Showing 1–6 of 6 results for author: Schmidt, D P

Improving CFD simulations by local machine-learned correction

Authors: Peetak Mitra, Majid Haghshenas, Niccolo Dal Santo, Conor Daly, David P. Schmidt

Abstract: High-fidelity computational fluid dynamics (CFD) simulations for design space explorations can be exceedingly expensive due to the cost associated with resolving the finer scales. This computational cost/accuracy trade-off is a major challenge for modern CFD simulations. In the present study, we propose a method that uses a trained machine learning model that has learned to predict the discretizat… ▽ More High-fidelity computational fluid dynamics (CFD) simulations for design space explorations can be exceedingly expensive due to the cost associated with resolving the finer scales. This computational cost/accuracy trade-off is a major challenge for modern CFD simulations. In the present study, we propose a method that uses a trained machine learning model that has learned to predict the discretization error as a function of largescale flow features to inversely estimate the degree of lost information due to mesh coarsening. This information is then added back to the low-resolution solution during runtime, thereby enhancing the quality of the under-resolved coarse mesh simulation. The use of a coarser mesh produces a non-linear benefit in speed while the cost of inferring and correcting for the lost information has a linear cost. We demonstrate the numerical stability of a problem of engineering interest, a 3D turbulent channel flow. In addition to this demonstration, we further show the potential for speedup without sacrificing solution accuracy using this method, thereby making the cost/accuracy trade-off of CFD more favorable. △ Less

Submitted 28 April, 2023; originally announced May 2023.

Comments: 7 pages, under review at ASME IMECE 2023 conference

Report number: vol. 87660, p. V009T10A062

Journal ref: In ASME International Mechanical Engineering Congress and Exposition, vol. 87660, p. V009T10A062. American Society of Mechanical Engineers, 2023

Validation and parameterization of a novel physics-constrained neural dynamics model applied to turbulent fluid flow

Authors: Varun Shankar, Gavin D. Portwood, Arvind T. Mohan, Peetak P. Mitra, Dilip Krishnamurthy, Christopher Rackauckas, Lucas A. Wilson, David P. Schmidt, Venkatasubramanian Viswanathan

Abstract: In fluid physics, data-driven models to enhance or accelerate solution methods are becoming increasingly popular for many application domains, such as alternatives to turbulence closures, system surrogates, or for new physics discovery. In the context of reduced order models of high-dimensional time-dependent fluid systems, machine learning methods grant the benefit of automated learning from data… ▽ More In fluid physics, data-driven models to enhance or accelerate solution methods are becoming increasingly popular for many application domains, such as alternatives to turbulence closures, system surrogates, or for new physics discovery. In the context of reduced order models of high-dimensional time-dependent fluid systems, machine learning methods grant the benefit of automated learning from data, but the burden of a model lies on its reduced-order representation of both the fluid state and physical dynamics. In this work, we build a physics-constrained, data-driven reduced order model for the Navier-Stokes equations to approximate spatio-temporal turbulent fluid dynamics. The model design choices mimic numerical and physical constraints by, for example, implicitly enforcing the incompressibility constraint and utilizing continuous Neural Ordinary Differential Equations for tracking the evolution of the differential equation. We demonstrate this technique on three-dimensional, moderate Reynolds number turbulent fluid flow. In assessing the statistical quality and characteristics of the machine-learned model through rigorous diagnostic tests, we find that our model is capable of reconstructing the dynamics of the flow over large integral timescales, favoring accuracy at the larger length scales. More significantly, comprehensive diagnostics suggest that physically-interpretable model parameters, corresponding to the representations of the fluid state and dynamics, have attributable and quantifiable impact on the quality of the model predictions and computational complexity. △ Less

Submitted 21 October, 2021; originally announced October 2021.

Comments: Submitted to Physical Review Fluids

Improved Methods for Mixing-Limited Spray Modeling

Authors: Majid Haghshenas, Peetak P. Mitra, Chu Wang, Fabien Tagliante, Lyle Pickett, David P. Schmidt

Abstract: The realization that interfacial features play little role in diesel spray vaporization and advection has changed the modus operandi for spray modeling. Lagrangian particle tracking has typically been focused on droplet behavior, with sub-models for breakup, collision, and interfacially limited vaporization. In contrast, the mixing-oriented spray models are constructed so that gas entrainment is t… ▽ More The realization that interfacial features play little role in diesel spray vaporization and advection has changed the modus operandi for spray modeling. Lagrangian particle tracking has typically been focused on droplet behavior, with sub-models for breakup, collision, and interfacially limited vaporization. In contrast, the mixing-oriented spray models are constructed so that gas entrainment is the limiting factor in the evolution of momentum and energy. In this work, a new spray model, ELMO (Eulerian-Lagrangian Mixing-Oriented), is implemented in a three-dimensional CFD code with two-way coupling with the gas phase. The model is verified and validated with three canonical sprays including spray A, H, and G from the ECN database. △ Less

Submitted 7 July, 2021; originally announced July 2021.

