Prediction Of Aerodynamic Coefficients Research Articles

Fast Prediction of Airfoil Aerodynamic Characteristics Based on a Combined Autoencoder

Aircraft airfoils are classified into two main categories: symmetrical and asymmetrical. Both types of airfoils have a significant impact on the flight performance and safety of the aircraft. The fast prediction of the aerodynamic coefficients and pressure distributions of airfoils is crucial for the design of aircraft. The traditional wind tunnel test and CFD methods have the disadvantages of high test cost and high time consumption. To solve these problems, a combined autoencoder (CAE) network is proposed in this paper, which can achieve the fast prediction of airfoil aerodynamic coefficients and pressure distributions. The network consists of an airfoil shape autoencoder (AE) network and a multilayer perceptron (MLP) network. Firstly, an autoencoder network reflecting the characteristics of the airfoil shape is established, and the effects of different latent variables on the performance of the autoencoder network are investigated. Then, the latent variables obtained from the autoencoder are concatenated with the inflow conditions such as the Reynolds number and the angle of attack to be used as inputs to the MLP network, and the aerodynamic coefficients of different airfoils in different inflow conditions are predicted. The effects of various latent variable inputs, as well as the direct input of the airfoil shape into the MLP network, on the prediction performance of aerodynamic coefficients are compared and analyzed. The optimal aerodynamic coefficient prediction network is then obtained. Finally, the CAE network is also applied to predict the pressure distributions of different airfoils in different inflow conditions and the effects of different latent variables and input conditions on the prediction performance of the pressure distributions are analyzed and compared with the advantages and disadvantages of the CAE network and the conditional variational autoencoder (CVAE) network. The results demonstrate that the proposed method is capable of accurately predicting aerodynamic characteristics in a shorter time, offering a valuable reference for the fast and efficient design of aircraft airfoils.

On the Fidelity of RANS-Based Turbulence Models in Modeling the Laminar Separation Bubble and Ice-Induced Separation Bubble at Low Reynolds Numbers on Unmanned Aerial Vehicle Airfoil

The operational regime of Unmanned Aerial Vehicles (UAVs) is distinguished by the dominance of laminar flow and the flow field is characterized by the appearance of Laminar Separation Bubbles (LSBs). Ice accretion on the leading side of the airfoil leads to the formation of an Ice-induced Separation Bubble (ISB). These separation bubbles have a considerable influence on the pressure, heat flux, and shear stress distribution on the surface of airfoils and can affect the prediction of aerodynamic coefficients. Therefore, it is necessary to capture these separation bubbles in the numerical simulations. Previous studies have shown that these bubbles can be modeled successfully using the Large Eddy Simulation (LES) and Direct Numerical Simulation (DNS) but are computationally costly. Also, for numerical modeling of ice accretion, the flow field needs to be recomputed at specific intervals, thus making LES and DNS unsuitable for ice accretion simulations. Thus, it is necessary to come up with a Reynolds-Averaged Navier–Stokes (RANS) equation-based model that can predict the LSBs and ISBs as accurately as possible. Numerical studies were performed to assess the fidelity of various RANS turbulence models in predicting LSBs and ISBs. The findings are compared with the experimental and LES data available in the literature. The structure of these bubbles is only studied from a pressure coefficient perspective, so an attempt is made in these studies to explain it using the skin friction coefficient distribution. The results indicate the importance of the use of transition-based models when dealing with low-Reynolds-number applications that involve LSB. ISB can be predicted by conventional RANS models but are subjected to high levels of uncertainty. Possible recommendations were made with respect to turbulence models when dealing with flows involving LSBs and ISBs, especially for ice accretion simulations.

