论文和技术资料

The increasing demand for large battery packs, driven by the expanding markets for electric vehicles (EVs) and grid-scale energy storage system, necessitates the development rapid battery pack development pipelines such as simulation. We present an integrated multiscale simulation framework that enables simulation of critical parameters at battery cell level depending on the battery pack historical operating conditions. The critical parameters cover locale temperature, state of charge (SoC), and state of health (SOH). In a representative EV battery pack, we predict its performance over repeated cycling to identify the bottlenecks with respect to its lifetime. The cycling consists of a repeated standardised driving pattern to reflect a realistic driving scenario. We show via simulation that the bottlenecks can be attributed to (1) busbar resistivity due to variations in welding/joining process, (2) topology of battery pack, and (3) thermal management system. The framework integrates numerical methods with state-of-the-art machine learning techniques to achieve high fidelity simulations while maintaining low computational costs. COMSOL’s surrogate modelling is a key-enabler in the integration of multiple scales, which span a single-cell model to a battery pack model. The battery chemistry model is based on the classical Doyle-Fuller-Newman pseudo-2D formulation [1] incorporating degradation mechanism (solid-electrolyte interface layer formation) [2] using COMSOL’s custom PDE interfaces.