Upcoming Seminars:
Professor Takeshi Sato
Advanced Manufacturing Technology Institute,
Kanazawa University, Japan
Rheo-SINDy: Finding a constitutive model from rheological data for complex fluids using sparse identification for nonlinear dynamics
September 16, 2024; NCRC South Atrium
Click here to add to Google Calendar
Abstract: Rheology plays a pivotal role in understanding the flow behavior of fluids by discovering governing equations that relate deformation and stress, known as constitutive equations. Despite the importance of these equations, current methods for deriving them lack a systematic methodology, often relying on sense of physics and incurring substantial costs. To overcome this problem, we propose a novel method named Rheo-SINDy, which employs the sparse identification of nonlinear dynamics (SINDy) algorithm for discovering constitutive models from rheological data. Rheo-SINDy was applied to five distinct scenarios, four with well-established constitutive equations and one without predefined equations. Our results demonstrate that Rheo-SINDy successfully identified accurate models for the known constitutive equations and derived physically plausible approximate models for the scenario without established equations. Notably, the identified approximate models can accurately reproduce nonlinear shear rheological properties, especially at steady state, including shear thinning. These findings validate the robustness of Rheo-SINDy in handling data complexities and underscore its efficacy as a tool for advancing the development of data-driven approaches in rheology. Full paper: Takeshi Sato, Souta Miyamoto, Shota Kato, arXiv:2403.14980
Past Seminars:
2024 Summer Seminar Series
Yi Dai
Graduate Student
(Allman Group)
Learning Classifier Models for Dynamic Reconfiguration of Modular Facilities
August 21, 2024; NCRC B32 Auditorium
Click here to add to Google Calendar
Abstract: The presence of discrete decisions in mixed-integer model predictive control (MPC) renders the optimization problem more significantly difficult to solve than a traditional continuous MPC problem. In this work, we present an approach to determine integer control decisions a priori to solving the MPC problem using data-driven machine learning algorithms. To demonstrate the efficacy of the proposed approach, we present the control performance of a benchmark system with three modular nonisothermal CSTR’s operating in four potential configurations. Compared to the MPC with fixed configurations, the MPC with machine learning classifiers for reconfiguration gives better control performance of shorter recovering time and smaller deviation from the setpoint.
Linghao Shi
Graduate Student
(Larson Group)
Pharmaceutical crystal growth with coarse-grained MD simulation
July 10, 2024; NCRC B32 Auditorium
Click here to add to Google Calendar
Abstract: Due to the challenges of simulating pharmaceutical crystal growth with brute-force MD simulation, we developed coarse-grained force fields for two kinds of representations. The first, a lower-resolution model, was coarse-grained via Iterative Boltzmann Inversion, whereas the second, a finer-resolution one, was achieved by the optimization of MARTINI force field. We were able to observe crystal growth with both force fields and compared the strengths and weaknesses of two different models.
Dr. Philipp Schönhöfer
Post doctoral Scholar
(Glotzer Group)
Geometrical Control of Active Matter: How to turn self-propelled colloids into colloidal robots
June 19, 2024; NCRC B32 Auditorium
Click here to add to Google Calendar
Abstract: It is well established that the collective and self-emergent dynamics of active swarms in biology, colloidal science and modern-day robotics can be traced back mostly to the local interactions between neighbouring particles. While nature and robotics scientists developed intricate strategies/algorithms for organisms/robots to communicate on the local scale and thereby coordinate their global dynamics, we still lack the ability to synthesize colloidal particles with a similar degree of complexity. In that regard, we study computationally how shape and other morphological properties of active particles and their environment influence both the local structures and global dynamics in crowded active particle systems. In particular, we demonstrate mechanisms that can control the clustering and collective migration of self-propelled particles by using models of both rigid and deformable particles.
