Spectral adjoint-based assimilation of sparse data in unsteady simulations of turbulent flows

Abstract

The URANS equations provide a computationally efficient tool to simulate unsteady turbulent flows for a wide range of applications. To account for the errors introduced by the turbulence closure model, recent works have adopted data assimilation (DA) to enhance their predictive capabilities. Recognizing the challenges posed by the computational cost of 4DVar DA for unsteady flows, we propose a 3DVar DA framework that incorporates a time-discrete Fourier transform of the URANS equations, facilitating the use of the stationary discrete adjoint method in Fourier space. Central to our methodology is the introduction of a corrective, divergence-free, and unsteady forcing term, derived from a Fourier series expansion, into the URANS equations. This term aims at mitigating discrepancies in the modeled divergence of Reynolds stresses, allowing for the tuning of stationary parameters across different Fourier modes. Our implementation is built upon an extended version of the coupled URANS solver in OpenFOAM, enhanced to compute adjoint variables and gradients. This design choice ensures straightforward applicability to various flow setups and solvers, eliminating the need for specialized harmonic solvers. A gradient-based optimizer is employed to minimize discrepancies between simulated results and sparse velocity reference data. The effectiveness of our approach is demonstrated through its application to flow around a two-dimensional circular cylinder at a Reynolds number of 3900. Our results highlight the method's ability to reconstruct mean flow accurately and improve the vortex shedding frequency through the assimilation of zeroth mode data. Additionally, the assimilation of first mode data further enhances the simulation's capability to capture low-frequency dynamics of the flow, and finally, it runs efficiently by leveraging a coarse mesh.

References in corpus (7)

• Turbulence Modeling in the Age of Data
• A Physics Informed Machine Learning Approach for Reconstructing Reynolds Stress Modeling Discrepancies Based on DNS Data
• A Random Matrix Approach for Quantifying Model-Form Uncertainties in Turbulence Modeling
• Nudging-based data assimilation of the turbulent flow around a square cylinder
• A variational data assimilation approach for sparse velocity reference data in coarse RANS simulations through a corrective forcing term
• Variational assimilation of sparse time-averaged data for efficient adjoint-based optimization of unsteady RANS simulations
• Mean- and unsteady-flow reconstruction with one or two time-resolved measurements

Cited by in corpus (1)

• Sparse data assimilation for under-resolved large-eddy simulations