Aerodynamic center of a wing at low Reynolds number

The location of the aerodynamic center, with respect to the center of gravity, determines the longitudinal static stability of an aerial vehicle. While most studies in the past focused on high Reynolds number (Re) flows, we consider flows at low Re (100≤Re≤6000) past an end-to-end wing and a finite wing of various semi-aspect ratio (0.25≤sAR≤5). It is shown that two-dimensional computations, even for an end-to-end wing, are inadequate for accurate prediction of aerodynamic coefficients. The chordwise pressure distribution as well as the variation of aerodynamic coefficients with angle of attack (α) at low Re are significantly different from that at high Re. Beyond a certain α, the flow is associated with a large recirculation zone, resulting in a non-linear variation of the aerodynamic coefficients with α. The conventional definition of the aerodynamic center is too restrictive at low Re. Owing to the non-linear variation of the pitching moment with change in α, it is not possible to determine one single location where it is invariant over a large range of α. The aerodynamic center is, therefore, determined over various sub-regimes of α. It varies significantly over the sub-regimes. The wing-tip vortex significantly affects the vortex shedding, flow structures, and pressure distribution on a finite wing. The locations of the aerodynamic center and center of pressure move fore, toward the leading edge, with a decrease in aspect ratio compared to that for an end-to-end wing.

Aerodynamic prediction and roughness implementation toward ice accretion simulation

Ice accretion in aircraft leads to significant separation and surface roughness. Therefore, the simulation of icing is highly intricate and challenging. The aerodynamic prediction of an iced wing requires an accurate turbulence model. In this paper, a modified transition model for the Reynolds-averaged Navier–Stokes equation is developed. The model is implemented using a linear eddy viscosity transition model as a baseline. It is modified to accurately predict the flow separation and simulate the influence of roughness. Different cases are numerically tested based on the modified model, including two types of iced airfoils (glaze ice and rime ice) and two rough geometries. The results indicate that the model demonstrates engineering accuracy in terms of the flow separation and aerodynamic coefficient prediction. Moreover, this model could be adapted for a wide range of ice shapes. In addition, the present model is capable of accounting for the influence of surface roughness, leading to significant improvements in heat transfer coefficients that match the experimental data well.

Geometry Representation Effectiveness in Improving Airfoil Aerodynamic Coefficient Prediction with Convolutional Neural Network

Many applications use symmetric or asymmetric airfoils, such as aircraft design, wind turbines, and heat transfer. Each airfoil has different aerodynamic coefficients. Obtaining the aerodynamic coefficients is a must to optimize the airfoil design. Engineers use various methods to get the airfoil aerodynamic coefficients. A prediction method is an approximation approach that effectively reduces time and cost. This article uses convolutional neural networks (CNN) to get approximation values of those coefficients. In CNN, we collect 8920 aerodynamic coefficients for 223 NACA 4 as labels in datasets by using XFOIL at  and  with varying angles of attacks starting  to  with increment of . The simulation results are compared to the experiment using E387 airfoil for validation. Then, airfoil geometries as part of input datasets were transformed into Grayscale and RGB images using the signed distance function (SDF) and mesh algorithm. Each airfoil representation was trained using an 80% dataset and tested using a 20% dataset with Adam as an optimizer to generate each prediction model using modified LeNet-5. We use three different layer depths in modified LeNet-5 to obtain the optimal layer number. There is no remarkable improvement when varying the depth layers, so four layers are used instead. Simulation results show that using an SDF with Fast Marching Method on CNN predicts the most effective for the airfoil’s lift, drag, and pitch moment coefficient with varying angles of attack simultaneously. One can extend the method by using SDF to recognize different flow conditions.

An Intelligent Method for Predicting the Pressure Coefficient Curve of Airfoil-Based Conditional Generative Adversarial Networks.

Recently, extensive studies have focused on analyzing aerodynamic performance due to its important impact on aircraft design. Most of these works compute the aerodynamic coefficient of the airfoil through computational fluid dynamics (CFD) simulation, which is too time-consuming. To reduce the computational time required, some intelligence-based methods have been presented. However, these methods also suffer from certain issues. First, most of them directly implement existing machine learning methods used to predict the aerodynamic coefficient without adding any improvements. Second, some methods convert the airfoil shape and aerodynamic curves into images, which may lead to curve distortion and the introduction of noise. Third, some methods learn the relationship between the airfoil shape and aerodynamic coefficients but ignore the influence of initial inflow conditions. Accordingly, to address these issues, we propose an intelligent method for predicting the pressure coefficients (Cp) of airfoil based on a conditional generative adversarial network (cGAN). More specifically, we first present a two-step data augmentation strategy designed to expand the original airfoil dataset. Subsequently, we design a novel cGAN-based neural network to predict the Cp curve. To the best of our knowledge, this is the first work to apply generative adversarial network (GAN) to aerodynamic coefficient prediction. Moreover, we design a new loss function to train our network. Extensive experimental results demonstrate that the Cp curve predicted by our method is very close to that generated via CFD simulation. More importantly, our method achieves a speedup close to 1000x compared with CFD simulation.

