The dissertation is devoted to the solution of a scientific and applied problem of improvement of service technology for automatic locomotive signaling by development and scientific substantiation of a method for monitoring the signal current by the spectral wavelet analysis for the detection defects of code current ALS and method of automation periodic control the ALS locomotive devices at the checkpoint. In the first section the traffic safety systems of trains operated in Ukraine are considered. A statistical analysis of the number of failures and malfunctions of all railway automation devices was performed. The main defects which affect the transmitting and receiving equipment of the system and cause a large number of interferences, as well as methods and means of protection against them are analyzed and classified. In the second section the method of controlling the influence of electromagnetic interference from PL on the operation of locomotive devices ALSN received further development, which differs from existing methods by measuring electrical interference in different types of ALSN code signals (green, yellow, red-yellow) with simultaneous measurement of the electromagnetic field along the railway track at the convergence section with PL followed by spectral analysis. The complex mathematical model of the communication channel between the rail circuit and ALSN input devices in the Simulink environment of the MatLAB package has been developed. The mathematical model of the signal-interference mixture operating at the input of the locomotive receiver of the system is given, pulse interference from local magnetic inhomogeneities of the rail line, quasi-harmonic interference from PL and stationary random interference, which are additively combined with each other, are analytically described. In the third section the methods of improving the analysis of ALS signal current using complex Fourier transform and short-term Fourier transform (window Fourier transform) are considered. The new method of monitoring the signal current of locomotive signaling has been developed and scientifically substantiated. It is based on the decoding the signal-noise mixture by means of spectral wavelet analysis, which allows to quickly detect defects of ALS code current and provides information about local signal features. Wavelet analysis of the obtained data showed that various deviations in the operation of the ALS system from the norms due to non-performance or non-compliance with maintenance and repair technology, sudden damage or failure directly in the process of train movement affects the time dependence and spectral component of the current, making it possible to recognize them. The wavelet analysis method can be used to filter ALS codes before decoding them in order to reduce the effect of interference on the signal current, as an additional method of signal processing. This method can also be used as a basis for developing the dynamic model of digital filter and ALS decoder with increased probability of receiving signals with interference, which will significantly increase the immunity to interference of ALS locomotive devices and bring the reliability of the system to a new level. In the fourth section it is proposed to improve the quality of ALS inspection by improving the maintenance of the system by automating the registration process of the changes in the readings of the locomotive traffic light and the algorithm for checking the ALS channel at the checkpoint. A method for monitoring the technical condition of ALS locomotive devices is proposed, which provides automatic registration of the readings of locomotive traffic lights by photo detector, measures the glowing time of traffic light lamps, the corresponding sequence of readings and, automatically analyzes the efficiency of system using a computer. A prototype of this device was developed, which is an automated hardware and software complex that controls the accuracy of transmission and decoding of ALS signals in the driver's cab at the checkpoint by comparing them with signals from track devices. The main results of the research of the dissertation are transferred for use in the distance of signaling and communication in Zaporizhzhia and introduced in the educational process of the Dnipro National University of Railway Transport named after Academician V. Lazaryan. Thus, the dissertation proposes effective methods for improvement of service technology of automatic locomotive signaling, which improve the quality of service locomotive devices, increase productivity and improve system immunity to interference, positively affecting railway safety, which becomes especially important in the conditions of the increased speeds of trains movement.