Comments: Presented at the ILASS-Americas 31st Annual Conference on Liquid Atomization and Spray Systems, May 2021. arXiv admin note: text overlap with arXiv:2107.03508

The Eulerian Lagrangian Mixing-Oriented (ELMO) Model

Authors: David P. Schmidt, Majid Haghshenas, Peetak P. Mitra, Chu Wang, P. Kelly Senecal, Fabien Tagliante, Lyle M. Pickett

Abstract: Past Lagrangian/Eulerian modeling has served as a poor match for the mixing limited physics present in many sprays. Though these Lagrangian/Eulerian methods are popular for their low cost, they are ill-suited for the physics of the dense spray core and suffer from limited predictive power. A new spray model, based on mixing limited physics, has been constructed and implemented in a multi-dimension… ▽ More Past Lagrangian/Eulerian modeling has served as a poor match for the mixing limited physics present in many sprays. Though these Lagrangian/Eulerian methods are popular for their low cost, they are ill-suited for the physics of the dense spray core and suffer from limited predictive power. A new spray model, based on mixing limited physics, has been constructed and implemented in a multi-dimensional CFD code. The spray model assumes local thermal and inertial equilibrium, with air entrainment being limited by the conical nature of the spray. The model experiences full two-way coupling of mass, momentum, species, and energy. An advantage of this approach is the use of relatively few modeling constants. The model is validated with three different sprays representing a range of conditions in diesel and gasoline engines. △ Less

Submitted 7 July, 2021; originally announced July 2021.

Comments: Submitted to International Journal of Multiphase Flow

Turbulence forecasting via Neural ODE

Authors: Gavin D. Portwood, Peetak P. Mitra, Mateus Dias Ribeiro, Tan Minh Nguyen, Balasubramanya T. Nadiga, Juan A. Saenz, Michael Chertkov, Animesh Garg, Anima Anandkumar, Andreas Dengel, Richard Baraniuk, David P. Schmidt

Abstract: Fluid turbulence is characterized by strong coupling across a broad range of scales. Furthermore, besides the usual local cascades, such coupling may extend to interactions that are non-local in scale-space. As such the computational demands associated with explicitly resolving the full set of scales and their interactions, as in the Direct Numerical Simulation (DNS) of the Navier-Stokes equations… ▽ More Fluid turbulence is characterized by strong coupling across a broad range of scales. Furthermore, besides the usual local cascades, such coupling may extend to interactions that are non-local in scale-space. As such the computational demands associated with explicitly resolving the full set of scales and their interactions, as in the Direct Numerical Simulation (DNS) of the Navier-Stokes equations, in most problems of practical interest are so high that reduced modeling of scales and interactions is required before further progress can be made. While popular reduced models are typically based on phenomenological modeling of relevant turbulent processes, recent advances in machine learning techniques have energized efforts to further improve the accuracy of such reduced models. In contrast to such efforts that seek to improve an existing turbulence model, we propose a machine learning(ML) methodology that captures, de novo, underlying turbulence phenomenology without a pre-specified model form. To illustrate the approach, we consider transient modeling of the dissipation of turbulent kinetic energy, a fundamental turbulent process that is central to a wide range of turbulence models using a Neural ODE approach. After presenting details of the methodology, we show that this approach outperforms state-of-the-art approaches. △ Less

Submitted 12 November, 2019; originally announced November 2019.

An Analysis of the Convergence of Stochastic Lagrangian/Eulerian Spray Simulations

Authors: David P. Schmidt, Frederick Bedford

Abstract: This work derives how the convergence of stochastic Lagrangian/Eulerian simulations depends on the number of computational parcels, particularly for the case of spray modeling. A new, simple, formula is derived that can be used for managing the numerical error in two or three dimensional computational studies. For example, kee** the number of parcels per cell constant as the mesh is refined yiel… ▽ More This work derives how the convergence of stochastic Lagrangian/Eulerian simulations depends on the number of computational parcels, particularly for the case of spray modeling. A new, simple, formula is derived that can be used for managing the numerical error in two or three dimensional computational studies. For example, kee** the number of parcels per cell constant as the mesh is refined yields an order one-half convergence rate in transient spray simulations. First order convergence would require a doubling of the number of parcels per cell when the cell size is halved. Second order convergence would require increasing the number of parcels per cell by a factor of eight. The results show that controlling statistical error requires dramatically larger numbers of parcels than have typically been used, which explains why convergence has been so elusive. △ Less

Submitted 30 January, 2018; originally announced January 2018.