The Fluid-Structures-Materials seminars during the Spring semester of 2023 will take place on Wednesdays,听4:00PM in AERO 111. A list of past and upcoming presentation is provided below.
We're looking forward to seeing you there.
The organizing committee,
Aviral Prakash, Thomas Calascione, Nils Wunsch and Jim Brasseur
Basu Dipak Parmar
PhD Candidate
Aerospace Engineering Sciences
葫芦娃视频
Abstract: Despite the increase in computational power, RANS models continue to be the workhorse in most industrial听and academic CFD solvers. Traditional听RANS models underperform in cases involving flow separation. adverse pressure听gradient and secondary flows. There is more high-fidelity data听available in the past decade, but it's hard to draw insight to develop better RANS models due to the sheer amount of data and complex physics. My research focuses on Data-Driven turbulence听modeling which utilizes high-fidelity data and state of the art machine learning听algorithms to develop听more accurate RANS models. The development of data-driven RANS models is described with consideration of invariant properties, Boussinesq approximation and stability. The DD-RANS models are tested on cases involving flow separation and perform better than the commonly used RANS models. A brief overview of interactive machine learning will be discussed, in which a machine learning model is interactively trained using an ongoing live high-fidelity simulation.
Connor Morency
PhD Student
Aerospace Engineering Sciences
葫芦娃视频
Abstract: The topics discussed in this presentation will include unstructured meshing techniques, discontinuity (shock) capturing, RANS turbulence modeling, and the setup, execution, and evaluation of supersonic CFD simulations. Specifically, the verification and validation of the finite-element CFD solver PHASTA at high supersonic speeds and the additional contributions resulting from this investigation are presented. PHASTA is verified via a mesh refinement study and validated via comparison to previous research regarding shock stand-off distance and turbulent supersonic boundary layers. A novel shock capturing technique using an artificial diffusion-based shock sensor is introduced. Shock stand-off distance and turbulent supersonic boundary layer parameters are estimated in a single simulation of a diamond-shaped 2D fin, showing good agreement to previous research.
Alireza Doostan and Jim Brasseur
Professors
Aerospace Engineering Sciences
葫芦娃视频
This听workshop will cover the follwing topics:
Jim Brasseur
Professor
Aerospace Engineering Sciences
葫芦娃视频
Abstract: This discussion will integrate three series of research studies into the nonsteady forces and moments that are generated at the rotor hub of utility scale wind turbines by the continual passage through the wind turbine rotor of the energy-dominant turbulence eddies that are embedded within the daytime atmospheric boundary layer (ABL).听 One series, developed at Penn State, applied large-eddy simulation (LES) of a wind turbine within the ABL to identify three characteristic time scales in the aerodynamic response to the passage of atmospheric turbulence eddies through the wind turbine rotor at the minute, second and sub-second time scales. These temporal scales are confirmed with analysis at Penn State of data from a GE field study. The third study was developed by comparing LES results with analysis at the 葫芦娃视频 (UCB) of field data obtained from the NREL/GE 1.5 MW wind turbine at听 the National Wind Technology Center, a site to the east of the Front Range of the Rocky Mountains. The NREL wind turbine responds to mountain-generated turbulence eddies embedded within westerly winds and confirms a key result from the LES-based study鈥攖hat the continual passage of the energy-containing atmospheric eddies creates time-variations in the axial moment on the main shaft (torque and power) that are fundamentally different from the non-axial moment components that directly force the main bearing. The LES explains the turbulence-rotor interactions underlying these fundamentally different responses. The implication is that that methods to suppress the potentially deleterious impacts of atmospheric turbulence on main bearing failure (the numerator in the levelized cost of energy (LCOE)) must differ fundamentally from methods designed to suppress turbulence-induced time variations in power (the denominator in LCOE). Current research at UCB integrates aerodynamics-based models with models of the main bearing to determine the extent to which atmospheric turbulence increases main bearing failure and LCOE.
