Model-Based Fault Diagnosis for Large-Scale Marine Battery Systems

Methods of fault diagnosis and prediction have become essential to the safe and economically viable operation of large-scale marine battery systems. However, these systems impose specific requirements limiting the usability of common approaches. In this study, a model-based fault diagnosis proce-dure is proposed, which utilizes a statistical system model in order to calculate the behavior deviations in a system requiring minimal computational effort and data availability. The proposed method is validated on both measured and artificially generated data of a 2 MWh system on a battery-hybrid vessel. The results show that the approach can successfully identify fault occurrences from the signal sequence.