DrivAerNet is a large-scale, high-fidelity CFD dataset of 3D industry-standard car shapes designed for data-driven aerodynamic design. It comprises 4000 high-quality 3D car meshes and their corresponding aerodynamic performance coefficients, alongside full 3D flow field information. It includes: - CFD Simulation Data: The raw dataset, including full 3D pressure, velocity fields, and wall-shear stresses, computed using 8-16 million mesh elements has a total size of $\sim$ 16TB. - Curated CFD Simulations: For ease of access and use, a streamlined version of the CFD simulation data is provided, refined to include key insights and data, reducing the size to $\sim$ 1TB. - 3D Car Meshes: A total of 4000 designs, showcasing a variety of conventional car shapes and emphasizing the impact of minor geometric modifications on aerodynamic efficiency. The 3D meshes and aerodynamic coefficients $\sim$ 84GB. - 2D slices include the car's wake in the $x$-direction and the symmetry plane in the $y$-direction $\sim$ 12GB.
The HEterogeneous Materials and Elastic Waves with Source variability in 3D (HEMEWS-3D) database comprises 30,000 simulations of elastic wave propagation in 3D geological domains. Each domain is parametrized by a different geological model built from a random arrangement of layers augmented by random fields that represent heterogeneities. Elastic waves originate from a randomly located pointwise source parametrized by a random moment tensor. For each simulation, ground motion is synthesized at the surface by a grid of virtual sensors. The high frequency of waveforms ($f_{max}$ = 5 Hz) allows for extensive analyses of surface ground motion. Existing and foreseen applications range from statistical analyses of the ground motion variability and machine learning methods on geological models to deep-learning-based predictions of ground motion that depend on 3D heterogeneous geologies and source properties.
We present a structured benchmark dataset for a representative vibroacoustic problem: Predicting the frequency response for vibrating plates. The vibrating plates benchmark dataset consists of in total 12,000 varied plate designs and accompanying vibration patterns, when the plates are excited by a harmonic force. These vibration platterns give the vibration velocity at every location of the plate orthogonal to its surface. The plate designs incorporate randomly placed beadings, indentations in the plate surface. The beadings stiffen the plates and completely change the resulting vibration patterns. Additionally, the size, thickness and damping loss factor of the plates are varied. The dataset is intended to further the development of surrogate modeling methods for partial differential equations in the field of vibroacoustics. It contains two settings, G-5000 and V-5000. Both incorporate the beading pattern variation but only G-5000 additionally includes the variation of plate size, thickness and damping loss factor.
Given the lack of consensus on a standard setof benchmarks for machine learning of PDEs, we propose a new suite of benchmarks here. Our aims in this regard are to ensure i) sufficient diversity among the types of PDE considered ii) access to training and test data is readily available for rapid prototyping and reproducibility iii) intrinsic computational complexity of problem to make sure that it is worthwhile to design fast surrogates to classical PDE solvers for a particular problem
Data Set Structure Fluid Structure Interaction(NS +Elastic wave) The TF_fsi2_results folder contains simulation data organized by various parameters (mu, x1, x2) where mu determines the viscosity and x1 and x2 are the parameters of the inlet condition. The dataset includes files for mesh, displacement, velocity, and pressure. This dataset structure is detailed below: plaintext TF_fsi2_results/ ├── mesh.h5 # Initial mesh ├── mu=1.0/ # Simulation results for mu = 1.0 │ ├── x1=-4/ # Inlet parameter x1 = -4 │ │ ├── x2=-4/ # Inlet parameter for x2 = -4 │ │ │ └── visualization/ │ │ │ ├── displacement.h5 # Displacements for mu=1.0, x1=-4, x2=-4 │ │ │ ├── velocity.h5 # Velocity field for mu=1.0, x1=-4, x2=-4 │ │ │ └── pressure.h5 # Pressure field for mu=1.0, x1=-4, x2=-4 │ │ ├── x2=-2/ │ │ │ └── visualization/ │ │ │ ├── displacement.h5 │ │ │ ├── velocity.h5 │ │ │ └── pressure.h5 │ │ └── ... # Other x2 values for x1 = -4 │ ├── x1=-2/ │ │ ├── x2=-4/ │ │ │ └── visualization/ │ │ │ ├── displacement.h5 │ │ │ ├── velocity.h5 │ │ │ └── pressure.h5 │ │ └── ... # Other x2 values for x1 = -2 │ └── ... # Other x1 values for mu = 1.0 ├── mu=5.0/ # Simulation results for mu = 5.0 │ └── ... # Similar structure as mu=1.0 └── mu=10.0/ # Simulation results for mu = 10.0 └── ... # Similar structure as mu=1.0 The dataset has readData.py and readMesh.py for loading the data. Also, the NsElasticDataset class in data_utils/data_loaders.py loads data automatically for all specified mus and inlet conditions (x1 and x2). Fluid Motions with Non-deformable Solid(NS) The data is in the folder TF_cfd2_results, and the organization is the same as above.