2016 Mechanical Engineering Abstracts

Poster Number: ME-01

Title: Fully Stressed Design Evolution Strategy for Layout Optimization of Truss Structures

Authors: Ali Ahrari; Kalyanmoy Deb

Abstract: Despite a huge number of publications on structural optimization, practitioners still prefer to rely on intuition-based try-and-error methods instead of rigorous optimization techniques. One of the critical factor that prompts such preference is that conventional test problems employed in academic research are too simple to simulate complexity of real structures. Furthermore, majority of optimization methods can handle the size optimization only, the potential saving from which is highly limited, when compared to the most sophisticated, and obviously the most challenging scenario, simultaneous topology, shape and size (TSS) optimization. In a recent study a method based on combination of optimality criteria and evolution strategies, called fully stressed design based on evolution strategies (FSD-ES), was proposed for TSS optimization of truss structures. In this research, an improved version of FSD-ES method, called FSD-ES-II, is proposed which can explicitly handle the displacement constraints and design requirements governed by standard design codes such as AISC-ASD. Performance of FSD-ES-II is compared with the best results available in the literature, often showing a significant superiority of the proposed approach. Second, realistic and large-scale test problems are developed to address a critical pitfall in academic research on structural optimization: the simplicity of conventional test problems. It is also demonstrated a huge saving in the cost can be achieved by more reliance on a potent optimization tool instead of human intuition. This study demonstrates a significant contribution of optimization in structural engineering and reduce the gap between academic research and practice on this topic.


Poster Number: ME-02

Title: Asynchronous Activation of a Canine Left Ventricular Model

Authors: Seyedborhan Alhosseinihamedani; Lik Chuan Lee

Abstract: Abnormal electrical activation of the left ventricle (LV) due to malfunction of bundle branches results in an asynchronous mechanical activation of the myocardium. Consequently, during several heart beats, different regions of the myocardium experience either an early (preexcitation) or a delayed electrical, and consequently, mechanical activation. This abnormality presents itself in the prolonged QRS complex in electrocardiography, and is associated with the conductivity of the tissue. This study investigates the effect of asynchronous activation of the myocardium in a canine left ventricle. The model was geometrically simplified to an ellipsoidal with constant thickness based on an existing literature. The simulation included an electrophysiology and a mechanics mesh which were coupled togethered. In the synchronous model, the LV was activated in endocardium in the septum region, whereas in the asynchronous model, activation occured in the epicardium in an opposite direction to that of the synchronous model. In the asynchronous model, the conductivity tensor was modified to present a prolonged activation. Results show that asynchronous activation of the heart slightly increased end systolic volume, while maintaining the same end diastolic volume. This resulted in a drop in stroke volume. Furthermore, the complete activation of LV was delayed from 20 msec in the synchronous model to 100 msec in the asynchronous model. This prolonged activation resulted in higher and more asymmetric strains in the asynchronous model. The results of strain at different locations are also presented and compared between two models. As a long-term goal we are looking for the possible growth stimuli that produce such an asymmetric pattern.


Poster Number: ME-03

Title: Gain Scheduling Control

Authors: Ali K. Al-Jiboory; Guoming Zhu

Abstract: This work develops a synthesis procedure for Static Output-Feedback (SOF) Gain-Scheduling (GS) controllers with guaranteed H2 performance for Linear Parameter Varying (LPV) systems, where the measurement of the scheduling parameters are affected by uncertainties. Unlike existing synthesis methods from literature, our conditions can handle the most general case where the time-varying parameters could affect all the open-loop matrices since the controller gain is synthesized via extra slack variables independently of the open-loop matrices. This is the main novelty of the developed approach. All scheduling parameters and associated uncertainties are modeled through multi-simplex framework, i.e., the Cartesian product of multiple unit simplexes. The control problem is solved through two-stage design procedure in terms of Parametrized Linear Matrix Inequalities (PLMIs). In the first stage, a state feedback GS is determined, then, this controller is used as an input parameter for the second stage, which synthesizes the Robust Gain-Scheduling (RGS) static output-feedback gain. Numerical examples demonstrate the effectiveness of the developed approach.


