Aerodynamic Simulation Research Articles

A Machine Learning-Based Approach for Predicting Aerodynamic Coefficients Using Deep Neural Networks and CFD Data

Artificial Intelligence (AI) and Machine Learning (ML) are increasingly being adopted across various fields, including aerodynamics, exhibiting impressive results in complex computational processes and improving prediction accuracy. This study introduces a novel method for airfoil performance assessment through the development and training of a deep Artificial Neural Network (ANN), used for predicting aerodynamic coefficients and pressure distributions, leveraging comprehensive data obtained by using a Computational Fluid Dynamics (CFD) solver. First, an automated CFD solver was developed for obtaining the extensive dataset needed for the effective training of the ANN. The automation process consisted in the generation of a geometry and a mesh, along with the successful integration of the open-source SU2 solver for conducting the aerodynamic simulations, chosen for its versatility and straightforward integration. Once various airfoil analyses were performed and a comprehensive dataset was obtained, data was normalized and the model was trained. Throughout the training process, several model configurations were tested, varying different architectures, hyperparameters and layer settings, until the best-performing layout was chosen. After broad testing and validation, the optimal configuration was identified as being the one to demonstrate the lowest error rates and the most accurate predictions on both training and unseen data, highlighting the model’s generalization capabilities. This Machine Learning-based approach, used as a substitute for traditional methods, provides remarkable accuracy and robustness, capturing complex behaviors and significantly reducing the computational costs associated with CFD simulations.

Optimization of Mechanical Deployable Reentry Vehicle based on Multi-Fidelity Aerodynamic-Trajectory Coupling Model

As the aerospace industry continues to advance, the demand for round-trip transportation between Earth and space, as well as deep space exploration missions, is increasing. The mechanical deployable reentry vehicle, a complex multidisciplinary system, experiences parameter coupling between trajectory design and aerodynamic simulation. The calculation of trajectory and the design of aerodynamic layout directly impact the flight process. To address issues such as biased trajectory data obtained without considering disciplinary coupling and the high computational costs associated with overall aerodynamic layout optimization, this paper fully considers the impact of global aerodynamic drag on the flight trajectory. A multi-fidelity optimization model for reentry vehicles, named Multi-Fidelity Aerodynamics-Trajectory Coupling (MF-ATC), is proposed. The model uses a large number of low-accuracy model samples to control computational costs, without considering the coupling relationship, and a small number of high-accuracy model samples to ensure approximation accuracy, achieving fast and precise prediction of flight trajectories. The results demonstrate that the MF-ATC model improves trajectory prediction accuracy and efficiently accomplishes aerodynamic shape optimization, effectively enhancing the vehicle's performance. The model shows great potential for engineering applications.

Acoustic and aerodynamic evaluation of DLR small-scale rotor configurations within GARTEUR AG26

AbstractThis paper presents the activities performed in the GARTEUR Action Group HC/AG-26 to study the acoustic and aerodynamic characteristics of small rotor configurations, including the influence of the rotor-rotor interactions. This paper will focus on comparisons between numerical activities and wind tunnel results on a small rotor provided by DLR. The wind tunnel models included a Rotor/Rotor/Pylon in isolated, tandem and coaxial configuration. The wind tunnel experiments for acoustics were performed in DLR’s Acoustic Wind Tunnel Braunschweig (AWB) and PIV test were performed in CIRA within a joint CIRA/DLR test program. For simulations, the numerical approaches from each partner are applied. The aerodynamic simulations necessary for the aeroacoustic predictions are conducted with various fidelity numerical methods, varying from lifting line to CFD. The acoustic values on the microphone positions are evaluated using Ffowcs Williams/Hawking (FW-H) formulation by all partners. The acoustic and aerodynamic predictions are compared to test data, including performance, PIV and acoustic directivity.

