融合电压特征与一致性评分的动力电池热失控预警研究OA

As a core energy unit of electric vehicles,the traction battery directly determines vehicles' reliable operation.To quickly identify thermal runaway risks in traction batteries,this paper proposes a multi-dimensional early-warning assessment method that integrates multi-dimensional voltage outlier features with a consistency scoring framework based on vehicle driving segments.First,two types of voltage anomaly indicators,namely the vertical outlier average and the rate outlier index,were introduced.Then,they were integrated with voltage,temperature,and internal resistance to build a battery pack consistency scoring framework based on the integration of subjective and objective weighting.Next,a set of multi-level early-warning conditions was designed to enable rapid localization and risk confirmation of abnormal cells.Finally,real-vehicle operational data were used to validate the effectiveness of the proposed method,and a comparative analysis with the sample entropy-based warning method was conducted.Results demonstrate the proposed method achieves higher accuracy and computational efficiency in early warnings.