Poster Number: ME-04

Title: Micro-Structurally Motivated Constitutive Model for Human Skin

Authors: Sheng Chen; A. Ni Annaidh; Sara Roccabianca

Abstract: The purpose of this work is to describe skin from the mechanical perspective by employing a constrained mixture theory approach. Skin can be described as a nonlinear, anisotropic, nearly incompressible elastic material. Families of collagen fibers aligned along preferential directions endow skin with its nonlinear, anisotropic behavior, while elastin gives skin its elastic response. to describe the nonlinear, orthotropic behavior of skin we employed a mass averaged strain energy function (SEF). We considered 3 collagen fiber families, one fiber family in the direction of the Langer’s line and the other two in diagonally symmetric direction with respect to the Langer’s Line. Collagen fibers’ mechanical behavior has been described by using a Fung-type exponential SEF. Elastin has been described by using a Neo-Hookean constitutive model. 43 samples from 6 human subjects were collected and mechanically tested by means of a quasi-static uniaxial tensile test. Samples were grouped based on subject and direction with respect to the Langer’s line (i.e., 0°, 45°, 90°). Nonlinear least-squares regression was used to determine best-fit values of the material parameters for the averaged data of each direction. The best-fit material parameters show that the mechanical behavior of skin is dominated by the presence of collagen fiber, represented by the mechanical parameters of collagen being several orders of magnitude bigger than the one for elastin. Root mean square errors show that the model proposed in this work gives a good fit to experimental data.


Poster Number: ME-05

Title: Non-Uniform Mapping in Genetic Algorithms

Authors: Yashesh Dhebar; Kalyanmoy Deb

Abstract: Genetic Algorithms (GA) come under the category of Evolutionary Algorithms wherein a population of solutions (individuals) gets refined in every generation (iteration). After each iteration of GA, the population of solutions passes through three operators - 1) Selection, 2) Cross Over and 3) Mutation. Fitness of individuals of this mutated population is then computed. Greediness and biasing in algorithm is introduced in a "selection" operator, where good solutions get more priority over bad solutions. We, in our study have introduced an additional operator which operates on a mutated population (i.e. when population has passed through all three stages mentioned previously). Here, the population is pushed towards a "best-so-far" point. The amount of push each solution in a population experiences depends on a "non-uniform mapping" function, which is dependent on a parameter "eta". Research was conducted for various kind of "mapping functions"; on unconstrained and constrained single objective optimization problems. In every case, the "non-uniform" mapping approach outperformed standard GA in terms of convergence speed and accuracy.


Poster Number: ME-06

Title: Quantifying Losses in Hand Function: A Model for Use in Rehabilitation

Authors: Joshua P. Drost; Tamara Reid Bush

Abstract: Currently, the methods used to measure changes in hand function are primarily surveys which are subjective. A model that maps both the force and motion abilities will allow clinicians to objectively compare changes in hand function throughout rehabilitation and treatment. Recently, we modeled the changes in the range of motions of the hand caused by arthritis; however, force data for each finger were not included. The goal of this work was to determine forces associated with the index finger and map these to the kinematic workspace of participants with and without reduced hand functionality. Sixteen “Healthy” participants and fifteen “Arthritic” participants were included in this study. Maximum forces were measured in nineteen trials over the range of motion. The subject was asked to pull towards the wrist for ten trials and push normal to the palm for nine trials in different positions. After collection, the data were analyzed in terms of the position (x,y,z coordinate), direction and magnitude of the force applied. Arthritic subjects applied significantly lower forces than healthy subjects (ANCOVA, p<0.001), and used a smaller volume of 3D space (MANOVA, p<0.001) and used a smaller range of fingertip directions when applying force (ANCOVA, p=0.0199). Future work will include the measurement of forces for all fingers, and mapping to the 3D kinematic model. Clinically, this model is highly innovative and useful: Measures of motion and force will be gathered prior to intervention; mid-way through rehabilitation and after rehabilitation is complete to determine what level of function was restored.


