Search

In this paper, we present two deep learning-based hybrid data-driven reduced order models for the prediction of unsteady fluid flows. The first model projects the high-fidelity time series data from a finite element Navier-Stokes solver to a low-dimensional subspace via proper orthogonal decomposition (POD). The time-dependent coefficients in the POD subspace are propagated by the recurrent net (closed-loop encoder-decoder updates) and mapped to a high-dimensional state via the mean flow field and POD basis vectors. This model is referred as POD-RNN. The second model, referred to as convolution recurrent autoencoder network (CRAN), employs convolutional neural networks (CNN) as layers of linear kernels with nonlinear activations, to extract low-dimensional features from flow field snapshots. The flattened features are advanced using a recurrent (closed-loop manner) net and up-sampled (transpose convoluted) gradually to high-dimensional snapshots. Two benchmark problems of the flow past a cylinder and flow past a side-by-side cylinder are selected as the test problems to assess the efficacy of these models. For the problem of flow past a single cylinder, the performance of both the models is satisfactory, with CRAN being a bit overkill. However, it completely outperforms the POD-RNN model for a more complicated problem of flow past side-by-side cylinders. Owing to the scalability of CRAN, we briefly introduce an observer-corrector method for the calculation of integrated pressure force coefficients on the fluid-solid boundary on a reference grid. This reference grid, typically a structured and uniform grid, is used to interpolate scattered high-dimensional field data as snapshot images. These input images are convenient in training CRAN. This motivates us to further explore the application of CRAN models for the prediction of fluid flows.

In this work the numerical stability of a streamline singular hyperbolic/saddle critical point (HSP) and its relationship with the divergence of pressure force/fluid flux are numerically investigated at low Reynolds numbers. Three canonical configurations at different Reynolds numbers are considered: (a) an isolated circular cylinder; (b) side-by-side circular cylinders and (c) a near-wall circular cylinder. The objective is to investigate the behavior of a HSP subjecting to imbalanced shear-layer interaction and different Reynolds numbers. It is found that a HSP evolves along the shear-layer interfaces and imposes adverse pressure gradients, which potentially deteriorates near-wake stability. The inherent characteristics of a HSP is linked with net positive fluid-fluid divergence and fluid three dimensionality. A vorticity-free stagnant zone is formed around a HSP, which cut the kinetic energy supply of shear layers in wake, project third-dimensional fluid fluxes and develops three-dimensional streamwise braids. These findings are confirmed and explained via the quantification of the fluid-flux divergence, the hydrodynamic responses of cylinder(s) and the secondary enstrophy. The primary focus in this article is to reveal the subtle analytical relationships between HSP, divergence of pressure force/fluid flux, fluid three dimensionality and imbalanced shear-layer interaction. To the knowledge of authors, so far these analytical relationships have not been reported in literature.

In this paper, an end-to-end nonlinear model reduction methodology is presented based on the convolutional recurrent autoencoder networks. The methodology is developed in the context of the overall data-driven reduced-order model framework proposed in the paper. The basic idea behind the methodology is to obtain the low dimensional representations via convolutional neural networks and evolve these low dimensional features via recurrent neural networks in the time domain. The high dimensional representations are constructed from the evolved low dimensional features via transpose convolutional neural networks. With an unsupervised training strategy, the model serves as an end to end tool which can evolve the flow state of the nonlinear dynamical system. The convolutional recurrent autoencoder network model is applied to the problem of flow past bluff bodies for the first time. To demonstrate the effectiveness of the methodology, two canonical problems namely the flow past a plain cylinder and the flow past side-by-side cylinders are explored in this paper. Pressure and velocity fields of the unsteady flow are predicted in future via the convolutional recurrent autoencoder model. The performance of the model is satisfactory for both the problems. Specifically, the multiscale nature and the gap flow dynamics of the side-by-side cylinders are captured by the proposed data-driven model reduction methodology. The error metrics, the normalized squared error, and the normalized reconstruction error are considered for the assessment of the data-driven framework., Comment: arXiv admin note: text overlap with arXiv:1808.01346 by other authors