2016/06/27 by Paul G. Constantine, Alireza Doostan, Constantine, Paul G. +1 · 1 citation
Engineering · Materials Science · #Advanced Battery Technologies Research #Computational Physics (physics.comp-ph) #FOS: Mathematics #FOS: Physical sciences #Machine Learning in Materials Science #Numerical Analysis (math.NA) #VLSI and FPGA Design Techniques
paper · pdf · doi:10.48550/arxiv.1606.08770
openalex publication_date 2016/06/27 · openalex created_date 2022/10/07 · openalex updated_date 2026/08/01
Renewable energy researchers use computer simulation to aid the design of\nlithium ion storage devices. The underlying models contain several physical\ninput parameters that affect model predictions. Effective design and analysis\nmust understand the sensitivity of model predictions to changes in model\nparameters, but global sensitivity analyses become increasingly challenging as\nthe number of input parameters increases. Active subspaces are part of an\nemerging set of tools for discovering and exploiting low-dimensional structures\nin the map from high-dimensional inputs to model outputs. We extend linear and\nquadratic model-based heuristic for active sub- space discovery to\ntime-dependent processes and apply the resulting technique to a lithium ion\nbattery model. The results reveal low-dimensional structure and sensitivity\nmetrics that a designer may exploit to study the relationship between\nparameters and predictions.\n