Jun 2026· Journal of Physics, Conference Series· Vol 3272, pp. 012014· 0 citations· 24 references
Physics
TL;DR
A hybrid modelling framework (PMG) that integrates Proper Orthogonal Decomposition, Multilayer Perceptron, Multilayer Perceptron, and Gaussian Process Regression is proposed that demonstrates superior predictive accuracy compared to traditional reduced-order models and deep learning-based approaches.
Abstract
Accurately modelling the nonlinear drag map within the design parameter space is a key challenge in preliminary nacelle design. However, due to sharp gradient variations in the drag distribution and the limited size of the available dataset, traditional surrogate models often suffer from insufficient predictive accuracy and poor generalization. To address this challenge, this study systematically evaluates the modelling performance of representative reduced-order models and deep learning–based approaches, and proposes a hybrid modelling framework (PMG) that integrates Proper Orthogonal Decomposition (POD), Multilayer Perceptron (MLP), and Gaussian Process Regression (GPR). The performance of various methods is evaluated and validated using high-resolution numerical simulations across the nacelle design parameter space. The results show that the PMG model reduces the required number of samples by 80% while accurately capturing the characteristics of complex drag distributions. Under small-sample conditions, the PMG model demonstrates superior predictive accuracy compared to traditional reduced-order models and deep learning-based approaches. This framework provides a promising approach for the rapid evaluation of preliminary nacelle designs.
Accurate machine-learning models for aerodynamic prediction are essential for accelerating shape optimization yet remain challenging to develop for complex three-dimensional configurations due to the high cost of generating training data. This work introduces a methodology for efficiently constructing accurate surrogate models for design purposes by first pretraining a large-scale model on diverse geometries and then fine-tuning it with a few more detailed task-specific samples. A Transformer-based architecture, AeroTransformer, is developed and tailored for large-scale training to learn aerodynamics. The methodology is evaluated on transonic wings, where the model is pretrained on SuperWing, a dataset of nearly 30,000 samples with broad geometric diversity, and subsequently fine-tuned to handle specific wing shapes perturbed from the Common Research Model. Results show that, with 450 task-specific samples, the proposed methodology achieves a 0.36% error on surface-flow prediction, reducing error by 84.2% compared to training from scratch. The influence of model configurations and training strategies is also systematically studied to provide guidance on effectively training and deploying such models under limited data and computational budgets. To facilitate reuse, we release the datasets and the pretrained models at https://github.com/tum-pbs/AeroTransformer . An interactive design tool is also built on the pretrained model and is available online at https://webwing.pbs.cit.tum.de .
In the present work, an alternative methodology was developed for the rapid prediction of pressure distributions over wings of low-speed aircraft. A hybrid neural network architecture, named “MARTHA” (Model for Airloads Reconstruction using a Trained Hybrid Architecture), was presented, which is composed of a Multilayer Perceptron and the decoder of an Autoencoder. Three compact representation models—Principal Component Analysis (PCA), Autoencoder (AE), and Variational Autoencoder (VAE)—were systematically evaluated to determine the optimal dimensionality reduction architecture; the AE was selected based on its superior reconstruction accuracy and training stability. The main feature of MARTHA is that it provides predictions of the differential pressure coefficient field in the form of monochrome images, where the pixel intensity directly represents the normalized pressure value. One of the main objectives of developing MARTHA was to create a rapid surrogate model that can approximate vortex lattice method (VLM) simulations in preliminary design and optimization tasks, particularly when thousands of wing configurations need to be evaluated. The key feature of the proposed model is its ability to predict the pressure distribution for trapezoidal wings of various geometries 101–104 times faster than numerical models, while maintaining accuracy (R2 = 0.9998). The data obtained are presented in a convenient format for their further use in CAE systems of strength analysis. To assess the practical utility of the proposed model, implementation cases were carried out using the finite element software ANSYS 18.2 for three wing configurations not present in the training dataset. The pressure fields predicted by MARTHA were mapped onto the wing meshes, and linear static structural analyses were performed. The obtained Von Mises stress distributions showed good agreement with the corresponding distributions obtained using numerical models.
O. Lukyanov, Damian Josue Guerra Guerra, J. G. Quijada Pioquinto et al.· Technologies· 0 citations
Purpose. To develop a hybrid methodology that integrates Artificial Neural Networks (ANN) with Finite Element Method (FEM) simulations for the rapid and accurate prediction of slope stability.
Methodology. A dataset of 1,000 FEM simulations was generated by systematically varying seven key input parameters: slope geometry (height and angle) and soil properties (cohesion, friction angle, unit weight, pore water pressure ratio, and reinforcement type). An ANN model with a (7-10-1) feedforward architecture was trained on this data.
