The given article considers the method of calculating the track geometry deformation with respect to uneven accumulation of residual deformations along the track. The technique proposes two significant changes in existing approaches to calculating the efficiency of the ballast layer. The transition from the approach of allowable stresses design in the ballast layer to the deformative approach of accumulations of track geometry deformations allows us to draw conclusions regarding the intervals of track tamping and the duration of ballast layer life cycle. The transition from the determinative to probabilistic approaches makes it possible to draw conclusions not only from the average unevenness, but also with regard to all possible facts of unevenness. The method is based on the mechanism of sudden and gradual deformations occurrence, which depends on a number of key factors: dynamic stresses on the ballast, non-uniformity of track elasticity, performance of current maintenance work. Based on the experimental studies results, the dependencies of sudden deformations and the intensity of gradual deformations on the level of stress on the ballast layer were established. The experimental results of the influence of the sub-ballast base elasticity on the intensity of accumulation of residual deformations are shown. On the basis of the developed method, the prediction of track geometry deterioration for a given structure of the track, the rolling stock and the permissible level of geometric deviations for track maintenance is presented.

In this paper, an application of computer vision and machine learning algorithms for common crossing frog diagnostics is presented. The rolling surface fatigue of frogs along the crossing lifecycle is analysed. The research is based on information from high-resolution optical images of the frog rolling surface and images from magnetic particle inspection. Image processing methods are used to pre-process the images and to detect the feature set that corresponds to objects similar to surface cracks. Machine learning methods are used for the analysis of crack images from the beginning to the end of the crossing lifecycle. Statistically significant crack features and their combinations that depict the surface fatigue state are found. The research result consists of the early prediction of rail contact fatigue.

The results of the study of the ballast layer consolidation after the work of ballast-tamping machines of different types are given in the article. The existing methods of determining the degree of consolidation of the ballast layer are analysed. The seismic method was improved by means of a complex dynamic and kinematic interpretation of the impulse response. For the dynamic interpretation with the use of statistical analysis, the features are selected so that they correspond to the degree of consolidation of the ballast layer. On the basis of researches, a device and software were developed that allow an automated evaluation of the ballast layer consolidation based on the kinematic and dynamic analysis of the measured impulse response. The measurements of the degree of the ballast layer consolidation after an operation of ballast-consolidation machines in different sequences allowed establishing the efficiency of the consolidation and the feasibility of the machines’ application.

Railway ballast tamping is one of the cost-expensive renewal and maintenance works of railway superstructure. The quality of ballast consolidation influences its resistance to residual deformations and long-term deterioration of track geometry. The process of ballast compaction along the sleeper under the vibration loading is complex and depends on many factors. The ballast flow processes under the vibration loading can produce both consolidation and un-consolidation of ballast material. The present study is devoted to the experimental investigation of ballast consolidation inhomogeneity. The method of ballast local consolidation measurement is proposed. The method is based on the velocity of impact wave propagation that is measured with device. The application of modern microcontroller and sensor techniques provided simple and reliable multi-point velocity measurements in a ballast layer. That enables well enough spatial resolution of ballast consolidation inhomogeneity. The measurement analysis has shown more than 4 times higher consolidation under the sleeper center than for unconsolidated ballast.

A machine learning approach for the recent detection of crossing faults is presented in the paper. The basis for the research are the data of the axle box inertial measurements on operational trains with the system ESAH-F. Within the machine learning approach the signal processing methods, as well as data reduction classification methods, are used. The wavelet analysis is applied to detect the spectral features at measured signals. The simple filter approach and sequential feature selection is used to find the most significant features and train the classification model. The validation and error estimates are presented and its relation to the number of selected features is analysed, as well.