objects within an urban road network. To achieve these results, the following approaches were employed: empirical analysis of official statistics on RTAs involving PMDs, a classification approach, probabilistic modeling, and regression forecasting. The primary data used in the study included the number of RTAs involving PMDs and recorded at selected URN objects. Particular emphasis was placed on quantitative description and prediction of PMD-related accident rates in mixed urban environments, especially in areas originally not designed for vehicular traffic. Results. A mathematical model for evaluating accident risk at urban URN objects, accounting for the new transport mode – personal mobility devices – has been developed. The adequacy and predictive reliability of the constructed models are confirmed by coefficients of determination (R²) ranging from 0.64 to 0.92. Discussion and Conclusion. The proposed model enables not only assessments of the current accident risk at URN objects but also the identification of temporal trends in its evolution. Thanks to the high accuracy of approximation, the model provides a reliable forecast for the expected number of PMD-involved RTAs at different types of URN objects.

  • Ball mill states classification using competitive neural networks

    by D. A. Poleshchenko, Y. A. Tsygankov

    A competitive neural network is proposed to be used to process a vibration acceleration signal from pilot ball mill pins in order to determine the filling level of its drum. It is experimentally proved that the neural network is able to detect the mill state characterized by its drum filling level under the condition of disturbances. Experimental results show the feasibility of the proposed method.