Optimization through rapid meta-model based transient thermal simulation of lithium ion battery cells

Today's battery electric vehicles (BEVs) use a broad variety of shapes and sizes of lithium-ion battery cells. One of the major concerns about a BEV battery pack is the process of aging, and the loss of driving range over the years of utilization. The phenomenon of capacity loss of lithium-ion batteries has been under scientific investigation for many years. Meanwhile, it is commonly accepted that the effects of temperature play an important role in this context [1-6]. Moreover, among others, the dynamic temperature behavior dictates the performance of the battery pack. The dimensioning of the cooling plays an important role in this manner [7]. In the early concept phase of a BEV and in this case the battery pack, many decisions must be made which have a major influence on the aforementioned aspects. Normally, simulation models are created to answer essential initial questions and to assist the cell selection process, as well as more general conceptual questions regarding the whole battery pack. More complex simulation models like 3D electrochemical or electrothermal finite element models require a great deal of computational power, especially when transient simulations are needed to answer important questions [8]. Therefore, a method is developed, which can speed up those simulations by a factor of more than 1,000 without losing to much accuracy of result. By creating metamodels with standardized pre-simulations of the models, the dynamic behavior can be emulated. It is shown, that by evaluating these meta-models in sequence, transient simulations can be executed without the time-consuming solving process of finite element models. Hence, it is possible to investigate a multitude of possible configurations in a much shorter and less computational expensive way.