Efficient prediction of turbulent flow quantities using a Bayesian hierarchical multifidelity model

Multifidelity models (MFMs) can be used to construct predictive models for flow quantities of interest (QoIs) over the space of uncertain/design parameters, with the purpose of uncertainty quantification, data fusion and optimization. For numerical simulation of turbulence, there is a hierarchy of methodologies ranked by accuracy and cost, where each methodology may have several numerical/modelling parameters that control the predictive accuracy and robustness of its resulting outputs. Compatible with these specifications, the present hierarchical MFM strategy allows for simultaneous calibration of the fidelity-specific parameters in a Bayesian framework as developed by Gohet al.(Technometrics, vol. 55, no. 4, 2013, pp. 501–512). The purpose of the MFM is to provide an improved prediction, mainly interpolation over the range covered by training data, by combining lower- and higher-fidelity data in an optimal way for any number of fidelity levels; even providing confidence intervals for the resulting QoI. The capabilities of the MFM are first demonstrated on an illustrative toy problem, and it is then applied to three realistic cases relevant to engineering turbulent flows. The latter include the prediction of friction at different Reynolds numbers in turbulent channel flow, the prediction of aerodynamic coefficients for a range of angles of attack of a standard airfoil and the uncertainty propagation and sensitivity analysis of the separation bubble in the turbulent flow over periodic hills subject to geometrical uncertainties. In all cases, based on only a few high-fidelity data samples, the MFM leads to accurate predictions of the QoIs. The result of the uncertainty quantification and sensitivity analyses are also found to be accurate compared with the ground truth in each case.

Prediction of aerodynamic coefficients of iced conductors based on composite image and convolutional neural network

The aerodynamic force of iced conductors is a key factor in the galloping study of transmission lines. At present, it is less efficient to obtain the aerodynamic coefficients of iced conductors by wind tunnel tests or numerical simulations. In this paper, an efficient method is proposed to predict the aerodynamic coefficients of iced conductors based on a convolutional neural network. The linear mapping relationships are constructed between the wind flow parameters and the shape image RGB matrix to generate composite images. Then the composite image and the convolutional neural network (CI–CNN) model is proposed. Due to the aerodynamic coefficients having a strong nonlinearity, the Fourier fitting function is used to describe the nonlinear characteristics of the aerodynamic coefficients changing with the attack angles. Then the Fourier composite image and convolutional neural network (FCI–CNN) model is proposed. The 1443 aerodynamic coefficients of crescent-shaped iced conductors from the wind tunnel test are taken as an example. The results show that both the FCI–CNN model and the CI–CNN model perform well during the estimation of the drag coefficient, especially the R2 calculated as 0.986 and 0.977, with MAE as 0.087 and 0.066, proving that the proposed method is effective. When predicting the lift coefficient, the R2 of the FCI–CNN model and the CI–CNN model are 0.931 and 0.838, respectively. So do the result in the prediction of the torque coefficient. Through the Den Hartog coefficient, the FCI–CNN model reflects the most important range of 170°–185° galloping wind attack angle compared to the wind tunnel test results. The FCI–CNN model gives better predictions than the CI–CNN model.

Multifidelity Prediction Framework with Convolutional Neural Networks Using High-Dimensional Data

This paper proposes two novel multifidelity neural network architectures developed for high-dimensional inputs such as computational flowfields. We employed a two-dimensional flow-varying transonic supercritical airfoil problem while exploring and comparing our methods with the former “multifidelity deep neural networks” from the literature. We call these novel methods “modified multifidelity deep neural networks” and “multifidelity convolutional neural networks.” The main objective of this study is to establish an advanced multifidelity prediction framework that can be applied to any computational data; however, here, we applied our methods to the prediction of aerodynamic coefficients using pressure coefficient fields around the airfoil. To generate the dataset, first, a coarse grid is employed using the SU2 Euler solver for low-fidelity data; then, a relatively finer grid is used to obtain the viscous solutions by using the Spalart–Allmaras turbulence model for the high-fidelity data. The performance metrics to compare the methods are determined as the test accuracy, the physical training time, and the size of the high-fidelity samples. The results demonstrate that the proposed multifidelity neural network architectures outperform the multifidelity deep neural networks in predictive modeling using high-dimensional inputs by improving the multifidelity prediction accuracy significantly.

