A coupled design optimization methodology for Li-ion batteries in electric vehicle applications based on FEM and neural networks

This paper focuses on developing a hybrid tool by using the finite element method (FEM) and the neural networks to improve the electrodes design for Li-ion battery better performances ad its lifetime. A design methodology approach based on a FEM based battery cell model is presented and applied in conjunction with the design of a neural network to optimize the electrodes design, in order to increase the usable capacity of a Li-ion battery over a range of charge-discharge current rates. It can be use for understanding the inter-dependence of chemical and mechanical degradation and coupling them to develop a useful tool to predict battery life. The effect of size, shape, charging and discharging conditions and material properties of electrode on the battery output voltage and temperature are analyzed.