Bolton Tran
Post doctoral Scholar
(Goldsmith Group)
Grand Canonical DFT for Modeling Electrochemical CO Reduction
May 22, 2024; NCRC B32 Auditorium
Click here to add to Google Calendar
Abstract: The electrochemical reduction of carbon monoxide (CO) is relevant in various efforts to electrifying the chemical industry using renewable energy sources. Modeling the kinetics of electrochemical CO reduction requires accounting for the electrode-electrolyte interface and the applied voltage, which conventional density functional theory (DFT) fails at. A family of grand canonical DFT (GC-DFT) approaches have emerged recently to address this electrochemical modeling challenge. In this talk, I will outline the inner working of a self-consistent GC-DFT approach, and its implementation to investigate CO adsorption and reactions on a copper electrode. I will also discuss some inherent drawbacks of GC-DFT pertaining to its continuum electrolyte description.
Past Seminars:
Inaugural 2023 Summer Seminar Series
Computational modeling as an enabling tool in nanomedicine and polymer
May 5, 2023; NCRC B32 Auditorium
Click here to add to Google Calendar
Abstract: The rapid advancement of computer hardware and simulation methods has made computational simulation an increasingly important tool in the research of soft materials. Molecular simulation, in particular, has become an irreplaceable tool to fill gaps in experiments and theory. However, the complex nature of soft materials and the broad range of spatial and temporal scales that their phenomena span present challenges for molecular modeling in designing reasonable systems and characterizing simulation results. In this talk, I will discuss my previous works on molecular simulations of nanomedicine and polymer to demonstrate how multiscale models and characterization tools can help address these challenges.
In Silico Characterization of Materials for use in Rechargeable Magnesium Batteries
June 9, 2023; NCRC B32 Auditorium
Click here to add to Google Calendar
Abstract: Non-aqueous Rechargeable Magnesium batteries (RMBs) represent a safer, cheaper and more powerful alternative to lithium-ion battery technology. However, at electrode/electrolyte interfaces (called Solid-Electrolyte Interfaces, SEIs), the high charge density of magnesium ions triggers the formation of ion impermeable deposits. To overcome the stability and reversibility issues caused by SEIs, the electrolytes used must be stable against reductive reactions at the anode/electrolyte interface during charge/discharge cycles in RMBs. In this talk, I discuss my investigations into ion agglomeration and transport charge carriers in Magnesium electrolytes in ethereal solvents under both equilibrium and operating conditions using large scale atomistic simulations.
The Many Ways to Assemble Open Crystals
July 7, 2023; NCRC B32 Auditorium
Click here to add to Google Calendar
Abstract: Open structures have a myriad of applications in engineering including adsorption and particle transport. However, the ability to tune the assembly of open structures, and thereby the properties of such structures, is often hindered due to a lack of fundamental understanding of why these building blocks self assemble into different crystals. In this talk, I show how to engineer assembly in open structures by either changing the shape of the building block for 2D colloidal crystals or engineering the interactions of the building block for 3D DNA nanocrystals. By understanding the interplay between completing structures and building block properties, I elucidate the assembly mechanisms that drive different crystals to form, opening new avenues for design of open crystal structures.
Self Assembly of Anisotropic Particles on Curved Interfaces
August 4, 2023; NCRC B32 Auditorium
Click here to add to Google Calendar
Abstract: Pickering emulsions are stabilized by the addition of colloidal particles, jammed at the interface between two phases. Much like how a soccer ball cannot be decorated solely with hexagonal patches, spherical particles that decorate a spherical surface cannot pack into a perfect hexagonal lattice. Here we simulate emulsion systems stabilized using faceted shape anisotropic colloidal particles on the surface of a sphere. Due to the shape anisotropy of the particles, they exhibit unique defect patterns that cannot be precisely analyzed using traditional analyses. We develop techniques to show how particle jamming and void creation disrupt particle packing for the five Platonic and four additional Archimedean shapes using both hard particle Monte Carlo and molecular dynamics simulation methods.
How can we model the physics of malaria blood?
August 25, 2023; NCRC B32 Auditorium
Click here to add to Google Calendar
Abstract: Blood isn’t a well-mixed suspension; instead, blood is a complex fluid with rich rheological behavior from deformable red blood cells colliding in flow. Many diseases, including malaria, alter the properties of red blood cells and change their distribution within the blood vessel in detrimental ways. I will briefly discuss why we want a simulated system for therapeutic development. Then I will walk through the simulation development, focusing on method selection and the benefits of using an immersed-boundary-lattice Boltzmann implementation. Lastly, I will present preliminary findings showing red blood cell margination changes during malaria infection.