Abstract by Hayden Oliver
Physics and Astronomy
Numerical Methods and their Application in Materials Discovery
Most of us were first introduced to numerical integration methods in second-semester calculus. We might remember the various rectangle rules, trapezoid method, or the higher-order Simpson's rule that we used to evaluate simple, one-dimensional integrations. Numerical integrals are a component of materials modeling, which is revolutionizing the world of material science and leading to many important discoveries and revelations about the world around us. In order to continue making such discoveries in the future, we need to improve time-intensive numerical integrals. We are developing new strategies to improve the error convergence of modern material simulation packages so that we can make more efficient use of computing resources. We demonstrate the efficacy of various adaptive mesh refinement and interpolation methods that show promising results for future materials discovery.