Residual life estimation of batteries in DC power systems of traction substations

Designing of intelligent traction substations relies on IT solutions implemented to improve the operational reliability of equipment. One of the ways to improve the reliability of traction substations is to develop condition monitoring systems allowing to detect developing defects and predict the equipment service life. The reliability of traction substations is determined by a number of solutions, including the direct current (DC) power systems with batteries. Today, the residual life of batteries is not determined during maintenance and the condition is assessed during a scheduled maintenance. In addition, the systems that reduce the labor intensity of determining the residual life of batteries are used to increase the reliability of the control current system and traction substations. AIM: This study aimed to develop a battery model based on the digital twin concept, where the battery's residual life is determined based on the recorded current, voltage, and temperature data with sufficient accuracy for engineering design purposes. METHODS: The study uses static battery operation observations processed using the experimental design, mathematical statistics, and numerical analysis methods. RESULTS: We propose a digital twin of a traction substation battery allowing to determine the residual life by monitoring the basic battery performance indicators in various conditions. CONCLUSION: The findings may be used to develop control and management systems based on a digital battery twin with condition monitoring capabilities implemented in the intelligent traction substations.