Developing reliable battery fault diagnosis and fault warning algorithms is essential to ensure the safety of battery systems. After years of development, traditional fault diagnosis techniques based on three-dimensional information of voltage, current and temperature have gradually encountered bottlenecks.
1. Online lithium-ion battery intelligent perception (LBIP): the model for thermal fault detection and localization was constructed, based on the Mask R–CNN instance segmentation model, and fine-tuned using a pre-trained model. Set the loss function, and optimize the network structure and network parameters in combination with the battery dataset;
Herein, the development of advanced battery sensor technologies and the implementation of multidimensional measurements can strengthen battery monitoring and fault diagnosis capabilities.
The integration of battery management systems (BMSs) with fault diagnosis algorithms has found extensive applications in EVs and energy storage systems [12, 13]. Currently, the standard fault diagnosis systems include data collection, fault diagnosis and fault handling , and reliable data acquisition [, , ] is the foundation.
This research built a lithium-ion battery thermal fault diagnosis model that optimized the original mask region-based convolutional neural network based on the battery dataset in both parameters and structure. The model processes the thermal images of the battery surface, identifies problematic batteries, and locates the problematic regions.
Not all emerging sensors have been systematically employed in research dedicated to battery safety. Some crucial parameters related to battery safety remain difficult to measure in-situ through sensors, such as lithium plating and internal micro-short circuits.