Fig. 1: State of health (SOH) evaluation for electric vehicles (EVs) based on historical field data.
From: Multi-modal framework for battery state of health evaluation using open-source electric vehicle data
a Raw operating data are uploaded to a cloud server from EVs. After data preprocessing, raw data are formulated into comma-separated values files. b Challenges in real-world EV SOH estimation. c Multi-modal feature engineering. Each SOH label is associated with one two-dimensional (2D) cell voltage feature map, two one-dimensional (1D) feature sequences related to charge capacity and temperature, and fifteen feature points. d Multi-modal SOH estimation framework. The multi-modal features are used as the input of the framework and consist of a 2D processing domain, a 1D processing domain, and a point feature processing domain.
