Model-based internal short circuit detection of lithium-ion batteries using standard charge profiles

As a latent risk, soft internal short circuit (ISCr) occured in lithium-ion batteries may cause thermal runaway with fire and explosion. To secure battery safety for users, detection of soft ISCr is important. However, ISCr detection of existing model-based methods is totally dependent on type of load currents. Therefore, the ISCr must be detected with standard charge profiles instead of particular load currents, which are measured with high sampling rate and have big fluctuations. In this paper, using constant currents and terminal voltages for charge of the battery, a model-based ISCr detection algorithm is proposed. Equivalent circuit model of the battery is used to estimate ISCr resistance as a fault index. The proposed method is verified in simulations for varied ISCr faults. Accuracy of the estimated ISCr resistances is above 93.37%, thereby leading to early detection of the ISCr.