About the forecasting of electrical energy consumption for traction of train using regression models differentiated by types of freight trains

In the current operational environment of the transportation sector, where energy-efficient organization of production processes plays a important role, it is essential to consider the characteristics of these processes that impact energy resource consumption. This article argues for the necessity of accounting for the features of different types of freight trains when establishing a system to manage energy resource expenditure for their traction. The article presents the results of testing the homogeneity of data samples on the value of specific energy consumption for traction of freight trains of various types in ranges of the average axle load of a car in a train at one of the sections of operation of locomotive crews of JSC Russian Railways. Using a stepwise regression method with variable exclusion, researchers developed regression models for different types of freight trains that describe their electricity consumption for traction. The forms of regression equations vary among different types of freight trains when compared within identical ranges of average axle load. This variation highlights the importance of a tailored approach to managing energy resource consumption for each train type. The evaluation of the quality of regression models, created from data on individual train trips and generalized data on freight train journeys, demonstrated that clustering data by train type enhances the accuracy of models used to calculate specific energy consumption for traction. Consequently, the research findings can aid in optimizing energy resource management and improving overall efficiency in railway operations.