Dust resuspension from the splash of a falling powder: A numerical aerodynamic simulation of a pellet falling onto a powder monolayer

Dust resuspension from the splash of a falling powder: A numerical aerodynamic simulation of a pellet falling onto a powder monolayer

Analysing a grid-forming storage hub for an offshore platform cluster supplied by wind energy

Abstract This paper performs an analysis of the electrification of an of-the-grid offshore oil and gas cluster of six platforms by means of a wind farm combined with a hydrogen energy hub. The latter, whose model is made publicly available, contains a grid-forming battery system and grid-following electrolysis and fuel cells units. State-of-the-art aerodynamic simulations are used for representing the wind farm power output, which encompasses wake losses and correlated wind fuctuations between turbines arising from farm-scale turbulence. One of the offshore platforms is represented by a detailed publicly available model named LEOGO. The other five platforms are represented as aggregated mixes of constant power and constant impedance loads. The computer simulations presented in the paper provide valuable insights into furthering the design and optimization of the hub concept. These insights include, among others, sizing of batteries depending on maximum ramp rates of fuel cells and electrolysers, identifying and mitigating instabilities induced by interactions among power electronic converters, and evaluating the consequences for the converters when those contribute with voltage support at the hub.

Accelerating aerodynamic simulations with a hybrid fine-tuned deep learning model

High-fidelity computational fluid dynamics simulations play an essential role in predicting complex aerodynamic flow fields, but their employment are hindered due to the high computational burdens involving fine spatial discretizations. While recent data-driven methods offer a promising avenue for performance improvements, they often face challenges related to excessive reliance on labeled data and insufficient accuracy. Consequently, we propose a novel hybrid model, which integrates a deep learning model into the fluid simulation workflow, harnessing the predictive capabilities to accelerate the fluid simulations. The acceleration is performed by a coarse-to-fine flow field mapping. To mitigate over-reliance on labeled data, the model is first pre-trained using pseudo-labeled data and then fine-tuned with a new designed attention mechanism. Acceleration efficiency of the hybrid model is demonstrated through two cases: aerodynamic simulations of an airfoil and a spherical blunt cone under varied operating conditions. Numerical experiments reveal that the proposed model achieves a substantial reduction in labeled data as well as prediction accuracy improvement, in comparison with traditional data-driven methods.

Enhancing improved advection upstream splitting method on triangular grids: A hybrid approach for improved stability and accuracy in compressible flow simulations

This paper introduces NAUSM+M+AUFS (New Improved Advection Upstream Splitting Method Plus Artificially Upstream Flux Vector Splitting), a novel hybrid computational scheme for simulating compressible flows on triangular grids. The AUSM+M (Improved Advection Upstream Splitting Method) method is enhanced through two key modifications to boost numerical stability and robustness in high Mach number and hypersonic flows. The first modification redefines the interfacial numerical sound velocity, reducing shock anomalies and improving shock-capturing by integrating velocity and characteristic sound speed parameters. The second modification addresses the insufficiency of the pressure flux dissipation term at supersonic speeds by introducing a formulation that increases dissipation proportionally to the Mach number, thereby enhancing performance in high-speed flows. These enhancements constitute the NAUSM+M method. The NAUSM+M+AUFS scheme combines the strengths of NAUSM+M and AUFS (Artificially Upstream Flux Vector Splitting) methods, particularly in overcoming the limitations of NAUSM+M in handling shock instabilities and the carbuncle phenomenon on structure triangular grids. A dynamic switching function adjusts the weighting between NAUSM+M and AUFS, optimizing accuracy and stability based on local flow conditions. Numerical tests demonstrate that NAUSM+M+AUFS significantly outperforms AUSM+M, NAUSM+M, and AUFS, effectively eliminating the carbuncle phenomenon and providing smooth shock wave contours. In steady flow analysis, the new hybrid method achieves convergence speeds comparable to AUFS and shows 15% to 45% superior convergence accelerating than AUSM+M, depending on the convergence rate. In addition, in steady flow analysis, the accuracy of NAUSM+M+AUFS is 46% better than that of AUFS. This approach represents a significant advancement, offering a robust, accurate, and efficient solution for high-speed aerodynamic simulations, with broad applicability across various compressible flow challenges.