Deep residual neural network for predicting aerodynamic coefficient changes with ablation

Data-driven methods for predicting aerodynamic coefficients of arbitrary shapes have received considerable attention due to their flexibility and scalability. This paper introduces a deep-learning framework based on a deep residual neural network and K-fold cross-validation for fast and accurate prediction of aerodynamic coefficients of a three-dimensional cone with shape changes due to ablation. The proposed neural network model is trained to learn the underlying relationship between shape transformations and the corresponding changes in aerodynamic coefficients. The shape transformations due to ablation are expressed as the difference between the nominal and ablated cones, measured in units of mesh coordinates. Multiple (K) models constructed based on the training process are combined to reduce the prediction variance effectively. The resulting ensemble model shows an improved prediction performance for various aerodynamic coefficients. To validate our methodology, we compare our model with the generic Multilayer Perceptron (MLP) with a varying number of neurons and the Gaussian process (GP) regression. The test results indicate that the proposed model predicts the aerodynamic coefficients more accurately than the baseline model (MLP/GP), indicating an improved generalization to unseen data.

A Manifold-Based Airfoil Geometric-Feature Extraction and Discrepant Data Fusion Learning Method

The geometrical shape of airfoils and the corresponding flight conditions are crucial factors for aerodynamic coefficient prediction. The obtained geometric-features of airfoils in most existing approaches (e.g., geometrical parameter extraction, polynomial description and deep learning) are in Euclidean space. State-of-the-art studies have shown that the curves or surfaces of an airfoil form a manifold in Riemannian space. Therefore, the features extracted by existing methods are not sufficient to reflect the geometric-features of airfoils. Meanwhile, flight conditions and geometric-features are greatly discrepant with different types. The discrepancy between these two factors must be considered and evaluated to improve the aerodynamic coefficient accuracy. Motivated by the advantages of manifold theory and multi-task learning (MTL), we propose a manifold-based airfoil geometric-feature extraction and discrepant data fusion learning method (MDF) to extract geometric-features of airfoils in Riemannian space (we call them manifold-features) and further fuse the manifold-features with flight conditions to predict aerodynamic coefficients. Experimental results show that our method can extract geometric-features of airfoils more accurately than existing methods, that the average mean square error (MSE) of airfoils rebuilt based on geometric-features is reduced by 41.30%, and that while keeping the same prediction accuracy level for the lift coefficient <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$C_{L}$</tex-math></inline-formula> , the MSE of the drag coefficient <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$C_{D}$</tex-math></inline-formula> predicted by MDF is further reduced by 54.56%.

Prediction and optimization of airfoil aerodynamic performance using deep neural network coupled Bayesian method

In this paper, we proposed an innovative Bayesian optimization (BO) coupled with deep learning for rapid airfoil shape optimization to maximize aerodynamic performance of airfoils. The proposed aerodynamic coefficient prediction model (ACPM) consists of a convolutional path and a fully connected path, which enables the reconstruction of the end-to-end mapping between the Hicks–Henne (H–H) parameterized geometry and the aerodynamic coefficients of an airfoil. The computational fluid dynamics (CFD) model is first validated with the data in the literature, and the numerically simulated lift and drag coefficients were set as the ground truth to guide the model training and validate the network model based ACPM. The average accuracy of lift and drag coefficient predictions are both about 99%, and the determination coefficient R2 are more than 0.9970 and 0.9539, respectively. Coupled with the proposed ACPM, instead of the conventional expensive CFD simulator, the Bayesian method improved the ratio of lift and drag coefficients by more than 43%, where the optimized shape parameters of the airfoil coincide well with the results by the CFD. Furthermore, the whole optimization time is less than 2 min, two orders faster than the traditional BO-CFD framework. The obtained results demonstrate the great potential of the BO-ACPM framework in fast and accurate airfoil shape optimization and design.

