Prerequisites for improving the methodology for rationing fuel consumption for maneuvering and economic traffic using machine learning models

Measures to normalize the consumption of diesel fuel for the operation of shunting locomotives are an integral part of the efficiency of the use of traction rolling stock. Due to the presence of problems in the formation of the fuel consumption rate, such as the performance of the shunting locomotive, the loading and technical condition of the diesel generator set, the operating conditions of locomotives, it is necessary to improve the existing fuel consumption rate determining methodology. The article presents the results of studies of the operating modes of shunting locomotives, the parameters of the spread of the values of the operating time of the diesel generator set in the load mode for shunting locomotives of the TEM2 and TEM18DM series are given. A mathematical model based on the bases of machine learning models is proposed, taking into account the degree of loading of the diesel generator set and the technical and economic characteristics of the locomotive series. The task of the proposed model is to determine the fuel consumption of a locomotive by an iterative calculation method, where the input parameters of the model are the performance indicators of the locomotive. It is established that the method of forming the fuel consumption rate for performing shunting operations should take into account the degree of loading of the diesel generator set and the technical and economic characteristics of the locomotive series. According to the results of the implementation of the proposed model for establishing specific fuel consumption rates for the studied locomotives based on the results of their operation, the average indicators of specific fuel consumption by locomotives by the positions of the driver's controller are obtained. The proposed method allows us to take into account the actual degree of loading of the diesel generator set and establish the correct fuel consumption rate for a certain period of operation of the locomotive, which represents the practical significance of the work. The presented research results are the beginning of work on the development of an intelligent system for diesel fuel rate rationing for the operation of shunting locomotives.