A semi-Markovian approach to evaluate the availability of low voltage direct current systems with integrated battery storage

This paper presents a methodology for modeling the availability of low-voltage direct current (LVDC) systems with battery storage, addressing the challenges posed by component failure through wear-out, the presence of many components, and the addition of battery storage to the system. While previous studies have explored various aspects of availability analysis for such systems, this methodology is unique in its comprehensive approach, combining component reliability models, semi-Markov availability models, the universal generating operator (UGO) method, and a stochastic battery reserve time analysis into a cohesive whole. This approach not only offers increased accuracy but also remains tractable and easy to understand. This methodology is validated by benchmarking against a Markov model implementation, demonstrating superior performance in terms of accuracy, where availability under-and over-estimations are avoided. This work has significant implications for maintenance planning and availability cost considerations of LVDC systems, while remaining computationally feasible.

Automated summary

This paper presents a methodology for modeling the availability of low-voltage direct current (LVDC) systems with battery storage. It addresses the challenges posed by component failure, the presence of many components, and the addition of battery storage to the system. The methodology combines component reliability models, semi-Markov availability models, the universal generating operator (UGO) method, and a stochastic battery reserve time analysis, offering increased accuracy while remaining tractable and easy to understand.