Accurate and reliable multifault diagnosis of battery packs is crucial to the safe operation of electric vehicles. To this end, this article proposes a systematically improved correlation coefficient (CC) method by utilizing multivariate statistical analysis and Bayesian probability theory under the framework of ensemble learning. Specifically, different window widths are first selected in an appropriate value range, and the CC signals between cross-cell voltages for each fixed window width are calculated to create different local submodels. Then, in each submodel, an independent component analysis-based fault diagnosis is implemented to obtain the local diagnostic result, and the results of all submodels are integrated as an ensemble fault probability (EFP) and an ensemble contribution rate (ECR) through Bayesian probabilistic ensemble interface. Once the EFP exceeds its threshold, a fault is considered to be detected and the ECR is subsequently applied along with the cross-cell sensor topology to identify the fault type (short-circuit or sensor fault) and locate the accurate position of the failed battery cell/sensor. In-depth theoretical analysis and sufficient comparative experiments on a real Lithium-ion battery packs test platform demonstrate that the proposed approach becomes more robust and reliable than the conventional CC-based methods, and also has better physical interpretability.

To facilitate the diagnosis of battery internal short circuit (ISC) using thermal behaviors, this work integrates several thermal effects, including the commonly ignored heat conduction hysteresis and radiation, to elaborate a lumped thermal model. Then, a pseudo-distributed model structure is built up to approximate the characteristics of real batteries by synthesizing multiple isomorphic electrical/thermal submodels with the extreme learning machine network. Besides, three kinds of configurable destructions are conducted to incur ISC consequences. From thermal and electrical model residuals, four ISC features are extracted and the multiclass relevance vector machine is utilized to assess ISC intensity, in which not only qualitative judgments are given but also quantitative confidences can be derived according to the posterior probabilities. Finally, experiments on 18650 Li-ion cells verify the reliability of the synthesized models and suggest that the diagnosis scheme can recognize ISC faults effectively with low grade and state misjudgment rates (14.59% and 3.13%, respectively).