A neural network approach for unstructured mesh quality evaluation

PurposeThe quality of the unstructured mesh has a considerable impact on the stability and accuracy of aerodynamic simulation in computational fluid dynamics (CFD). Typically, engineers spend a significant portion of their time on mesh quality evaluation to ensure a valid, high-quality mesh. The extensive manual interaction and a priori knowledge required to undertake an accurate and timely evaluation process have become a bottleneck in the idealized efficient CFD workflow. This paper aims to introduce a neural network-based quality evaluation approach for unstructured meshes to enable higher efficiency and the level of automation.Design/methodology/approachThe paper investigates the capability of deep neural networks for the quality evaluation of unstructured meshes. For training the network, we build a training dataset for mesh quality learning algorithms. The dataset contains a rich variety of unstructured aircraft meshes with different mesh sizes, densities, cell distribution, growth ratios and cell numbers to ensure its diversity and availability. We also design a neural network, AircraftNet, to learn the effect of mesh quality on the convergent properties of the numerical solutions. The proposed network directly manipulates raw point data in mesh source files rather than passing it to an intermediate data representation. During training, AircraftNet extracts non-linear quality features from high-dimensional data spaces and then automatically predicts the overall quality of the input unstructured mesh.FindingsThe paper provides a series of experimental results on GPUs. It shows that AircraftNet is able to effectively analyze the quality-related features like mesh density and distribution from the extracted features and achieve high prediction accuracy on the proposed dataset with even a small number of training runs.Research limitations/implicationsBecause of the limited training dataset, the research results may lack generalizability. Therefore, researchers are encouraged to test the proposed propositions further.Originality/valueThe paper publishes a benchmarking dataset for mesh quality learning algorithms and designs a novel neural network approach for unstructured mesh quality evaluation.

Neural fields for rapid aircraft aerodynamics simulations

This paper presents a methodology to learn surrogate models of steady state fluid dynamics simulations on meshed domains, based on Implicit Neural Representations (INRs). The proposed models can be applied directly to unstructured domains for different flow conditions, handle non-parametric 3D geometric variations, and generalize to unseen shapes at test time. The coordinate-based formulation naturally leads to robustness with respect to discretization, allowing an excellent trade-off between computational cost (memory footprint and training time) and accuracy. The method is demonstrated on two industrially relevant applications: a RANS dataset of the two-dimensional compressible flow over a transonic airfoil and a dataset of the surface pressure distribution over 3D wings, including shape, inflow condition, and control surface deflection variations. On the considered test cases, our approach achieves a more than three times lower test error and significantly improves generalization error on unseen geometries compared to state-of-the-art Graph Neural Network architectures. Remarkably, the method can drastically accelerate high fidelity numerical solver, achieving a five order of magnitude speedup on the RANS transonic airfoil dataset.

Multi-task Gaussian Processes based transient aerodynamic load reconstruction for maglev vehicle using acceleration response

The accurate estimation of aerodynamic loads is crucial for developing high-speed maglev trains, which can reach speeds of up to 600 km/h. Traditionally, this estimation has been achieved through computational aerodynamic simulation or direct pressure measurement. In this paper, we propose a novel framework for reconstructing transient aerodynamic loads on maglev vehicles using on-board acceleration measurements. In the framework, an inverse mathematical model that correlates the measured acceleration and external aerodynamic loads is derived from a well-calibrated maglev vehicle model. To avoid the ill-posed problem when solving the inverse mathematical model, a multi-task Gaussian Processes method is proposed, in which all reconstructed transient aerodynamic loads are treated as Gaussian Processes and the closed-form posterior distributions of these aerodynamic loads could be calculated. To validate the proposed framework, a set of transient vibration data collected from an operational maglev train passing through a double-track tunnel is utilized for load reconstruction. The results demonstrate that the framework offers a cost-effective and efficient means to obtain aerodynamic loads, highlighting its practical relevance for aerodynamic field testing in the context of evolving high-speed maglev technologies.

