Aerodynamic Design Optimization and Shape Exploration using Generative Adversarial Networks

Chen, Wei; Chiu, Kevin; Fuge, Mark

doi:10.2514/6.2019-2351

Cited by 71 publications

(44 citation statements)

References 56 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finally, in both methods, we use the soft vicinal loss (it was observed that the soft vicinal loss performed better in both methods across all examples) and pick and based on the rule of thumb method described by [11]. In the Airfoil example we used a residual neural network (ResNet) [16] trained on the dataset as the estimator model and a BézierGAN [8] to generate airfoils. In this example we refer to the continuously conditioned Bézier-GAN as 'CcGAN' and the BézierGAN with L as 'PcDGAN'.…”

Section: Model Configurationmentioning

confidence: 99%

“…This assumption is impractical in many engineering design settings since performance is usually continuous. For example, in turbine design, some important performance metrics are the power coefficient, the pressure coefficient, and the cavitation number [23]; in aerodynamic design, performance is measured by lift to drag ratio or inverse lift coefficient [8]; in beam design, common metrics are compliance and natural-frequency [2] -these metrics are all continuous variables. To use those metrics as the condition in conditional generative models, past work [1,25] proposes to discretize the continuous values of the metrics to discrete bins.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

PcDGAN

Nobari

Chen

Ahmed

2021

Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery &Amp; Data Mining

Self Cite

View full text Add to dashboard Cite

Engineering design tasks often require synthesizing new designs that meet desired performance requirements. The conventional design process, which requires iterative optimization and performance evaluation, is slow and dependent on initial designs. Past work has used conditional generative adversarial networks (cGANs) to enable direct design synthesis for given target performances. However, most existing cGANs are restricted to categorical conditions. Recent work on Continuous conditional GAN (CcGAN) tries to address this problem, but still faces two challenges: 1) it performs poorly on non-uniform performance distributions, and 2) the generated designs may not cover the entire design space. We propose a new model, named Performance Conditioned Diverse Generative Adversarial Network (PcDGAN), which introduces a singular vicinal loss combined with a Determinantal Point Processes (DPP) based loss function to enhance diversity. PcDGAN uses a new self-reinforcing score called the Lambert Log Exponential Transition Score (LLETS) for improved conditioning. Experiments on synthetic problems and a real-world airfoil design problem demonstrate that PcDGAN outperforms state-of-the-art GAN models and improves the conditioning likelihood by 69% in an airfoil generation task and upto 78% in synthetic conditional generation tasks and achieves greater design space coverage. The proposed method enables efficient design synthesis and design space exploration with applications ranging from CAD model generation to metamaterial selection. CCS CONCEPTS• Applied computing → Aerospace; • Computing methodologies → Machine learning algorithms; Machine learning approaches.

show abstract

Section: Model Configurationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

PcDGAN

Nobari

Chen

Ahmed

2021

Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery &Amp; Data Mining

Self Cite

View full text Add to dashboard Cite

show abstract

“…The results suggest a great potential for the nonlinear MD-CNN-AE to be used for feature extraction of flow fields in lower dimension. Chen et al [136] applied CNN-based GANs to reduce the dimension of input airfoil shapes to realize the design optimization of an airfoil. CNNs are better than traditional approaches (e.g.…”

Section: B Feature Extractionmentioning

confidence: 99%

Neural Networks-Based Aerodynamic Data Modeling: A Comprehensive Review

Zhang

Xiang

et al. 2020

IEEE Access

View full text Add to dashboard Cite

This paper reviews studies on neural networks in aerodynamic data modeling. In this paper, we analyze the shortcomings of computational fluid dynamics (CFD) and traditional reduced-order models (ROMs). Subsequently, the history and fundamental methodologies of neural networks are introduced. Furthermore, we classify the neural networks based studies in aerodynamic data modeling and illustrate comparisons among them. These studies demonstrate that neural networks are effective approaches to aerodynamic data modeling. Finally, we identify three important trends for future studies in aerodynamic data modeling: a) the transformation method and physics informed models will be combined to solve high-dimensional partial differential equations; b) in the research area of steady aerodynamic response predictions, model-oriented studies and data-integration-oriented studies will become the future research directions, while in unsteady aerodynamic response predictions, radial basis function neural networks (RBFNNs) are the best tools for capturing the nonlinear characteristics of flow data, and convolutional neural networks (CNNs) are expected to replace long short-term memories (LSTMs) to capture the temporal characteristics of flow data; and c) in the field of steady or unsteady flow field reconstructions, the CNN-based conditional generative adversarial networks (cGANs) will be the best frameworks in which to discover the spatiotemporal distribution of flow field data. INDEX TERMS Aerodynamics, convolutional neural networks, neural networks, generative adversarial networks, recurrent neural networks.

show abstract

“…Deep learning [39] refers to neural networks with a large number of hidden layers. Deep learning has gained significant popularity in recent years as advances in computing power, such as massively parallel GPU architectures, and custom software, such as TENSOR-FLOW [40], have enabled the training of deep neural networks to process large datasets with impressive results in the machine learning [38,41] and design communities [37,42]. However, training deep learning networks requires access to very large amounts of training data to fit thousands of associated network parameters.…”

Section: Neuralmentioning

confidence: 99%

A Comparative Evaluation of Supervised Machine Learning Classification Techniques for Engineering Design Applications

Sharpe

Wiest

Wang

et al. 2019

Journal of Mechanical Design

View full text Add to dashboard Cite

Supervised machine learning techniques have proven to be effective tools for engineering design exploration and optimization applications, in which they are especially useful for mapping promising or feasible regions of the design space. The design space mappings can be used to inform early-stage design exploration, provide reliability assessments, and aid convergence in multiobjective or multilevel problems that require collaborative design teams. However, the accuracy of the mappings can vary based on problem factors such as the number of design variables, presence of discrete variables, multimodality of the underlying response function, and amount of training data available. Additionally, there are several useful machine learning algorithms available, and each has its own set of algorithmic hyperparameters that significantly affect accuracy and computational expense. This work elucidates the use of machine learning for engineering design exploration and optimization problems by investigating the performance of popular classification algorithms on a variety of example engineering optimization problems. The results are synthesized into a set of observations to provide engineers with intuition for applying these techniques to their own problems in the future, as well as recommendations based on problem type to aid engineers in algorithm selection and utilization.

show abstract

Aerodynamic Design Optimization and Shape Exploration using Generative Adversarial Networks

Cited by 71 publications

References 56 publications

PcDGAN

PcDGAN

Neural Networks-Based Aerodynamic Data Modeling: A Comprehensive Review

A Comparative Evaluation of Supervised Machine Learning Classification Techniques for Engineering Design Applications

Contact Info

Product

Resources

About