Data-driven models refer to predictive or discriminative models established based on a large amount of historical or real-time data, using statistical methods, machine learning, or artificial intelligence. These models can model battery behavior without the need for explicit physical mechanism expressions. In the battery industry, such models are widely used in performance prediction, health assessment, safety monitoring, and intelligent control.