Aerodynamic Simulation Study of a Space Vehicle with Atmosphere-Breathing Electric Propulsion in a Free Molecular Gas Flow

Aerodynamic Simulation Study of a Space Vehicle with Atmosphere-Breathing Electric Propulsion in a Free Molecular Gas Flow

Assessment of Breakwater as a Protection System against Aerodynamic Loads Acting on the Floating PV System

Offshore floating photovoltaic (FPV) systems are subjected to significant aerodynamic forces, especially during extreme wind conditions. Accurate estimation of these forces is crucial for the proper design of mooring lines and connection systems. In this study, detailed CFD simulations were performed for various PV panel configurations, and using these CFD simulation correlations were developed to estimate lift and drag forces as a function of the number of panels. These correlations provide valuable tools for designing large-scale FPV systems with multiple PV modules. Additionally, this study investigates the potential of using breakwaters to reduce aerodynamic forces on FPV systems. Breakwaters, typically used to mitigate wave impacts, can also serve as wind barriers, significantly reducing wind forces before they reach the FPV array. Aerodynamic simulations with and without a breakwater were conducted using CFD to assess this effect. The results show a substantial reduction in lift and drag coefficients, especially for angles of attack up to 10 degrees, demonstrating the effectiveness of the breakwater in protecting the FPV system. However, beyond this threshold, the effectiveness of the breakwater of 2 m reduces. These findings highlight the importance of strategic breakwater placement and heights and their role in enhancing FPV system resilience. The insights gained from this study are critical for optimizing breakwater design and placement, ensuring the structural integrity and performance of FPV systems in varying environmental conditions. The data generated will also contribute to future design improvements for floating PV systems.

Aerodynamic Noise Simulation of a Super-High-Rise Building Facade with Shark-Like Grooved Skin.

The wind-driven aerodynamic noise of super-high-rise building facades not only affects the experience of use inside the building but also reduces the life cycle of building facade materials to some extent. In this paper, we are inspired by the micro-groove structure of shark skin with damping and noise reduction properties and apply bionic skin to reduce the aerodynamic noise impact of super-high-rise buildings. The aerodynamic noise performance of smooth and super-high-rise building models with bionic grooves is simulated via CFD to investigate the noise reduction performance of different bionic groove patterns, such as I-shape, ∪-shape, V-shape, and ∩-shape patterns, and their corresponding acoustic noise reduction mechanisms. This study showed that the bionic shark groove skin has a certain noise reduction effect, and the I-shaped groove has the best noise reduction effect. By applying bionic skin, the aerodynamic noise of super-high-rise buildings can be effectively reduced to improve the use experience and environmental quality of the buildings and provide a new research idea and application direction for the aerodynamic noise reduction design of building facades.

Aerodynamic Analysis and Simulation for a Z-Shaped Folding Wing UAV

The potential of folding wing unmanned aerial vehicles (UAVs) in enhancing the adaptability of complex missions is substantial. Analyzing the aerodynamic characteristics of these UAVs is vital for optimizing the design of their folding wing configuration and airfoil shape. In this study, a model of a Z-shaped folding wing UAV was established with the objective of enhancing UAV maneuverability. An integrated vortex lattice method (VLM) was employed to analyze the aerodynamic characteristics of the Z-shaped folding wing and numerical simulations were utilized to validate the obtained results, which prove the effectiveness of the integrated VLM in studying aerodynamic characteristics at a high angle of attack of the Z-shaped folding wing UAV. This is a new way to reduce computation time while maintaining accuracy to calculate such complex folding structures in high fidelity. Furthermore, the Z-shaped folding wing configuration demonstrates enhanced maneuverability at high angles of attack, and a Simulink flight simulation model based on the aerodynamic coefficients of the Z-shaped folding wing verifies this property. This analysis can properly predict the post-stall characteristics for the Z-shaped folding wing UAV to ensure the safety of the vehicle during flight.

