Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userA Point-Cloud Deep Learning Framework for Prediction of Fluid Flow Fields on Irregular Geometries
50
0
0.0
(
0
)
Ask ChatGPT about the research
No Arabic abstract
We present a novel deep learning framework for flow field predictions in irregular domains when the solution is a function of the geometry of either the domain or objects inside the domain. Grid vertices in a computational fluid dynamics (CFD) domain are viewed as point clouds and used as inputs to a neural network based on the PointNet architecture, which learns an end-to-end mapping between spatial positions and CFD quantities. Using our approach, (i) the network inherits desirable features of unstructured meshes (e.g., fine and coarse point spacing near the object surface and in the far field, respectively), which minimizes network training cost; (ii) object geometry is accurately represented through vertices located on object boundaries, which maintains boundary smoothness and allows the network to detect small changes between geometries; and (iii) no data interpolation is utilized for creating training data; thus accuracy of the CFD data is preserved. None of these features are achievable by extant methods based on projecting scattered CFD data into Cartesian grids and then using regular convolutional neural networks. Incompressible laminar steady flow past a cylinder with various shapes for its cross section is considered. The mass and momentum of predicted fields are conserved. For the first time, our network generalizes the predictions to multiple objects as well as an airfoil, even though only single objects and no airfoils are observed during training. The network predicts the flow fields hundreds of times faster than our conventional CFD solver, while maintaining excellent to reasonable accuracy.
rate research
Read More
A new kind of Lagrangian diagnostic family is proposed and a specific form of it is suggested for characterizing mixing: the maximal extent of a trajectory (MET). It enables the detection of coherent structures and their dynamics in two- (and potentially three-) dimensional unsteady flows in both bounded and open domains. Its computation is much easier than all other Lagrangian diagnostics known to us and provides new insights regarding the mixing properties on both short and long time scales and on both spatial plots and distribution diagrams. We demonstrate its applicability to two dimensional flows using two toy models and a data set of surface currents from the Mediterranean Sea.
The most essential characteristic of any fluid is the velocity field v(r) and this is particularly true for macroscopic quantum fluids. Although rapid advances have occurred in quantum fluid v(r) imaging, the velocity field of a charged superfluid - a superconductor - has never been visualized. Here we use superconductive-tip scanning tunneling microscopy to image the electron-pair density r{ho}_S(r) and velocity v_S(r) fields of the flowing electron-pair fluid in superconducting NbSe2. Imaging v_S(r) surrounding a quantized vortex finds speeds reaching 10,000 km/hr. Together with independent imaging of r{ho}_S(r) via Josephson tunneling, we visualize the supercurrent density j_S(r)=r{ho}_S(r)v_S(r), which peaks above 3 x 10^7 A/cm^2. The spatial patterns in electronic fluid flow and magneto-hydrodynamics reveal hexagonal structures co-aligned to the crystal lattice and quasiparticle bound states, as long anticipated. These novel techniques pave the way for electronic fluid flow visualization in many other quantum fluids.
There is great interest to develop artificial intelligence-based protein-ligand affinity models due to their immense applications in drug discovery. In this paper, PointNet and PointTransformer, two pointwise multi-layer perceptrons have been applied for protein-ligand affinity prediction for the first time. Three-dimensional point clouds could be rapidly generated from the data sets in PDBbind-2016, which contain 3 772 and 11 327 individual point clouds derived from the refined or/and general sets, respectively. These point clouds were used to train PointNet or PointTransformer, resulting in protein-ligand affinity prediction models with Pearson correlation coefficients R = 0.831 or 0.859 from the larger point clouds respectively, based on the CASF-2016 benchmark test. The analysis of the parameters suggests that the two deep learning models were capable to learn many interactions between proteins and their ligands, and these key atoms for the interaction could be visualized in point clouds. The protein-ligand interaction features learned by PointTransformer could be further adapted for the XGBoost-based machine learning algorithm, resulting in prediction models with an average Rp of 0.831, which is on par with the state-of-the-art machine learning models based on PDBbind database. These results suggest that point clouds derived from the PDBbind datasets are useful to evaluate the performance of 3D point clouds-centered deep learning algorithms, which could learn critical protein-ligand interactions from natural evolution or medicinal chemistry and have wide applications in studying protein-ligand interactions.
We investigate the flow of a nano-scale incompressible ridge of low-volatility liquid along a chemical channel: a long, straight, and completely wetting stripe embedded in a planar substrate, and sandwiched between two extended less wetting solid regions. Molecular dynamics simulations, a simple long-wavelength approximation, and a full stability analysis based on the Stokes equations are used, and give qualitatively consistent results. While thin liquid ridges are stable both statically and during flow, a (linear) pearling instability develops if the thickness of the ridge exceeds half of the width of the channel. In the flowing case periodic bulges propagate along the channel and subsequently merge due to nonlinear effects. However, the ridge does not break up even when the flow is unstable, and the qualitative behavior is unchanged even when the fluid can spill over onto a partially wetting exterior solid region.
Surface effects become important in microfluidic setups because the surface to volume ratio becomes large. In such setups the surface roughness is not any longer small compared to the length scale of the system and the wetting properties of the wall have an important influence on the flow. However, the knowledge about the interplay of surface roughness and hydrophobic fluid-surface interaction is still very limited because these properties cannot be decoupled easily in experiments. We investigate the problem by means of lattice Boltzmann (LB) simulations of rough microchannels with a tunable fluid-wall interaction. We introduce an ``effective no-slip plane at an intermediate position between peaks and valleys of the surface and observe how the position of the wall may change due to surface roughness and hydrophobic interactions. We find that the position of the effective wall, in the case of a Gaussian distributed roughness depends linearly on the width of the distribution. Further we are able to show that roughness creates a non-linear effect on the slip length for hydrophobic boundaries.
suggested questions
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
