Bio-Inspired Computational Design of Vascularized Electrodes for High-Performance Fast-Charging Batteries Optimized by Deep Learning

Slow ionic transport and high voltage drop (IR drop) of homogeneous porous electrodes are the critical causes of severe performance degradation of lithium-ion batteries at high charging rates. Nature has already provided plenty of examples like vascular structures to solve this kind of multi-variable transport optimization problem. In this presentation, it is numerically demonstrated that a bio-inspired vascularized porous electrode can simultaneously solve these two problems by introducing low tortuous channels and graded porosity, which can be verified by porous electrode theory. Despite the immense geometrical parameter space of the vascularized electrodes, recent progress in the machine learning algorithm has accelerated the optimization of the complicated topological structures. To optimize the vasculature structural parameters, artificial neural networks are employed to accelerate the computation of possible structures 84 times with high accuracy. Furthermore, an inverse-design searching library is compiled to find the optimal vascular structures under different industrial fabrication and design criteria, based on the structure-property relationships learned by artificial neural networks from the numerical simulation. The prototype delivers a customizable package containing optimal geometric parameters and their uncertainty and sensitivity analysis. Finally, the full-vascularized cell shows a 66% improvement in charging capacity compared to the traditional homogeneous cell under 3.2 C current density in a numerical simulation. This computational research provides an innovative methodology to solve the fast-charging problem in batteries and broaden the applicability of deep learning algorithms informed by numerical simulations to different scientific or engineering areas.