A non-linear unsteady vortex-lattice method for rotorcraft applications

Abstract The present work aims to extend the capabilities of DUST, a mid-fidelity aerodynamic solver developed at Politecnico di Milano, for the aerodynamic simulation of rotorcraft applications. With this aim, a numerical element was implemented in the solver obtained by a coupling between the potential unsteady vortex lattice method and viscous aerodynamic data of aerofoil sections available from two-dimensional high-fidelity computational fluid dynamics (CFD) simulations or experimental wind-tunnel tests. The paper describes the mathematical formulation of the method as well as a validation of the implementation performed by comparison with both high-fidelity CFD simulation results and experimental data obtained over aerodynamics and aeroelastic fixed-wing benchmarks. Then, the method was used for the evaluation of the aerodynamic performance of two rotorcraft test cases, i.e. the full-scale proprotor of the XV-15 tiltrotor operating in different flight conditions and two propellers in tandem with overlapping disks. Simulation results comparison with high-fidelity CFD and data from wind tunnel tests highlighted the potentialities and advantages of the implemented approach to be used for the design and investigation of rotorcraft configurations characterised by consistent viscosity effects.

A novel real-time hybrid testing method for Twin-Rotor Floating Wind Turbines with Single-Point Mooring systems

The Floating Wind Turbines (FWT) are at the forefront of innovation, aiming to enhance power capacity and reduce costs. Among these innovations, the Twin-Rotor Single-Point Mooring Floating Wind Turbines (TR-SPM FWT) stands out. However, the unique structural design and dynamic characteristics of TR-SPM FWT pose challenges for traditional FWT model testing methods. To address this, a real-time hybrid testing (RTHT) method is proposed, employing the Multi-drive Aerodynamic Loading Simulator (MALS), specifically tailored for TR-SPM FWT model tests. MALS incorporates load distribution rules to accurately simulate aerodynamic loads, with its precision and response speed comprehensively evaluated. Furthermore, to enable multiple MALS units to replicate complicated aerodynamic behaviors concurrently, interference between channels and synchronization of units are examined. A load correction model is introduced and a RTHT system is devised, considering the floater yaw feedback of TR-SPM FWT. Ultimately, the proposed method is successfully applied to investigate the dynamic response of a TR-SPM FWT under diverse wind and wave conditions, demonstrating the reliability of the proposed test method and system.

Aerodynamic simulation on roof-mounted intelligent awareness system of intelligent vehicle

With the rapid development of intelligent driving technology, it is urgent to conduct research on the distribution and motion characteristics of snow particles in the area of the intelligent vehicle awareness system under harsh snow and wind conditions. In this paper, the Unsteady Reynolds-Averaged Navier-Stokes (URANS) was first used to conduct 1:12 scaled aerodynamic simulation of the intelligent vehicle. It is found that the error of aerodynamic drag coefficient between simulation and test is 5.6%, indicating that the current numerical simulation model can predict the aerodynamic characteristics of intelligent vehicle accurately. Based on this, URANS was used coupled with Lagrangian Multiphase Model (LMP) to investigate the snow accumulation near the roof-mounted intelligent awareness system under different vehicle speeds and snowfall intensities. The study found that with the increase of snowfall intensity from moderate to heavy and vehicle speed at 40 km/h, the incidence of snow particles on various components significantly increases, and the number of adhered particles also increased accordingly. The particle adhesion rate of the awareness system has decreased by 1.24%, while the number of adhered snow particles increased significantly at the back of the roof-mounted intelligent awareness system and the wake window. Under moderate snow condition, as the vehicle speed increased from 40 to 60 km/h, the number of incident particles and adhered particles in the awareness system have significantly decreased by 24,592 and 317,019, respectively, but the particle adhesion rate has increased by a significant 2.88%. Compared to the same snowfall intensity but low speed, the range of particle distribution is significantly reduced and more scattered.

