Grid Sufficiency in Large Eddy Simulations as a Hypothesis Test

Abstract

The purpose of this Short Communication is three-fold: to make the case that the question of grid sufficiency in LES should be viewed as a hypothesis test, to demonstrate how this can be done using standard tools from statistics, and to illustrate how this leads to transparent ways to both judge grid sufficiency and to make decisions about how much averaging time is required to make that judgment.

Keywords:

• Grid sufficiency
• hypothesis test
• large eddy simulation
• quantities-of-interest
• statistical analysis

Acknowledgments

Insightful comments by Ana Kupresanin and Daniel Israel are gratefully acknowledged.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by grant DE-NA0003993 as part of the Predictive Science Academic Alliance Program (PSAAP) managed by the NNSA Office of Advanced Simulation and Computing (ASC).