Filters








4,887 Hits in 4.6 sec

Latent-space Physics: Towards Learning the Temporal Evolution of Fluid Flow [article]

Steffen Wiewel, Moritz Becher, Nils Thuerey
2019 arXiv   pre-print
We propose a method for the data-driven inference of temporal evolutions of physical functions with deep learning. More specifically, we target fluid flows, i.e.  ...  We demonstrate for the first time that dense 3D+time functions of physics system can be predicted within the latent spaces of neural networks, and we arrive at a neural-network based simulation algorithm  ...  Supplemental Document for Latent Space Physics: Towards Learning the Temporal Evolution of Fluid Flow Appendix A: Long-short Term Memory Units and Dimensionality A central challenge for deep learning problems  ... 
arXiv:1802.10123v3 fatcat:m6jbm5xzrbci3ehkb3vxm7j2rq

Latent Space Simulation for Carbon Capture Design Optimization [article]

Brian Bartoldson, Rui Wang, Yucheng Fu, David Widemann, Sam Nguyen, Jie Bao, Zhijie Xu, Brenda Ng
2021 arXiv   pre-print
This raises the possibility of a fast, accurate replacement for a CFD simulator and therefore efficient approximation of the IAs required by CCS design optimization.  ...  While the IA associated with a particular CCS design can be estimated via a computational fluid dynamics (CFD) simulation, using CFD to derive the IAs associated with numerous CCS designs is prohibitively  ...  Latent space physics: Towards learning the temporal evolution of fluid flow. In Computer Graphics Forum, volume 38, pages 71–82, 2019. [29] Jared Willard, Xiaowei Jia, Shaoming Xu, Michael S.  ... 
arXiv:2112.11656v1 fatcat:tvecpx3ivvexnpq7yvmf6bo4f4

Latent Space Subdivision: Stable and Controllable Time Predictions for Fluid Flow [article]

Steffen Wiewel, Byungsoo Kim, Vinicius C. Azevedo, Barbara Solenthaler, Nils Thuerey
2020 arXiv   pre-print
We propose an end-to-end trained neural networkarchitecture to robustly predict the complex dynamics of fluid flows with high temporal stability.  ...  By selectively overwriting parts of the predicted latent space points, our proposed method is capable to robustly predict long-term sequences of complex physics problems.  ...  Source code and video: https://ge.in.tum.de/publications/ latent-space-subdivision/.  ... 
arXiv:2003.08723v1 fatcat:ovdf7njbhfh5lpeswa6olgubxm

Physics perception in sloshing scenes with guaranteed thermodynamic consistency [article]

Beatriz Moya, Alberto Badias, David Gonzalez, Francisco Chinesta, Elias Cueto
2021 arXiv   pre-print
RNNs learn from history the required hidden information to correlate the limited information with the latent space where the simulation occurs.  ...  of the observed fluid in real-time.  ...  The authors also thank the support of ESI Group through the project UZ-2019-0060.  ... 
arXiv:2106.13301v2 fatcat:vllo3bmvrbfvfjbmuuidhemtxi

Pattern formation in nonequilibrium physics

J. P. Gollub, J. S. Langer
1999 Reviews of Modern Physics  
The authors describe and compare some of these phenomena, offer reflections on their similarities and differences, and consider possibilities for the future development of this field.  ...  Remarkable and varied pattern-forming phenomena occur in fluids and in phase transformations.  ...  In fact, the phase-space paradigm can give only a caricature of the real physics because of the large number of relevant degrees of freedom involved in most turbulent flows.  ... 
doi:10.1103/revmodphys.71.s396 fatcat:xtg46tfefjbcfa4lhvdepiojwu

Pattern Formation in Nonequilibrium Physics [chapter]

J. P. Gollub, J. S. Langer
1999 More Things in Heaven and Earth  
The authors describe and compare some of these phenomena, offer reflections on their similarities and differences, and consider possibilities for the future development of this field.  ...  Remarkable and varied pattern-forming phenomena occur in fluids and in phase transformations.  ...  In fact, the phase-space paradigm can give only a caricature of the real physics because of the large number of relevant degrees of freedom involved in most turbulent flows.  ... 
doi:10.1007/978-1-4612-1512-7_43 fatcat:xm36onqwb5f77gj2kzp5xrwati

Physics-based Deep Learning [article]

Nils Thuerey and Philipp Holl and Maximilian Mueller and Patrick Schnell and Felix Trost and Kiwon Um
2021 arXiv   pre-print
This digital book contains a practical and comprehensive introduction of everything related to deep learning in the context of physical simulations.  ...  As much as possible, all topics come with hands-on code examples in the form of Jupyter notebooks to quickly get started.  ...  Latent-space Physics: Towards Learning the Temporal Evolution of Fluid Flow. Comp. Grap. Forum, 38(2):12, 2019. URL: https://ge.in. tum.de/publications/latent-space-physics/.  ... 
arXiv:2109.05237v2 fatcat:dm2wyckg6fcxzhsxi4hmo76sny

Robust learning from noisy, incomplete, high-dimensional experimental data via physically constrained symbolic regression

Patrick A. K. Reinbold, Logan M. Kageorge, Michael F. Schatz, Roman O. Grigoriev
2021 Nature Communications  
We illustrate this using an experimental weakly turbulent fluid flow where only the velocity field is accessible.  ...  We also show that this hybrid approach allows reconstruction of the inaccessible variables – the pressure and forcing field driving the flow.  ...  The experimental data used in this work were produced by Jeff Tithof. The magnetic field measurements were performed with assistance from Charles Haynes. Author contributions  ... 
doi:10.1038/s41467-021-23479-0 pmid:34050155 fatcat:uppkvycsqnhw3b7xlcit3onmrq