Two-level parallel load balancing strategy for accelerating DSMC simulations in near-continuum gases

The Direct Simulation Monte Carlo (DSMC) algorithm is widely employed for simulating rarefied gas flows and is increasingly applied in near-continuum regimes for research and engineering purposes. However, its computational demands, notably load imbalance and extended simulation time, hinder widespread adoption. Addressing these challenges, this paper introduces the Two-Level parallel load balancing strategy. This novel approach combines thread-level and multi-process parallelism to enhance load balancing and reduce simulation time. Key features include a thread-level load-decoupling strategy implemented via OpenMP and a multi-process load balancing mechanism employing distributed memory via MPI. Building upon our previous [Formula: see text] [L. Li, W. Ren and B. Zhang, J. Aeronaut. Astronaut. Aviat. Ser. A 46, 88 (2014)] approach, the load balancing mechanism utilizes Stop At Risk (SAR) criteria for repartitioning with METIS. Additionally, a specialized data transmission mechanism utilizing MPI nonblocking communication minimizes global communication between processes. Validation and evaluation are performed using four hypersonic flow cases around a cylinder and sphere, demonstrating significant improvements. Notably, the proposed strategy achieves [Formula: see text] enhancement over the [Formula: see text] strategy under 512 CPU cores compared to 16 CPU cores, and reduces between-process communication time with [Formula: see text]. These advancements contribute to enhancing the effectiveness of the DSMC algorithm in near-continuum aerodynamic simulations.

Optimisation Design of Thermal Test System for Metal Fibre Surface Combustion Structure

The metal fibre surface combustion structure has the characteristics of strong thermal matching ability, short response time, and strong shape adaptability. It has more advantages in the thermal test of complex hypersonic vehicle surface inlet, leading edge, etc. In this paper, a method of aerodynamic thermal simulation test based on metal fibre surface combustion is proposed. The aim of the study was to create a uniform target heat flow on the inner wall surface of a cylindrical specimen by matching the gas jet flow rate and the geometry of the combustion surface. The research adopted the optimisation design method based on the surrogate model to establish the numerical calculation model of a metal fibre combustion jet heating cylinder specimen. One hundred sample points were obtained through Latin hypercube sampling, and a database of design parameters and heat flux was established through numerical simulation. The kriging surrogate model and the non-dominated sequencing genetic optimisation algorithm with elite strategy were adopted. A bi-objective optimisation design was carried out with the optimisation objective of the coincidence between the predicted and the target heat flux on the inner wall of the specimen. The results showed that the average relative errors of heat flow density on the specimen surface were 8.8% and 6% through the leave-one-out cross-validation strategy and the validation of six test sample points, respectively. The relative error values in most regions were within 5%, which indicates that the established kriging surrogate model has high prediction accuracy. Under the optimal solution conditions, the numerical calculation results of the heat flow on the inner wall of the specimen were in good agreement with the target heat flow values, with an average relative error of less than 5% and a maximum value of less than 8%. These results show that the optimisation design method based on the kriging surrogate model can effectively match the thermal test parameters of metal fibre combustion structures.

Numerical simulation of the transition flight aerodynamics of cross-shaped quad-tiltrotor UAV

In order to enhance the stability of the tilt transition process, a new configuration of Quad-Tiltrotor UAV was presented in this paper. Firstly, numerical simulation was used to calculate and analyze the aerodynamic interaction between the front rotor/fuselage/rear rotor during the transition state mode. The calculation model of the isolated rotor, front-rear rotor, front rotor-fuselage, and front rotor-rear rotor-fuselage combination states are established. Besides, the effects of pitch, roll, and yaw moment on the fuselage at different tilt angles are analyzed. It is concluded that the front rotor is the leading factor in the aerodynamic interference of the whole UAV in the different combination states. The research results can provide a reference for the optimization design of the overall layout, structure, and flight control strategy of the cross-shaped quad-tiltrotor UAV, and can also provide solutions for the logistics application of urban air traffic.