Enhanced State Estimation for turbulent flows combining Ensemble Data Assimilation and Machine Learning

TL;DR

This paper introduces a combined data assimilation and machine learning approach to enhance turbulent flow state estimation, achieving accurate predictions with reduced computational costs by leveraging low-fidelity models and sparse sensor data.

Contribution

The paper presents a novel integration of ensemble data assimilation with physics-informed machine learning to improve turbulent flow estimation from low-fidelity models and sparse data.

Findings

The combined approach accurately predicts flow features in turbulent channel flow.

Significant reduction in computational costs compared to high-fidelity simulations.

Method demonstrates robustness and potential for future applications in fluid dynamics.

Abstract

A novel strategy is proposed to improve the accuracy of state estimation and reconstruction from low-fidelity models and sparse data from sensors. This strategy combines ensemble Data Assimilation (DA) and Machine Learning (ML) tools, exploiting their complementary features. ML techniques rely on the data produced by DA methods during analysis phases to train physics-informed corrective algorithms, which are then coupled with the low-fidelity models when data from sensors is unavailable. The methodology is validated via the analysis of the turbulent plane channel flow test case for $Re_\tau \approx 550$. Here, the low-fidelity model consists of coarse-grained simulations coupled with the Immersed Boundary Method (IBM), while observation is sampled by a highly refined body-fitted calculation. The analysis demonstrates the capabilities of the algorithm based on DA and ML to accurately predict the flow features with significantly reduced computational costs. This approach exhibits potential for future synergistic applications of DA and ML, leveraging the robustness and efficiency of ML models alongside the physical interpretability ensured by DA algorithms.