Machine learning for physics-informed generation of dispersed multiphase flow using generative adversarial networks

Siddani, Bhargav Sriram; Balachandar, S.; Moore, W. C.; Yang, Yunchao; Fang, Ruogu

doi:10.1007/s00162-021-00593-9

Cited by 35 publications

(16 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…PhysGAN have been applied to quantify real-world data uncertainty in various domains, which includes flood prediction [17], blood alcohol concentration prediction [18], and porous media flow modeling [19]. PhysGANs have not been used to quantify uncertainty in traffic state estimation.…”

Section: Related Workmentioning

confidence: 99%

Quantifying Uncertainty In Traffic State Estimation Using Generative Adversarial Networks

Mo¹,

Fu²,

Di³

2022

Preprint

View full text Add to dashboard Cite

This paper aims to quantify uncertainty in traffic state estimation (TSE) using the generative adversarial network based physics-informed deep learning (PIDL). The uncertainty of the focus arises from fundamental diagrams, in other words, the mapping from traffic density to velocity. To quantify uncertainty for the TSE problem is to characterize the robustness of predicted traffic states. Since its inception, generative adversarial networks (GAN) has become a popular probabilistic machine learning framework. In this paper, we will inform the GAN based predictions using stochastic traffic flow models and develop a GAN based PIDL framework for TSE, named "PhysGAN-TSE". By conducting experiments on a real-world dataset, the Next Generation SIMulation (NGSIM) dataset, this method is shown to be more robust for uncertainty quantification than the pure GAN model or pure traffic flow models. Two physics models, the Lighthill-Whitham-Richards (LWR) and the Aw-Rascle-Zhang (ARZ) models, are compared as the physics components for the PhysGAN, and results show that the ARZ-based PhysGAN achieves a better performance than the LWR-based one.

show abstract

Section: Related Workmentioning

confidence: 99%

Quantifying Uncertainty In Traffic State Estimation Using Generative Adversarial Networks

Mo¹,

Fu²,

Di³

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Thus, the first step of the curation process is to make each of the N particle within the domain to be the reference particle and consider a local volume around it for the purposes of microscale modelling. The local volume is typically much smaller than the entire computational box and determines the number of M closest neighbors that are within which will be considered in the microscale modeling [9,24,29]. The location of the M neighbors within the local volume define the microscope environment of the reference particle, while averages within the local volume are used to define the mesoscale Re and φ.…”

Section: Particle-resolved Simulation-data and Curationmentioning

confidence: 99%

“…The flow field around any reference particle can be recreated by adding its own superposable wake and those of each neighbor. Siddani et al [24] pursued this flow prediction task using conditional generative adversarial network (GAN) [25,26]. The volumetric representation of the M neighbors ((M +1)-body input) and the convolutional neural network (CNN) architecture enabled the approach in reconstructing the complex flow through the random bed of particles far more accurately than the superposable wake approach.…”

Section: Introductionmentioning

confidence: 99%

Point-particle drag, lift, and torque closure models using machine learning: hierarchical approach and interpretability

Siddani¹,

Balachandar²

2022

Preprint

Self Cite

View full text Add to dashboard Cite

Developing deterministic neighborhood-informed point-particle closure models using machine learning has garnered interest in recent times from dispersed multiphase flow community. The robustness of neural models for this complex multi-body problem is hindered by the availability of particle-resolved data. The present work addresses this unavoidable limitation of data paucity by implementing two strategies: (i) by using a rotation and reflection equivariant neural network and (ii) by pursuing a physics-based hierarchical machine learning approach. The resulting machine learned models are observed to achieve a maximum accuracy of 85% and 96% in the prediction of neighborinduced force and torque fluctuations, respectively, for a wide range of Reynolds number and volume fraction conditions considered. Furthermore, we pursue force and torque network architectures that provide universal prediction spanning a wide range of Reynolds number (0.25 ≤ Re ≤ 250) and particle volume fraction (0 ≤ φ ≤ 0.4). The hierarchical nature of the approach enables improved prediction of quantities such as streamwise torque, by going beyond binary interactions to include trinary interactions.

show abstract

“…To improve training efficiency, there is a growing trend integrating the model-based model and the data-driven model together, namely, physics-informed deep learning (PIDL) [12]. Physics-informed GAN is the most widely-used physics-informed generative model, which has been applied to solving stochastic differential equations [18,19,4] and quantifying uncertainty in various domains [15,14]. In contrast, we only find one paper that applies PIDL to normalizing flow [10].…”

Section: Introductionmentioning

confidence: 99%

TrafficFlowGAN: Physics-informed Flow based Generative Adversarial Network for Uncertainty Quantification

Mo¹,

Fu²,

Xu³

et al. 2022

Preprint

View full text Add to dashboard Cite

This paper proposes the TrafficFlowGAN, a physics-informed flow based generative adversarial network (GAN), for uncertainty quantification (UQ) of dynamical systems. TrafficFlowGAN adopts a normalizing flow model as the generator to explicitly estimate the data likelihood. This flow model is trained to maximize the data likelihood and to generate synthetic data that can fool a convolutional discriminator. We further regularize this training process using prior physics information, so-called physics informed deep learning (PIDL). To the best of our knowledge, we are the first to propose an integration of flow, GAN and PIDL for the UQ problems. We take the traffic state estimation (TSE), which aims to estimate the traffic variables (e.g. traffic density and velocity) using partially observed data, as an example to demonstrate the performance of our proposed model. We conduct numerical experiments where the proposed model is applied to learn the solutions of stochastic differential equations. The results demonstrate the robustness and accuracy of the proposed model, together with the ability to learn a machine learning surrogate model. We also test it on a real-world dataset, the Next Generation SIMulation (NGSIM), to show that the proposed TrafficFlowGAN can outperform the baselines, including the pure flow model, the physics-informed flow model, and the flow based GAN model.

show abstract

Machine learning for physics-informed generation of dispersed multiphase flow using generative adversarial networks

Cited by 35 publications

References 27 publications

Quantifying Uncertainty In Traffic State Estimation Using Generative Adversarial Networks

Quantifying Uncertainty In Traffic State Estimation Using Generative Adversarial Networks

Point-particle drag, lift, and torque closure models using machine learning: hierarchical approach and interpretability

TrafficFlowGAN: Physics-informed Flow based Generative Adversarial Network for Uncertainty Quantification

Contact Info

Product

Resources

About