Yasara Dharmadasa
PhD Candidate
Aerospace Engineering Sciences
葫芦娃视频
Abstract: Deployable space structures are crucial for overcoming the volume limitations of launch vehicles and enabling the construction of large and complex space systems. A few of the design considerations are achieving a high packaging ratio, being stowed for a long duration, surviving launch vibrations, a flawless deployment, and being operational in the space environment. To meet these requirements, it is essential to have a clear understanding of the mechanics at the material and component levels. Our work focuses on the high curvature folding of composite flexures (for booms and elastic hinges) and thin sheets (for deployable membranes). We design a scalable hinge architecture using high-strain composites, develop analytic frameworks to assess their performance and fabricate test prototypes. Our analytic frameworks provide a guideline for achieving maximum compaction without material failure or loss of structural integrity during the operational stage. Thin sheets can achieve a high compaction ratio by folding in specific (origami-inspired) patterns, and the presence of residual creases affects the deployment dynamics and the deployed geometry of the sheet. We study the mechanics of a single crease by investigating the formation process and the subsequent unfolding using experiments. The observations are rationalized with analytic formulations and finite element modeling, deriving relations to the material and geometrical parameters of the sheet.
Kenneth Jansen
Professor
Aerospace Engineering Sciences
葫芦娃视频
Abstract: For several decades, scale-resolving turbulence simulations have served as grand challenge problems capable of saturating the largest available computer鈥檚 resources for as long (or longer) than the resource was available.听 Larger machines have allowed听expansion of geometric and flow complexity.听 In this talk, we will discuss the development of finite element methods for application to scale-resolving simulations on extreme-scale computational resources.听 The methods developed provide very low dissipation and higher order accurate discretizations. Their ability to use unstructured grids to match grid resolution to the local needs of the scale-resolving simulation makes them particularly attractive and efficient for complex geometry flows, and more fundamental flows with a large spatial variation in the smallest required scale that must be resolved.听 It is important to note that getting the solver to scale well and make efficient use of the emerging hardware is only one of the challenges of 鈥渆xtreme-scale computing.鈥� Other significant challenges addressed in this talk include preparing the inputs for simulations at this scale (pre-processing) and extracting meaningful insight from the massive spatial-temporal data stream that a petascale, and soon exascale, turbulence simulation produces.听This talk will discuss the general and specific challenges that have been addressed for unstructured grids in this area and will close with applications to boundary layers of aerodynamic interest.
John Evans
Associate Professor
Aerospace Engineering Sciences
葫芦娃视频
Abstract: Immersed finite element methods simplify the finite element solution of partial differential equations on complex and/or deforming domain geometries by relaxing the requirement that the finite element approximation space be defined on a body-fitted mesh.听 As a consequence, immersed finite element methods allow for the simulation of physical systems that are difficult, if not impossible, to simulate using classical finite element methods.听 However, the computer implementation of an immersed finite element method remains a challenging and time-consuming task, even for domain experts.听 The EXHUME (EXtraction for High-order Unfitted finite element MEthods) software library was introduced specifically to ease the burden of implementing immersed finite element methods.听 In particular, EXHUME enables one to transform classical finite element codes into immersed finite element codes with minimal implementation effort, empowering a larger community of scientists and engineers to employ immersed finite element methods in their own work.听 I will begin with an introduction to a new interpolation-based approach to immersed finite element analysis, the key technology behind the EXHUME software project.听 The efficacy of interpolation-based immersed finite element analysis will then be illustrated using example problems from heat conduction, structural mechanics, and fluid mechanics.听 Finally, the EXHUME software tools for generating the data structures necessary to perform interpolation-based finite element analysis will be presented, as well as how these tools have been leveraged to enable interpolation-based finite element analysis within the popular open-source finite element analysis platform, 鈥淔EniCS.鈥�
Manoj Settipalli
PhD Candidate
Aerospace Engineering Sciences
葫芦娃视频
Abstract: Quantum memories play a critical role in quantum information processing (QIP), facilitating the storage and retrieval of quantum information for on-chip and long-distance quantum communications, with potential applications ranging from aerospace to neuronal sensing. It is now well established that quantized vibrations (phonons) in mechanical oscillators can behave quantum mechanically under specific conditions and can play an important role in QIP. Bulk acoustic wave (BAW) phonons, which vibrate within the bulk of a material, are promising candidates for storing quantum information due to their long lifetimes. In this work, we investigate a hybrid photon-magnon-phonon system, where BAW phonons are excited in a Gadolinium Iron Garnet (GGG) substrate by quantized electron spin-waves (magnons) in a Yttrium Iron Garnet (YIG) thin film and are coupled to microwave photons. Recent experiments on a millimeter scale YIG/GGG device at room temperature show that the memories are limited by the phonon lifetime of 0.2 渭s. It is expected that operating in the milliKelvin regime will allow the memories to have a higher phonon lifetime, limited by diffraction, which is less understood. We present theoretical and numerical analysis to predict the diffraction-limited BAW phonon lifetime limits, mode shapes, and their coupling strengths to magnons. We further analyze the effects of magnon-phonon coupling and phonon lifetimes on the dynamics of the hybrid system using Heisenberg-Langevin equations and discuss their experimental implications.
听
Nils Wunsch
Phd Student
Aerospace Engineering Sciences
葫芦娃视频
Outline:
听
Marisa Petrusky
Aerospace Engineering Sciences
葫芦娃视频
Outline:
Mohammad Mokhtarzadeh Khanegahi
PhD Student
Aerospace Engineering Sciences
葫芦娃视频
Abstract: Multiscale topology optimization is a computational design method that uses optimization algorithms to generate efficient designs of structures at multiple length scales, from the microscale to the macroscale. Although the advent of novel additive manufacturing technologies offers a powerful tool to fabricate multiscale materials, bridging the scales between different scales in terms of geometry and mechanics still presents challenges. An example of a multiscale topology optimization method is the two-scale design method, whereby the properties of the microscale are averaged to obtain a property of the macroscale. Full-scale approaches are another paradigm in which, through finer discretization, multiscale structures naturally emerge. Although two-scale methods have been successfully applied, they pose challenges in terms of connectivity and compatibility of microstructures. Multiple post-processing steps are required to reinterpret the results into a mono-scale design. We consider simultaneous optimization of macroscale and microscale properties using homogenization methods while ensuring connectivity of microstructures. Our research focuses on fiber reinforced composites, where the performance of the system is optimized by simultaneously altering macroscale geometry and microscale fiber orientation at the microscale. The structural geometry is represented by a level set, which is approximated by quadratic B-spline functions. The fiber orientation field is parameterized with higher order B-splines on hierarchically refined meshes. Different levels of refinement are used to control the discretization of the fiber orientation field and ensure a smooth fiber layout. Furthermore, the parallel alignment of fiber paths is enforced by imposing novel penalty terms, which are incorporated into the optimization process. Numerical examples demonstrate with two-dimensional and three-dimensional configurations that the proposed method is efficient in simultaneously optimizing the macroscopic shape and the fiber layout while improving manufacturability by promoting parallel fiber paths.
Kwang-Chun Park
Professor
Aerospace Engineering Sciences
葫芦娃视频
Abstract:听Aristotle states 鈥渋f we are ignorant of what a motion is, we are of necessity ignorant of what nature is.鈥� Galileo declares dynamics is 鈥渁 subject of never-ending听听interest, perhaps the most important in nature, one which has engaged the minds of all听the great philosophers鈥︹�� This talk offers a guided tour into the origins of motions听and dynamics as conceived by great western philosophers, and chronologically听navigates through the great ideas dealing with its principles, theoretical developments,听and arriving at a maturing plateau. Today, dynamics once again is opening new doors听into expanding and exciting new 鈥渁rts,鈥� viz., data-based science and engineering.听