Based on the voltage data, this paper develops a fault warning algorithm for electric vehicle lithium-ion battery packs based on K-means and the Fréchet algorithm. And the actual collected EV driving data are used to verify.
A sensor fault detection and isolation scheme for battery pack is presented. The proposed diagnostic scheme is with low computational effort. Adaptive extended Kalman filter is applied to help generate the residual. The residuals are evaluated by a statistical inference method. The effectiveness of the proposed scheme is experimentally validated.
However, the portability of the method is poor. The authors in ref (26) use the Kernel Principal Component Analysis (KPCA) approach to train a nonlinear data model for internal short-circuit detection of lithium-ion batteries. However, the method requires a large amount of historical data for offline training.
Therefore, this paper develops a data-driven early warning algorithm for lithium-ion batteries based on data driven for minor faults. Based on the voltage data, this paper develops a fault warning algorithm for electric vehicle lithium-ion battery packs based on K-means and the Fréchet algorithm.
Accurate evaluation of Li-ion battery safety conditions can reduce unexpected cell failures. Here, authors present a large-scale electric vehicle charging dataset for benchmarking existing algorithms, and develop a deep learning algorithm for detecting Li-ion battery faults.
It also provided a fault detection scheme for a single battery cell to detect the current or voltage sensor fault with extended Kalman filter, but with the difficulty of achieving the fault isolation. However, the topic of sensors FDI for a battery pack has not been addressed in the literature.