Aerodynamic Shape Design by Evolutionary Optimization and Support Vector Machines

Andrés-Pérez, Esther; Carro-Calvo, L.; Salcedo‐Sanz, Sancho; Martin-Burgos, Mario J.

doi:10.1007/978-3-319-21506-8_1

Cited by 8 publications

(4 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We see a range of SVM applications in support of aerodynamic analysis and optimization. For example, Andrés-Pérez et al [135] used SVM as a surrogate model to estimate objective function (lift-drag ratio), in combination with an evolutionary algorithm for ASO problems. SVM surrogate managed to address 14 input parameters for a 2D case and 36 inputs for a 3D case.…”

Section: Support Vector Machinesmentioning

confidence: 99%

Machine Learning in Aerodynamic Shape Optimization

Li¹,

Martins²

2022

Preprint

View full text Add to dashboard Cite

Large volumes of experimental and simulation aerodynamic data have been rapidly advancing aerodynamic shape optimization (ASO) via machine learning (ML), whose effectiveness has been growing thanks to continued developments in deep learning. In this review, we first introduce the state of the art and the unsolved challenges in ASO. Next, we present a description of ML fundamentals and detail the ML algorithms that have succeeded in ASO. Then we review ML applications contributing to ASO from three fundamental perspectives: compact geometric design space, fast aerodynamic analysis, and efficient optimization architecture. In addition to providing a comprehensive summary of the research, we comment on the practicality and effectiveness of the developed methods. We show how cutting-edge ML approaches can benefit ASO and address challenging demands like interactive design optimization. However, practical large-scale design optimizations remain a challenge due to the costly ML training expense. A deep coupling of ML model construction with ASO prior experience and knowledge, such as taking physics into account, is recommended to train ML models effectively.

show abstract

Section: Support Vector Machinesmentioning

confidence: 99%

Machine Learning in Aerodynamic Shape Optimization

Li¹,

Martins²

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…However, this UAV database is limited and as there is no quantitative value or performance calculation for comparison, it is inadequate like other attempts in the literature. Another major gap in the literature is that almost all of the studies are limited to certain parts of the aircraft, such as the 2D airfoil or the 3D wing [17][18][19][20][21][22][23][24][25][26][27][28][29]. Approximate models for aerodynamic coefficients of complete aircraft geometry used in aerodynamic shape optimization are very limited.…”

Section: Introductionmentioning

confidence: 99%

AI-Driven Multidisciplinary Conceptual Design of Unmanned Aerial Vehicles

Karali,

Inalhan,

Tsourdos

2024

AIAA SCITECH 2024 Forum

View full text Add to dashboard Cite

This paper presents a multidisciplinary conceptual design framework for unmanned aerial vehicles based on artificial intelligence-driven analysis models. This approach leverages AIdriven analysis models that include aerodynamics, structural mass, and radar cross-section predictions to bring quantitative data to the initial design stage, enabling the selection of the most appropriate configuration from various optimized concept designs. Due to the design optimization cycle, the initial dimensions of key components such as the wing, tail, and fuselage are provided more accurately for later design activities. Simultaneously, the generated structure enables more suitable design point selection through the feedback loop within the design iteration. Therefore, in addition to reducing design costs, this approach also offers a substantial time advantage in the overall design process.

show abstract

“…These methods can be broadly categorized into two groups: traditional surrogate models and neural network models. Traditional surrogate models, such as polynomial response surface [2], kriging [3], and support vector regression (SVR) [4], typically involve fitting a prediction function between the airfoil shape and the aerodynamic characteristics. However, traditional surrogate models are unable to handle a large volume of training data [5].…”

Section: Introductionmentioning

confidence: 99%

Fast Prediction of Airfoil Aerodynamic Characteristics Based on a Combined Autoencoder

Wang,

Qian,

Zhao

et al. 2024

Symmetry

View full text Add to dashboard Cite

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.

show abstract

Aerodynamic Shape Design by Evolutionary Optimization and Support Vector Machines

Cited by 8 publications

References 28 publications

Machine Learning in Aerodynamic Shape Optimization

Machine Learning in Aerodynamic Shape Optimization

AI-Driven Multidisciplinary Conceptual Design of Unmanned Aerial Vehicles

Fast Prediction of Airfoil Aerodynamic Characteristics Based on a Combined Autoencoder

Contact Info

Product

Resources

About