CNN-based flow control device modelling on aerodynamic airfoils

Wind energy has become an important source of electricity generation, with the aim of achieving a cleaner and more sustainable energy model. However, wind turbine performance improvement is required to compete with conventional energy resources. To achieve this improvement, flow control devices are implemented on airfoils. Computational fluid dynamics (CFD) simulations are the most popular method for analyzing this kind of devices, but in recent years, with the growth of Artificial Intelligence, predicting flow characteristics using neural networks is becoming increasingly popular. In this work, 158 different CFD simulations of a DU91W(2)250 airfoil are conducted, with two different flow control devices, rotating microtabs and Gurney flaps, added on its Trailing Edge (TE). These flow control devices are implemented by using the cell-set meshing technique. These simulations are used to train and test a Convolutional Neural Network (CNN) for velocity and pressure field prediction and another CNN for aerodynamic coefficient prediction. The results show that the proposed CNN for field prediction is able to accurately predict the main characteristics of the flow around the flow control device, showing very slight errors. Regarding the aerodynamic coefficients, the proposed CNN is also capable to predict them reliably, being able to properly predict both the trend and the values. In comparison with CFD simulations, the use of the CNNs reduces the computational time in four orders of magnitude.

Accuracy of Küchemann’s prediction for the locus of aerodynamic centres on swept wings compared to an inviscid panel method

AbstractThe locus of aerodynamic centres of a finite wing is the collection of all spanwise section aerodynamic centres, and depends on aspect ratio, wing sweep and planform shape. This locus is of great importance in the positioning of vortex elements in lifting-line theory. Traditionally, these vortex elements are placed along the quarter-chord of a wing, leading to inaccurate predictions of aerodynamic coefficients for swept wings due to the discontinuity in the line of vorticity at the wing root. An analytical solution was presented by Küchemann in 1956 to determine the locus of aerodynamic centres as a function of sweep. While experimental studies have been performed to visualise this locus, no large amount of data is available to fully evaluate the accuracy of Küchemann’s analytical solution. In the present study, a numerical approach is taken using a high-order panel method for inviscid, incompressible flow to calculate the locus of aerodynamic centres for elliptic wings over a wide range of sweep angles, aspect ratios and profile thicknesses. An inviscid panel method is chosen over full CFD solutions because of their ability to isolate the inviscid phenomena. Küchemann’s prediction is compared to this numerical data. The root mean square error is calculated for each wing in a broad design space to determine the accuracy of Küchemann’s theory. It is shown to be remarkably accurate over the range of cases studied, with the root mean square error staying below 4% for all wings with aft sweep and aspect ratios higher than $R_A=5$ . The actual difference between Küchemann’s prediction and numerical data is lower than that for the majority of the span for many of the wing designs considered, with the RMS error being skewed by the results at the tip. Results demonstrate that Küchemann’s analytical equations can be used as an accurate approximation for the locus of aerodynamic centres and could be used in modern numerical lifting-line algorithms*.

A twin-decoder structure for incompressible laminar flow reconstruction with uncertainty estimation around 2D obstacles

Over the past few years, deep learning methods have proved to be of great interest for the computational fluid dynamics community, especially when used as surrogate models, either for flow reconstruction, turbulence modeling, or for the prediction of aerodynamic coefficients. Overall exceptional levels of accuracy have been obtained but the robustness and reliability of the proposed approaches remain to be explored, particularly outside the confidence region defined by the training dataset. In this contribution, we present an autoencoder architecture with twin decoder for incompressible laminar flow reconstruction with uncertainty estimation around 2D obstacles. The proposed architecture is trained over a dataset composed of numerically-computed laminar flows around 12,000 random shapes, and naturally enforces a quasi-linear relation between a geometric reconstruction branch and the flow prediction decoder. Based on this feature, two uncertainty estimation processes are proposed, allowing either a binary decision (accept or reject prediction), or proposing a confidence interval along with the flow quantities prediction (u, v, p). Results over dataset samples as well as unseen shapes show a strong positive correlation of this reconstruction score to the mean-squared error of the flow prediction. Such approaches offer the possibility to warn the user of trained models when provided input shows too large deviation from the training data, making the produced surrogate model conservative for fast and reliable flow prediction.