Social physics [article]

Marko Jusup, Petter Holme, Kiyoshi Kanazawa, Misako Takayasu, Ivan Romic, Zhen Wang, Suncana Gecek, Tomislav Lipic, Boris Podobnik, Lin Wang, Wei Luo, Tin Klanjscek (+3 others)
2021 arXiv   pre-print
Recent decades have seen a rise in the use of physics-inspired or physics-like methods in attempts to resolve diverse societal problems.  ...  Such a rise is driven both by physicists venturing outside of their traditional domain of interest, but also by scientists from other domains who wish to mimic the enormous success of physics throughout  ...  Fluid-dynamical models. Macroscopic, or fluid-dynamic, models of traffic only use traffic density ρ, flow Q, and average velocity v as variables describing the system.  ... 
arXiv:2110.01866v1 fatcat:ccfxyezl6zgddd6uvrxubmaxua

Subspace Graph Physics: Real-Time Rigid Body-Driven Granular Flow Simulation [article]

Amin Haeri, Krzysztof Skonieczny
2021 arXiv   pre-print
Also, a new direction for real-time physics modeling is the use of deep learning.  ...  This research advances machine learning methods for modeling rigid body-driven granular flows, for application to terrestrial industrial machines as well as space robotics (where the effect of gravity  ...  Latent space physics: Towards learning the temporal evolution of fluid flow. Computer Graphics Forum 38, 71–82.  ... 
arXiv:2111.10206v1 fatcat:vntwi2zoevfsrfmvohvydhqqte

Spline-PINN: Approaching PDEs without Data using Fast, Physics-Informed Hermite-Spline CNNs [article]

Nils Wandel, Michael Weinmann, Michael Neidlin, Reinhard Klein
2021 arXiv   pre-print
In this paper, we propose to approach the solution of PDEs based on a novel technique that combines the advantages of two recently emerging machine learning based approaches.  ...  First, physics-informed neural networks (PINNs) learn continuous solutions of PDEs and can be trained with little to no ground truth data. However, PINNs do not generalize well to unseen domains.  ...  [32] learn to solve the Reynolds-averaged Navier-Stokes equations, but the specific approach prevents a generalization beyond airfoil flows and discards the temporal evolution of the fluid state.  ... 
arXiv:2109.07143v1 fatcat:gtmhnavy2rab5irenkzbqh457a

Extracting Interpretable Physical Parameters from Spatiotemporal Systems using Unsupervised Learning [article]

Peter Y. Lu, Samuel Kim, Marin Soljačić
2019 arXiv   pre-print
The architecture is trained end-to-end and extracts latent parameters that parameterize the dynamics of a learned predictive model for the system.  ...  We demonstrate an unsupervised learning technique for extracting interpretable physical parameters from noisy spatiotemporal data and for building a transferable model of the system.  ...  The equation describes a diffusing quantity that is also affected by the flow or drift of the system, e.g. dye diffusing in a moving fluid. We consider the case of a constant velocity field.  ... 
arXiv:1907.06011v2 fatcat:a6r25xyk2nhqvamqgo3mwlkwqy

DisCo: Physics-Based Unsupervised Discovery of Coherent Structures in Spatiotemporal Systems [article]

Adam Rupe, Nalini Kumar, Vladislav Epifanov, Karthik Kashinath, Oleksandr Pavlyk, Frank Schlimbach, Mostofa Patwary, Sergey Maidanov, Victor Lee, Prabhat, James P. Crutchfield
2019 arXiv   pre-print
DisCo provides a scalable unsupervised physics-based representation learning method that decomposes spatiotemporal systems into their structurally relevant components, which are captured by the latent  ...  Complementary and alternative to supervised machine learning approaches, unsupervised physics-based methods based on behavior-driven theories hold great promise.  ...  underlying physical and causal mechanisms, and are better able to predict the occurrence and evolution of these phenomena over time.  ... 
arXiv:1909.11822v1 fatcat:zcmyoz6nbvdm3mgefb5rjy4gwi

A deeper look into natural sciences with physics-based and data-driven measures

Davi Röhe Rodrigues, Karin Everschor-Sitte, Susanne Gerber, Illia Horenko
2021 iScience  
In this perspective, we discuss some state-of-the-art data-driven tools to analyze latent effects in data and explain their applicability in natural science, focusing on two recently introduced, physics-motivated  ...  and a latent network of glymphatic channels from the mouse brain microscopy data.  ...  ACKNOWLEDGMENTS We would like to thank the group of M. Klä ui, in particular J. Zazvorká , for discussions and contributing with the experimental data used in Figure 4 .  ... 
doi:10.1016/j.isci.2021.102171 pmid:33665584 pmcid:PMC7907479 fatcat:l47w7qmbkfg4bexhcx7jdlcgqa

RELATIVISTIC QUARK PHYSICS

Johann Rafelski
1998 Relativistic Aspects of Nuclear Physics  
We present a brief survey of the development of nuclear physics towards relativistic quark physics.  ...  This is followed by a thorough discussion of the quest for the observation of the dissolution of nuclear matter into the deconfined quark matter (QGP) in relativistic nuclear collisions.  ...  This research program is supported by US-Department of Energy under grant DE-FG03-95ER40937 .  ... 
doi:10.1142/9789814528917_0016 fatcat:cqwwcieqojcs5iai4u6mcgyeae
« Previous Showing results 1 — 15 out of 4,887 results