An extended approach to multiscale simulation of structured heat exchangers for enhanced battery cooling

The battery market experienced an annual growth of 30 – 40% in recent years, particularly due to the increasing demand for electric vehicles. Nowadays, the most common type are lithium-ion batteries due to the high energy density and number of possible charge cycles. To maximize the service life of a lithium-ion battery pack it is essential that the operating temperature is kept between 15 – 35 °C. Therefore, a battery thermal management system is indispensable. The present work introduces an extended multiscale simulation approach that can be seen as an extension of the classical heat exchanger cell method . It enables detailed results for field variables of an entire battery pack at relatively low computational cost. Only information about the flow pattern inside the heat exchanger and the inlet temperature of the system is required, making the methodology applicable to complex or newly developed structured heat exchangers. The battery pack is divided into repetitive units (unit cells), which are simulated in detail by coupling computational fluid dynamics and a finite element method (CFD-FEM). The individual unit cells are then interconnected to yield the results for the entire battery pack. This approach overcomes the computational limitations associated with large and complex geometries. The presented methodology is applied to a 18650 battery pack consisting of 448 individual battery cells and a cooling plate with sinusoidally structured cooling channels under different load and operating conditions. The present work demonstrates that the extended multiscale integration approach can be used flexibly to evaluate the cooling performance of a battery pack under different conditions, providing valuable information for the design and analysis of cooling plates. For validation, a full-scale simulation of the battery pack was conducted. As the total cooling heat flow and the total pressure drop resulting from the large-scale integration model only deviated from the full-scale simulation by 2.84% and 1.17%, respectively, it can be concluded that this extended multiscale simulation approach presents a good compromise between computational effort and model accuracy. • Multiscale simulation of lithium-ion battery packs. • Integration of structured heat exchangers into cooling plates. • Detailed flow and temperature fields for entire battery packs. • Extension of the cell method to unit cells with complex flow patterns. • Foundation for optimizing battery thermal management systems.