Development of organizational and technological construction diagrams of railway track facilities with the use of artificial neural networks

Abstract

The article is devoted to the research of the development procedure of organizational and technological construction diagrams of railway track facilities with the use of artificial neural networks. Organizational and technological diagrams allow to present the construction technological process in a visual form, determine the sequence of the facility erection in parts and methods of work production. In the digitalization conditions of organizational and technological documentation development there is a need for the rapid formation of visual materials for the construction works implementation. It is the visual aspects that allows to effectively assess the development and control of the technological processes implementation. The solution of these problems is possible by improving the existing engineering and technical support system of railway construction through the introduction of an engineering-intellectual support subsystem of technological processes of railway track facilities construction. One of the subsystem’s tasks is the effective use of automation tools with the artificial intelligence elements. The development procedure analysis of organizational and technological diagrams allowed to determine the artificial neural network application possibilities to obtain a visual representation of the technological process taking into account its spatial partitioning into grips. For this purpose the neural network structure, the program module realizing the algorithm of its training, the lists of input data and output parameters were developed. For the neural network training the data sample prepared in advance in the coded format based on the existing organizational and technological diagrams of railway track facilities construction was used. Based on the theoretical study results, practical development aspects of organizational and technological diagrams with the use of neural network on the example of the railway roadbed embankments erection are presented. The results given in the article are obtained in the course of dissertation research carried out by the author.