4 results
Now showing1 - 4 of 4
- Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Conditional neural field-based reduced order model for dynamic ditching load prediction(American Institute of Physics (AIP), 2026-10-22); ; ; Grid-based neural networks such as convolutional autoencoders are widely used in dimension reduction-based surrogate models for computational fluid dynamics. Recently, the use of coordinate-based approaches like conditional neural fields (CNFs) has emerged. Their independence of the spatial discretization is an advantageous feature and a key motivation for computational fluid dynamic applications. This paper discusses the spatiotemporal prediction of aircraft ditching loads using a CNF approach. The model employs a Fourier feature mapping of input coordinates to capture high frequency content. Based on feature-wise linear modulation, the output is conditioned on a latent vector, which is learned by autodecoding. The model is evaluated using two datasets for the dynamic loads on the fuselage of a German Aerospace Center (DLR) D150 aircraft. The first dataset relates to a single fixed spatial discretization, and the second dataset includes data from different discretizations. On the first dataset, the model is compared to a grid-dependent convolutional autoencoder-based model, and both approaches achieve similar reconstruction accuracy, but the CNF requires significantly fewer parameters. When paired with a long short-term memory network in the latent space, both approaches achieve similar spatiotemporal prediction accuracy. Results for the second dataset demonstrate the ability of the CNF-based approach to reconstruct ditching loads accurately for heterogeneous spatial discretizations. In particular, the model can generalize to discretizations not used for training. This allows for flexible use of training datasets generated for different geometries and/or discretizations and the use of the surrogate model to predict loads for different configurations.Publicationtype: Journal Article - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Disentangled latent spaces for reduced order models using deterministic autoencodersData-driven reduced-order models based on autoencoders generally lack interpretability compared to classical methods such as the proper orthogonal decomposition. More interpretability can be gained by disentangling the latent variables and analyzing the resulting modes. For this purpose, probabilistic β-variational autoencoders (β-VAEs) are frequently used in computational fluid dynamics and other simulation sciences. Using a benchmark periodic flow dataset, we show that competitive results can be achieved using non-probabilistic autoencoder approaches that either promote orthogonality or penalize correlation between latent variables. Compared to probabilistic autoencoders, these approaches offer more robustness with respect to the choice of hyperparameters entering the loss function. We further demonstrate the ability of a non-probabilistic approach to identify a reduced number of active latent variables by introducing a correlation penalty, a function also known from the use of β-VAE. The investigated probabilistic and non-probabilistic autoencoder models are finally used for the dimensionality reduction of aircraft ditching loads, which serves as an industrial application in this work.Publicationtype: Journal Article93 44 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: without files , Machine Learning Based Prediction of Ditching Loads(American Institute of Aeronautics and Astronautics (AIAA), 2025); ; ; Approaches are presented to predict dynamic ditching loads on aircraft fuselages using machine learning. The employed learning procedure is structured into two parts, the reconstruction of the spatial loads using a convolutional autoencoder (CAE) and the transient evolution of these loads in a subsequent part. Different CAE strategies are assessed and combined with either long short-term memory (LSTM) networks or Koopman operator based methods to predict the transient behavior. The training data are compiled by an extension of the momentum method of von Karman and Wagner, and the rationale of the training approach is briefly summarized. The application included refers to a full-scale fuselage of a DLR-D150 aircraft for a range of horizontal and vertical approach velocities at 6 deg incidence. Results indicate a satisfactory level of predictive agreement for all four investigated surrogate models examined, with the combination of an LSTM and a deep decoder CAE showing the best performance.Publicationtype: Journal Article80 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Machine learning based prediction of ditching loads(2024-10-11); ; ; We present approaches to predict dynamic ditching loads on aircraft fuselages using machine learning. The employed learning procedure is structured into two parts, the reconstruction of the spatial loads using a convolutional autoencoder (CAE) and the transient evolution of these loads in a subsequent part. Different CAE strategies are assessed and combined with either long short-term memory (LSTM) networks or Koopman-operator based methods to predict the transient behaviour. The training data is compiled by an extension of the momentum method of von-Karman and Wagner and the rationale of the training approach is briefly summarised. The application included refers to a full-scale fuselage of a DLR-D150 aircraft for a range of horizontal and vertical approach velocities at 6° incidence. Results indicate a satisfactory level of predictive agreement for all four investigated surrogate models examined, with the combination of an LSTM and a deep decoder CAE showing the best performance.Publicationtype: Preprint75 151