FR3D: Three-dimensional Flow Reconstruction and Force Estimation for Unsteady Flows Around Extruded Bluff Bodies via Conformal Mapping Aided Convolutional Autoencoders
Imperial College London（帝国理工学院）
Ali Girayhan Özbay, Sylvain Laizet
备注：
29 pages, 10 figures. Accepted at International Journal of Heat and Fluid Flow
In many practical fluid dynamics experiments, measuring variables such as velocity and pressure is possible only at a limited number of sensor locations, \textcolor{black}{for a few two-dimensional planes, or for a small 3D domain in the flow}. However, knowledge of the full fields is necessary to understand the dynamics of many flows. Deep learning reconstruction of full flow fields from sparse measurements has recently garnered significant research interest, as a way of overcoming this limitation. This task is referred to as the flow reconstruction (FR) task. In the present study, we propose a convolutional autoencoder based neural network model, dubbed FR3D, which enables FR to be carried out for three-dimensional flows around extruded 3D objects with different cross-sections. An innovative mapping approach, whereby multiple fluid domains are mapped to an annulus, enables FR3D to generalize its performance to objects not encountered during training. We conclusively demonstrate this generalization capability using a dataset composed of 80 training and 20 testing geometries, all randomly generated. We show that the FR3D model reconstructs pressure and velocity components with a few percentage points of error. Additionally, using these predictions, we accurately estimate the Q-criterion fields as well lift and drag forces on the geometries.
We propose a method to construct a reduced order model with machine learning for unsteady flows. The present machine-learned reduced order model (ML-ROM) is constructed by combining a convolutional neural network autoencoder (CNN-AE) and a long short-term memory (LSTM), which are trained in a sequential manner. First, the CNN-AE is trained using direct numerical simulation (DNS) data so as to map the high-dimensional flow data into low-dimensional latent space. Then, the LSTM is utilized to establish a temporal prediction system for the low-dimensionalized vectors obtained by CNN-AE. As a test…
A series of reduced-order numerical simulations on a specific bluff body type (v-gutters) in a subsonic duct flow is done to assess the unsteady wake dynamics. Two of the v-gutter's geometrical parameters are varied: the v-gutter's base angle ($θ$) and the size of a slit ($ξ$) at the leading-edge of the v-gutter. Turbulent flow kinematics and pressure field are analyzed to evaluate the unsteadiness at a freestream Mach number of $M_\infty = 0.25$ and a freestream Reynolds number based on bluff body's transverse length (L) of $Re_L=0.1 \times 10^6$. Five v-gutter angles are considered ($θ$, $^\…
We present an efficient deep learning technique for the model reduction of the Navier-Stokes equations for unsteady flow problems. The proposed technique relies on the Convolutional Neural Network (CNN) and the stochastic gradient descent method. Of particular interest is to predict the unsteady fluid forces for different bluff body shapes at low Reynolds number. The discrete convolution process with a nonlinear rectification is employed to approximate the mapping between the bluff-body shape and the fluid forces. The deep neural network is fed by the Euclidean distance function as the input a…
The impact of fluidic actuation on the wake and drag of a three-dimensional blunt body is investigated experimentally. Jets blowing tangentially to the main flow allow to force the wake with variable frequency and amplitude. Depending on the forcing conditions, two flow regimes can be distinguished. First, in case of broadband actuation with frequencies comprising the natural wake time scale, the convection of the jet structures enhances wake entrainment, shortens the length of the recirculating flow and increases drag. Secondly, at higher actuation frequencies, shear-layer deviation leads to …
A networked oscillator based analysis is performed for periodic bluff body flows to examine and control the transfer of kinetic energy. Spatial modes extracted from the flow field with corresponding amplitudes form a set of oscillators describing unsteady fluctuations. These oscillators are connected through a network that captures the energy exchanges amongst them. To extract the network of interactions among oscillators, amplitude and phase perturbations are impulsively introduced to the oscillators and the ensuing dynamics are analyzed. Using linear regression techniques, a networked oscill…
This paper is concerned with the development of a hybrid data-driven technique for unsteady fluid-structure interaction systems. The proposed data-driven technique combines the deep learning framework with a projection-based low-order modeling. While the deep learning provides low-dimensional approximations from datasets arising from black-box solvers, the projection-based model constructs the low-dimensional approximations by projecting the original high-dimensional model onto a low-dimensional subspace. Of particular interest of this paper is to predict the long time series of unsteady flow …
Cross-flow, or vertical-axis, turbines are a promising technology for capturing kinetic energy in wind or flowing water and their inherently unsteady fluid mechanics present unique opportunities for control optimization of individual rotors or arrays. To explore the potential for beneficial interactions between turbines in an array, coherent structures in the wake of a single two-bladed cross-flow turbine are examined using planar stereo particle image velocimetry in a water channel. First, the mean wake structure of this high chord-to-radius ratio rotor is described, compared to previous stud…
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 n…
This work develops a low-dimensional nonlinear stochastic model of symmetry-breaking coherent structures from experimental measurements of a turbulent axisymmetric bluff body wake. Traditional model reduction methods decompose the field into a set of modes with fixed spatial support but time-varying amplitudes. However, this fixed basis cannot resolve the mean flow deformation due to variable Reynolds stresses, a central feature of Stuart's nonlinear stability mechanism, without the further assumption of weakly nonlinear interactions. Here, we introduce a parametric modal basis that depends on…
Cluster-based reduced-order modelling of a mixing layer
提出了一种新的基于聚类的非定常流降阶建模策略。CROM结合了Gunzburger小组（Burkardt et al. 2006）开创的聚类分析和Eckhardt小组（Schneider et al. 2007）在流体动力学中引入的过渡矩阵模型。CROM构成了一个潜在的替代POD模型和推广的Ulam-Galerkin方法经典地用于动力系统，以确定一个有限秩近似的Perron-Frobenius算子。该策略分两步处理时间分辨的流快照序列。首先，快照数据被聚类成状态空间中的少量代表性状态，称为质心。这些质心将状态空间划分为互补的非重叠区域（质心Voronoi单元）。从标准算法出发，确定聚类的概率，并通过分析转移矩阵对状态进行排序。其次，使用马尔可夫过程动态地建模状态之间的转换。然后通过对马尔可夫过程的精细分析来提取物理机制，例如使用有限时间李雅普诺夫指数和熵方法。该CROM框架被应用到Lorenz吸引子（作为说明性的例子），空间发展的不可压缩混合层和三维湍流尾流的钝体的速度场。对于这些例子，CROM被证明是以无监督的方式识别非平凡的准吸引子和过渡过程。CROM有许多潜在的应用系统识别的物理机制的复杂的动态，比较流动演变模型，识别的前兆，希望和不希望的事件，并利用非线性驱动动力学的流量控制应用。
