Amanda Howard

Amanda Howard

Mathematician

Computational Fluid Dynamics

Suspension Modeling

Densely packed suspensions of spherical particles are a chaotic dynamical system with fascinating evolution. I focus on non-homogenous flows, particularly flows with sharp gradients in the volume fraction or bidisperse particle sizes or weights. My most recent work focuses on using machine learning to extract more information from simulations or experiments.

Rotating_cloud from Amanda on Vimeo.

Relevant publications:

  1. Howard, Amanda A., Dong, Justin, Patel, Ravi, D’Elia, Marta, Maxey, Martin R., & Stinis, Panos. (2023). Machine learning methods for particle stress development in suspension Poiseuille flows. Rheologica Acta, 1-28.

  2. Howard, Amanda A., Maxey, Martin R., & Gallier, Stany. (2022). A bidisperse suspension balance model. Phys. Rev. Fluids, 7, 12.

  3. Reyes, Brandon, Howard, Amanda A., Perdikaris, Paris, & Tartakovsky, Alexandre M. (2021). Learning unknown physics of non-Newtonian fluids. Phys. Rev. Fluids, 6, 073301.

  4. Howard, Amanda A., Maxey, Martin R., & Yeo, Kyongmin (2018). Settling of heavy particles in concentrated suspensions of neutrally buoyant particles under uniform shear. Fluid Dynamics Research, 4, 041401.

  5. Howard, Amanda A. & Maxey, Martin R. (2018). Simulation study of particle clouds in oscillating shear flow. J. Fluid Mech, 852, 484-506.

  6. Cui, Francis R., Howard, Amanda A., Maxey, Martin R. & Tripathi, Anubhav (2017). Dispersion of a suspension plug in oscillatory pressure-driven flow. Phys. Rev. Fluids, 2, 094303.

Nonlocal models for multiphase flow

We propose a non-local model for surface tension obtained in the form of an integral of a molecular-force-like functionadded to the Navier-Stokes momentum conservation equation. The non-local surface tension model has an inherent length scale that can be tuned to model both macro-scale and nano-scale fluids.

The non-local surface tension offers computational advantages over local surface tension models, because it relies on computing integrals instead of approximating a smoothed Dirac delta function, which can be present difficulties when the interface width is changing.

Relevant publications:

  1. Howard, Amanda A., & Tartakovsky, Alexandre M. (2021). A conservative level set method for N-phase flows with a free-energy-based surface tension model. Journal of Computational Physics, 426, 109955.

  2. Howard, Amanda A., & Tartakovsky, Alexandre M. (2020). Non-local model for surface tension in fluid-fluid simulations. Journal of Computational Physics, 109732.

  3. Howard, Amanda A., Zhou, Yongcheng, & Tartakovsky, Alexandre M. (2019). Analytical steady-state solutions for pressure with a multiscale non-local model for two-fluid systems. arXiv:1905.08052.