In recent years, the data-driven turbulence model has attracted widespread concern in fluid mechanics. The existing approaches modify or supplement original by machine learning based on experimental/numerical data, order to augment capability of present models. Different from previous researches, this paper directly reconstructs a mapping function between turbulent eddy viscosity and mean flow variables neural networks completely replaces partial differential equation model. On other hand,...

Galloping is a type of fluid-elastic instability phenomenon characterized by large-amplitude low-frequency oscillations the structure. The aim present study to reveal underlying mechanisms galloping square cylinder at low Reynolds numbers ( $Re$ ) via linear stability analysis (LSA) and direct numerical simulations. LSA model constructed coupling reduced-order fluid with structure motion equation. relevant unstable modes are first yielded LSA, then development evolution these investigated...

This study proposes an improvement in the performance of reduced-order models (ROMs) based on dynamic mode decomposition to model flow dynamics attractor from a transient solution. By combining higher order (HODMD) with efficient selection criterion, HODMD criterion (HODMDc) ROM is able identify dominant patterns high accuracy. helps us develop more parsimonious structure, allowing better predictions dynamics. The method tested solution NACA0012 airfoil buffeting transonic flow, and its good...

Aeroacoustic noise is a major concern in wind turbine design that can be minimized by optimizing the airfoils shape rotating blades. To this end, we present framework for airfoil optimization to reduce trailing edge of Far-field evaluated using Amiet's theory coupled with TNO-Blake model calculate wall pressure spectrum and fast turn-around XFOIL simulations evaluate boundary layer parameters. The computational first validated NACA0012 at 0° angle attack. Particle swarm used find optimized...

It is well known that the occurrence of vortex-induced vibration (VIV) at a subcritical Reynolds number, which lower than 47, induced by fluid-structure interaction. However, for free flow, this phenomenon disappears number about 20. The current study provides an explanation to disappearance VIV capturing evolution purely fluid modes numbers between 12 and 55. To ensure accurate mode extraction, dynamic decomposition technique utilized. Results show stable von Kármán vortex shedding exists...

Transonic buffet is a phenomenon of aerodynamic instability with shock wave motions which occurs at certain combinations Mach number and mean angle attack, limits the aircraft flight envelope. The objective this study to develop modelling method for unstable flow oscillating waves moving boundaries, perform model-based feedback control two-dimensional by means trailing-edge flap oscillations. System identification based on ARX algorithm first used derive linear model input–output dynamics...

This work proposes a data-driven reduced-order modeling algorithm for complex, high-dimensional, and unsteady fluid systems with exogenous input control. is variant of dynamic mode decomposition (DMD), which an equation-free method identifying coherent structures complex flow dynamics. Compared existing methods, the proposed improves capability predicting evolution near unstable equilibrium state. The achieved by two steps. First, system matrix without identified standard DMD to represent...

Wind-tunnel experiment plays a critical role in the design and development phases of modern aircraft, which is always limited by prohibitive cost. In contrast, numerical simulation, as an important alternative research paradigm, mimics complex flow behaviors but less accurate compared to experiments. This leads recent emerging interest applying data fusion for aerodynamic prediction. particular, prediction aerodynamics with lower computational cost can be achieved fusing experimental...

In this work, we developed a novel framework for incorporating the near-wall non-overlapping domain decomposition (NDD) method with machine learning technique. It allows solution to be calculated Robin-type (slip) wall boundary condition on relatively coarse mesh and then corrected in region by solving thin boundary-layer equations fine subgrid. Through an estimated turbulent viscosity profile provided neural network, proposed can easily extended different turbulence models achieve...

Aerodynamic data can be obtained from different sources, which vary in fidelity, availability and cost. As the fidelity of increases, cost acquisition usually becomes higher. Therefore, to obtain accurate unsteady aerodynamic model with very low desired level accuracy, this paper proposes an multi-fidelity modeling framework. The approach integrates ideas fusion, modeling, nonlinear system identification machine learning. Data fusion reduces total generation for construction, while a...

Ice accretion on wind turbine blades and wings changes the effective shape of airfoil considerably deteriorates aerodynamic performance. However, unsteady performance iced is often difficult to predict. In this study, simulated under different pitching amplitudes reduced frequencies. order efficiently predict icing conditions, a multi-fidelity reduced-order model based multi-task learning proposed. The implemented using lift moment coefficient clean as low-fidelity data. Through few data...