Multigrid sequential data assimilation for the large-eddy simulation of a massively separated bluff-body flow

TL;DR

This paper demonstrates that multigrid sequential data assimilation, using Ensemble Kalman Filter, can significantly enhance the accuracy of large-eddy simulations of complex turbulent flows around bluff bodies, even with sparse observations.

Contribution

It introduces a multigrid sequential data assimilation approach for LES that improves predictive accuracy in complex turbulent flows with limited observational data.

Findings

Data assimilation improves LES accuracy at reduced grid resolutions.

The method achieves symmetric statistical results despite sparse observations.

Significant accuracy improvements are observed away from sensor locations.

Abstract

The potential for data-driven applications to scale-resolving simulations of turbulent flows is assessed herein. Multigrid sequential data assimilation algorithms have been used to calibrate solvers for Large Eddy Simulation for the analysis of the high-Reynolds-number flow around a rectangular cylinder of aspect ratio 5:1. This test case has been chosen because of a number of physical complexities which elude accurate representation using reduced-order numerical simulation. The results for the statistical moments of the velocity and pressure flow field show that the data-driven techniques employed, which are based on the Ensemble Kalman Filter, are able to significantly improve the predictive features of the solver for reduced grid resolution. In addition, it was observed that, despite the sparse and asymmetric distribution of observation in the data-driven process, the data augmented results exhibit perfectly symmetric statistics and a significantly improved accuracy also far from the sensor location.