Neural network model for the solution of tasks of technical diagnostics of the transport telecommunication network

The research purposeis to develop an approach to assessing the reliability of a train radio communication channel based on the representation of a radio communication network in the form of a model that includes both traditional elements (various types of equipment, a wire channel) and a radio channel, which is proposed to be described by the availability of radio coverage, which varies depending on the location of the mobile subscriber. In the research methods of the theory of reliability, and, in particular, the assessment of the availability factor of the elements of the radio communication system and communication channels, dynamically changing during the movement of the mobile radio station. In addition, to assess the energy of radio channels, the theory of radio wave propagation was applied, including on the basis of industry recommendations for calculating radio communication networks of JSC Russian Railways. As result of the research, a method was proposed for assessing the dependence of the reliability of a train radio communication channel (in terms of readiness) on the track coordinate. In this case, the radio communication channel dynamically changes the composition of the elements when the mobile subscriber moves along the railway section. Simultaneously, for each point of the section, the calculation formula is also dynamically changed, which makes it possible to find the value of the readiness of the communication channel. The proposed approach allows to identify the sections of technological radio communication networks that are weakest from the point of view of communication reliability and take this information into account when planning work to improve the reliability of communication networks.

show abstract

Identification and modeling of processes for automated control of functional diagnostics of metal structures

Osadchyi¹,

Lurie²,

Boskin³

et al. 2020

Visnyk TNTU

View full text Add to dashboard Cite

The issues of control and identification of multidimensional and closed systems of metal structures diagnostics have been considered. The main varieties of identification and modeling of processes for managing the functional diagnostics of metal structures dealing with the improvement of the input information quality due to limited restrictions satisfaction, searching for additional a posteriori information in the process of dynamic changes of the controlled object, forecasting changes in the internal structure of the material have been specified and discussed in detail. The scheme of a priori information in the form of parameterized mappings of inputs and outputs interactions of the diagnostic system has been presented. The results of simulation are associated with the selection of variables measured diagnostic parameters. A methodology for quantitative estimations of posterior inflow information due to entropy values has been developed. Using the marked apertures of modelling and forecasting at each stage of identification, it is possible to obtain a mathematical model adequate to the real situation.

show abstract

Neural network model for the solution of tasks of technical diagnostics of the transport telecommunication network

Cited by 3 publications

References 0 publications

Substantiation of the Safety of Applying Mobile Communication Technologies for Train Automation

Substantiation of the Safety of Applying Mobile Communication Technologies for Train Automation

On the assessment of the reliability of the train radio communication channel

Identification and modeling of processes for automated control of functional diagnostics of metal structures

Contact Info

Product

Resources

About