In the dissertation work the research of electromagnetic influence of rolling stock on rail circuits is carried out and methods of increase of functional safety of rail circuits are developed.
The scientific novelty of the dissertation is to solve the scientific and practical problem of improving the functional safety of rail circuits in the operation of new types of rolling stock by ensuring their electromagnetic compatibility.
For the first time: developed a mathematical model of the propagation of electrical noise from several units of EMF in a longitudinally inhomogeneous asymmetric rail line, which allows to determine the degree of influence of noise on the rail circuits; developed a method of improving the functional safety of rail circuits, based on the use of wavelet analysis of signal current and classifier based on artificial neural networks, which allows to detect excessive noise and distortion of signal current and prevent dangerous failures of broadcasters; developed a method to increase the functional safety of broadcasting by monitoring the signal current using an adaptive fuzzy neural network, which allows to detect and control the development of rail circuit defects, starting from their origin, resulting in timely prevention of system transition to the limit state transition to maintenance of broadcasting stations taking into account their actual condition.
Improved: mathematical model of hardware and software complex for measuring electrical interference from EMF in rails, which differs from existing ones in that it takes into account the mutual influence of system components, which allows to scientifically substantiate the choice of complex parameters to ensure the required dynamic range, accuracy and resolution amplitude, frequency and duration of electrical interference; mathematical model of the influence of harmonic interference on the track receiver of the tonal rail circuit, which differs from the existing ones in that it takes into account the amplitude, frequency and phase of traction current harmonics, which allows to determine the limiting parameters of electrical interference, exceeding which can cause failure in normal and control modes of operation.
The introduction substantiates the relevance of the topic, formulates the purpose, objectives of the study, outlines the scientific novelty, the practical significance of the results of the dissertation, information about testing and publication of research results.
The first section analyzes the functional safety of rail circuits and electromagnetic compatibility (EMC) of railway automation systems with traction network, in particular, with the electric rolling stock of railways, and also considers the impact of electromagnetic interference from electric rolling stock on the functional safety of rail circuits.
In the second section, the mathematical model of the hardware-software complex (APC) for measuring electromagnetic interference from EMF in rails is improved. The problem of rational choice of parameters of agro-industrial complex consists in strict requirements of regulatory documents concerning accuracy of measurement of parameters of harmonics of traction current (TS).
Theoretical and experimental studies allowed to scientifically substantiating the choice of parameters of the measuring complex, which allowed to provide the necessary resolution and accuracy of determining the parameters of harmonic interference, as well as to develop a mathematical model of electrical interference from several units of rolling stock in longitudinally inhomogeneous asymmetric line.
In the third section, a mathematical model of electrical interference from EMF in an asymmetric rail line is developed. The need for such a model is due to the fact that to investigate the influence of various factors on the EMC of new types of rolling stock with rail wheels during operational tests is almost impossible, in addition, such testing is quite expensive.
The fourth section presents a model of the effect of harmonic interference on the receiver of the tonal rail circuit. As a result of research the limiting current of disturbances at an input of the track receiver of TRK with frequency in a band ± (0..0,5) Hz concerning frequency of a signal current.
The fifth section considers methods for improving the functional safety of rail circuits under the influence of electromagnetic interference. The paper proposes two methods for solving this problem, namely the method of monitoring TRC using wavelet transform and classifier based on artificial neural networks and the method of continuous monitoring of signal current of TRC to detect distortions in the initial stages of their occurrence using adaptive fuzzy neural network.
The development of a method for monitoring defects in the process of their development allows to move further to the maintenance of rail circuits, taking into account the actual condition.
