Iranian Journal of Electrical and Electronic Engineering

In this paper, a stochastic approach is proposed for reliability assessment of bidirectional DC-DC converters, including the fault-tolerant ones. This type of converters can be used in a smart DC grid, feeding DC loads such as home appliances and plug-in hybrid electric vehicles (PHEVs). The reliability of bidirectional DC-DC converters is of such an importance, due to the key role of the expected increasingly utilization of DC grids in modern Smart Grid. Markov processes are suggested for reliability modeling and consequently calculating the expected effective lifetime of bidirectional converters. A three-leg bidirectional interleaved converter using data of Toyota Prius 2012 hybrid electric vehicle is used as a case study. Besides, the influence of environment and ambient temperature on converter lifetime is studied. The impact of modeling the reliability of the converter and adding reliability constraints on the technical design procedure of the converter is also investigated. In order to investigate the effect of leg increase on the lifetime of the converter, single leg to five-leg interleave DC-DC converters are studied considering economical aspect and the results are extrapolated for six and seven-leg converters. The proposed method could be generalized so that the number of legs and input and output capacitors could be an arbitrary number.