Battery Grouping Based on Dynamic Gaussian Mixture Model

The lithium-ion battery group contains dozens of similar batteries, the inconsistency of characteristics of individual batteries can seriously affect the performance and life of the whole battery pack. In this paper, we propose a dynamic mixed Gaussian model for battery groping. Firstly, the static termination voltage and the capacity are used for clustering according to conventional method. Then clustering results utilize the whole battery discharge sequence and proposed the method for battery grouping. Finally, through actual production standards, the model structure is dynamically adjusted to meet the demand for the battery grouping number. Compared with the actual production, the proposed method proves the effectiveness of the method.