Hanyu Li

Postdoctoral Researcher @Lawrence Livermore National Laboratory

Livermore, CA, US

Developed a topology optimization framework to design full-cell batteries and supercapacitors and obtained up to 750% increase in energy storage and 80% decrease in overall cell resistance.• Collaborated with experimental team to produce optimized supercapacitor electrodes via additive manufacturing and confirmed through experiment a 10X improvement on energy storage and power output as compared to commercial supercapacitors.• Enabled massive parallelization capability on existing Discontinuous Galerkin discretization within in-house C++ finite element library with automatic differentiation, to model advection-dominated reactive scalar transport system and investigate optimal structure of tritium breeders that supports self-sustainable fusion reaction• Implemented shape optimization and level-set topology optimization approach to design porous electrodes with improved manufacturability by avoiding fuzzy material interfaces.• Developed a python package for large scale three dimensional Dolye-Fuller-Newman model to simulate various architected porous electrode structures for lithium-ion batteries, and identified the TMPS gyroid as the optimal structure that promotes uniform lithium deposition and reduces ohmic loss.• Developed thermomechanics and porous electrodes physics modules with automatic differentiation capability for in-house GPU-enabled finite element software package.• Composed proposal to improve convergence rate and prediction accuracy of physics-informed neural operator using a posteriori error estimates, with application to PDE-constrained optimization.