PROBLEMS AND PROSPECTS FOR THE DEVELOPMENT OF MODERN METHODS AND MEANS OF MONITORING THE THERMAL STATE OF LOCOMOTIVE TRACTION MOTORS

This article examines the current challenges and future directions in monitoring the thermal state of locomotive traction electric  motors, particularly in the context of increasing power demands and operational reliability requirements. Thermal  overload is identified as a key factor contributing to insulation degradation and motor failure. A classification of the main  sources of heat loss in DC traction motors is provided, along with an analysis of their impact on cooling efficiency and thermal  stability. The paper presents a comparative overview of modern monitoring approaches, including direct methods  (temperature sensors) and indirect methods (temperature estimation via electrical parameters). The specific challenges of  applying these methods to DC traction motors are highlighted, especially the technical limitations of installing sensors in  rotating components. The study emphasizes the role of thermal modeling, such as equivalent thermal circuit models and the finite element  method, which allow detailed visualization of temperature distribution within motor components and support  the design of effective thermal control systems. Special attention is given to the use of artificial neural networks and Digital  Twin technologies for predictive monitoring. These intelligent systems, trained on operational or laboratory data, can  accurately estimate internal temperatures in inaccessible zones and enable early fault detection. The integration of ANN-based models with real-time data from sensors and simulation results forms a foundation for adaptive condition-based maintenance. This work generalizes the latest research trends and outlines future prospects for developing hybrid monitoring systems that  combine classical models, sensor technologies, and machine learning algorithms, considering the structural and operational  features of traction motors used in domestic locomotive fleets.