Research on battery inconsistency evaluation based on improved local outlier factor and fuzzy matrix

Inconsistency between cells can seriously damage the overall performance of the battery pack and even cause safety accidents. An inconsistency evaluation model oriented to real-world vehicles is necessary owing to the contribution to battery management and maintenance. This paper proposes a comprehensive inconsistency evaluation strategy based on improved local outlier factor (ImLOF) algorithm and fuzzy matrix. Battery cell voltage is used as the basis for evaluation, and the abnormal rate distribution of each cell is analyzed by the established cell anomaly detection model. Each cell is set as an evaluation indicator to assess the state of inconsistency of the whole battery pack based on a fuzzy matrix and evaluation factor. The evaluation strategy is validated by full life cycle data from a real-world vehicle, and the results show that the proposed method can be effectively applied to evaluate the inconsistency of battery packs in real vehicles. • Inconsistency evaluation can provide important basis for battery management. • A inconsistency evaluation strategy based on ImLOF and fuzzy matrix is proposed. • Battery cell voltage is used as the basis for inconsistency evaluation. • Evaluation factor based on evaluation matrix is introduced to quantify inconsistency • The strategy is validated by full life cycle data from a real-world vehicle.