Findings. The model demonstrated exceptional predictive performance, achieving a near-perfect correlation coefficient (R 0.999997) and an extremely low mean squared error (MSE = 3.6828 10-6), showing close agreement with the FEM-computed factors of safety (FOS). A comprehensive sensitivity analysis based on analysis of variance identified the pore water pressure ratio as the dominant controlling parameter, contributing approximately 77 % to the variability of FOS, followed by cohesion with a contribution of about 13 %. Complementary correlation analysis revealed that cohesion exhibits the strongest linear correlation with FOS (r = 0.83), whereas the pore water pressure ratio shows a relatively weak linear correlation, highlighting its pronounced nonlinear and interaction-driven influence on slope stability. These results demonstrate that the proposed ANN–FEM hybrid framework provides a powerful, efficient, and reliable tool for slope stability assessment and parametric analysis. The methodology is particularly well suited for engineering applications requiring rapid decision-making, large-scale evaluations, and uncertainty analysis.
Originality. The core originality of this research is its development of a robust ANN–FEM hybrid framework applied to a large, systematically generated dataset of 1,000 slope simulations. Unlike many studies, it comprehensively incorporates seven critical input variables, including the often underrepresented pore water pressure. Furthermore, its scientific rigor is enhanced by a dual interpretability strategy that combines analysis of variance for quantifying parameter contribution and correlation heatmaps for distinguishing linear effects from nonlinear ones, providing deeper insight into slope stability mechanisms.
Practical value. This study provides engineers with a fast and reliable tool to predict slope safety in seconds instead of running time-consuming FEM simulations, making it highly valuable for real-time decision-making and large parametric studies. Practically, it also shows that controlling pore water pressure (through drainage) is the most effective risk-reduction strategy, while cohesion offers a predictable way to improve slope stability in design.
F. Benayoun, M. Feligha, S. Bekkouche et al.· Naukovyi Visnyk Natsionalnoh...· 0 citations
To address the problems in hull form optimization where an increase in design variables requires approximation models to be rebuilt from scratch and historical CFD samples are insufficiently utilized, this paper proposes an incremental approximation model construction method. Based on the additive decomposition property of high-dimensional model representation, this method breaks through the rigid structure limitations of traditional approximation models. When dimension expansion occurs in the design space, it fully inherits existing low-order component models and historical sample point databases, requiring only local supplementary sampling and incremental construction for new variables and their strong coupling terms, thereby achieving adaptive cross-dimensional updates of the approximation model. The method is validated through numerical test functions and a container ship hull line resistance optimization case study. Results show that, compared with traditional full-dimensional approximation models, this method reduces full-space resampling overhead and maintains high prediction accuracy while reducing CFD sample requirements, providing an efficient modeling approach for ship hydrodynamic optimization in high-dimensional dynamic spaces.
In mechanical design optimization, ANN-based surrogate models are increasingly used to replace computationally expensive simulations such as the Finite Element Method (FEM) and Computational Fluid Dynamics (CFD). However, their performance strongly depends on the quality of training data (Design of Experiments - DoE) and ANN configuration. Many studies still rely on random sampling and trial-and-error approaches, leading to suboptimal accuracy. This paper develops an ANN-based surrogate model for the structural design of a large-scale 3D concrete printer frame. A dataset is generated using an automated computational module integrating MATLAB with ANSYS APDL. The ANN models are used to evaluate the influence of design parameters on structural displacements and natural frequencies. Comparative results show that the ANN model based on random sampling (RS) and trial-and-error yields a maximum relative error of approximately 8%. The use of Latin Hypercube Sampling (LHS) reduces this error to below 3.5%, while the combination of LHS and Reduced Grid Search (RGS) achieves the best performance, with the maximum relative error predominantly below 2%. These results demonstrate that combining LHS with hyperparameter optimization significantly improves the accuracy and robustness of surrogate models, providing a reliable alternative to FEM in structural design optimization.
D. Ta, V. B. Phung, H. Dang et al.· 2026 11th International Conf...· 0 citations
Neural surrogates offer a promising route to accelerating computationally expensive simulations governed by partial differential equations across science and industry. Their practical deployment, however, is limited by unreliable predictions under out-of-distribution (OOD) conditions. We develop a solver-coupled surrogate-Newton framework that uses surrogate predictions as high-quality initial guesses for Newton-Krylov iterations, thereby combining rapid global flow-field prediction with high-accuracy numerical convergence at the terminal stage. On an OOD benchmark comprising geometries sampled from actual transonic airfoil optimization trajectories, the framework lowers the median residual L_2 ratio by over seven orders of magnitude while substantially reducing field and aerodynamic errors. In practical supercritical airfoil optimization, it improves online prediction reliability while achieving a 15.5-fold generation-level speedup over CFD. We further test the framework's extension to three dimensions using a flying-wing dataset. Together, these studies demonstrate the potential of surrogate-Newton coupling to deliver accurate, efficient and scalable steady CFD across industrial workflows.
Ming Lei, Weishao Tang, Yufei Zhang et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