Multi-Objective Optimization of Low Reynolds Number Airfoil Using Convolutional Neural Network and Non-Dominated Sorting Genetic Algorithm

The airfoil is the prime component of flying vehicles. For low-speed flights, low Reynolds number airfoils are used. The characteristic of low Reynolds number airfoils is a laminar separation bubble and an associated drag rise. This paper presents a framework for the design of a low Reynolds number airfoil. The contributions of the proposed research are twofold. First, a convolutional neural network (CNN) is designed for the aerodynamic coefficient prediction of low Reynolds number airfoils. Data generation is discussed in detail and XFOIL is selected to obtain aerodynamic coefficients. The performance of the CNN is evaluated using different learning rate schedulers and adaptive learning rate optimizers. The trained model can predict the aerodynamic coefficients with high accuracy. Second, the trained model is used with a non-dominated sorting genetic algorithm (NSGA-II) for multi-objective optimization of the low Reynolds number airfoil at a specific angle of attack. A similar optimization is performed using NSGA-II directly calling XFOIL, to obtain the aerodynamic coefficients. The Pareto fronts of both optimizations are compared, and it is concluded that the proposed CNN can replicate the actual Pareto in considerably less time.

On the application of artificial neural network for the development of a nonlinear aeroelastic model

Aeroelastic Computational Fluid Dynamics simulations have traditionally been associated to a high computational cost, making them prohibitive in a initial phase of the design. Analytic models, which may not be accurate for nonlinear aerodynamics, have traditionally been utilized in order to size those structures. Recently, some authors have proposed the use of artificial neural networks to reduce the error in the prediction of aerodynamic coefficients of bluff bodies, which have separated flow over a substantial part of its wetted surface. This article proposes a method based on neural networks for calculating the dynamic aerodynamic coefficients of a flat plate. The procedure, which is applied for different network typologies (feed-forward and long-short term memory neural networks), is, then, coupled with a structural solver in order to create an aeroelastic reduced order model. The results are compared with CFD aeroelastic simulations, showing a high reduction of computational cost (99%) without penalties in the accuracy. The instabilities are captured and the mean deformation, amplitude and frequency of the motion are predicted. In addition, the different neural network models are compared evidencing that for the aeroelastic calculation feed-forward networks are most efficient in terms of accuracy and computational cost.

An application of neural networks to the prediction of aerodynamic coefficients of aerofoils and wings

This work proposes a novel multi-output neural network for the prediction of aerodynamic coefficients of aerofoils in two dimensions and wings in three dimensions. Contrary to existing neural networks that are often designed to predict aerodynamic quantities of interest, the proposed network considers as output the pressure at a number of selected points on the aerodynamic shape. The proposed multi-output neural network is compared with other approaches found in the literature. Furthermore, a detailed comparison of the proposed neural network with the popular proper orthogonal decomposition method is presented. The numerical results, involving high dimensional problems with flow and geometric parameters, show the benefits of the proposed approach.

On the application of surrogate regression models for aerodynamic coefficient prediction

Computational fluid dynamics (CFD) simulations are nowadays been intensively used in aeronautical industries to analyse the aerodynamic performance of different aircraft configurations within a design process. These simulations allow to reduce time and cost compared to wind tunnel experiments or flight tests. However, for complex configurations, CFD simulations may still take several hours using high-performance computers to deliver results. For this reason, surrogate models are currently starting to be considered as a substitute of the CFD tool with a reasonable prediction. This paper presents a review on surrogate regression models for aerodynamic coefficient prediction, in particular for the prediction of lift and drag coefficients. To compare the behaviour of the regression models, three different aeronautical configurations have been used, a NACA0012 airfoil, a RAE2822 airfoil and 3D DPW wing. These databases are also freely provided to the scientific community to allow other researchers to make further comparison with other methods.

Data Mining and Machine Learning Techniques for Aerodynamic Databases: Introduction, Methodology and Potential Benefits

Machine learning and data mining techniques are nowadays being used in many business sectors to exploit the data in order to detect trends, discover certain features and patters, or even predict the future. However, in the field of aerodynamics, the application of these techniques is still in the initial stages. This paper focuses on exploring the benefits that machine learning and data mining techniques can offer to aerodynamicists in order to extract knowledge from the CFD data and to make quick predictions of aerodynamic coefficients. For this purpose, three aerodynamic databases (NACA0012 airfoil, RAE2822 airfoil and 3D DPW wing) have been used and results show that machine-learning and data-mining techniques have a huge potential also in this field.