Poster Number: ME-07

Title: Role of Computational Modeling in Understanding Arterial Adaptation

Authors: Hailu Getachew; Seungik Baek

Abstract: Adaptation of arteries, to sustained changes in blood flow and pressure, highly affects the cardiovascular circulation at different levels. Among many arterial adaptation studies that used mathematical models, a Constrained Mixture Approaches is becoming popular for modeling growth and remodeling of arteries, in which vascular walls are assumed to be made up of different constituents. By proper choice of the number of the constituents, we could predict arterial adaptation in various physiological ranges. Specifically, vascular smooth muscle proliferation and collagen turnover are critical for the vascular adaptation of arteries. In the computations, we investigate possible ranges of the kinetic parameters in collagen synthesis and several possible functions of degradation and their consequences; and then we narrow the possible relations between collagen turnover and arterial adaptation. This study shows that we could incorporate the chemical reactions involving collagens and muscle cells for arterial responses to external stimuli using relatively idealized models and yet, capture the key changes in arterial mass and geometry. It also indicates that both rates of collagen synthesis and degradation are tightly regulated by the mechanical stress. We believe that the model used in our analysis is beneficial to cardiovascular researchers, in the sense that it could be used as a tool for studying, for example, the reaction of arteries in the human brain to hypertension and hemodynamics disorders. Further investigation of the causes that affect rates of collagen turnover and muscle mass changes could enable to develop treatments for cardiovascular patients.

This work was supported in part by National Science Foundation (CMMI-1150376 and CBET-1148298) for SB and for JH


Poster Number: ME-08

Title: Computational Fluid Dynamic Simulation of Human Carotid Artery Bifurcation Based on Anatomy and Volumetric Blood Flow Rate Measured with Magnetic Resonance Imaging

Authors: Hamidreza Gharahi; Byron A. Zambrano; David C. Zhu; J. Kevin DeMarco; Seungik Baek

Abstract: Blood flow patterns and local hemodynamic parameters have been widely associated with the onset and progression of atherosclerosis in the carotid artery. Assessment of these parameters can be performed noninvasively using cine phase-contrast (PC) magnetic resonance imaging (MRI). In addition, in the last two decades, computational fluid dynamics (CFD) simulation in three dimensional models derived from anatomic medical images has been employed to investigate the blood flow in the carotid artery. This study developed a workflow of a subject-specific CFD analysis using MRI to enhance estimating hemodynamics of the carotid artery. Time-of-flight (toF) MRI scans were used to construct three-dimensional computational models. PC-MRI measurements were utilized to impose the boundary condition at the inlet and a 0-dimensional lumped parameter model was employed for the outflow boundary condition. The choice of different viscosity models of blood flow as a source of uncertainty was studied, by means of the axial velocity, wall shear stress, and oscillatory shear index. The sequence of workflow in CFD analysis was optimized for a healthy subject using PC-MRI. Then, a patient with carotid artery stenosis and the hemodynamic parameters were examined. The simulations indicated that the lumped parameter model used at the outlet gives physiologically reasonable values of hemodynamic parameters. Moreover, the dependence of hemodynamics parameters on the viscosity models was observed to vary for different geometries. Other factors, however, may be required for a more accurate CFD analysis, such as the segmentation and smoothness of the geometrical model, mechanical properties of the artery’s wall, and the prescribed velocity profile at the inlet.


Poster Number: ME-09

Title: Analysis and Modeling of a Turbulent Jet Ignition System for Internal Combustion Engines

Authors: Masumeh Gholamisheeri; Gerald Gentz; Bryce Thelen; Elisa Toulson

Abstract: The behavior of transient, compressible and combusting premixed methane-air jets was experimentally studied with high speed imaging in a Rapid Compression Machine (RCM). The jets were generated with a turbulent jet ignition system, which is a prechamber initiated combustion system. Prechamber combustion systems are advanced ignition systems which are under investigation in order to replace current automotive ignition systems. In the last several decades various prechamber combustion systems have been produced and investigated to increase efficiency and reduce pollutants through lean, low temperature combustion. The prechamber is a small volume chamber where an injector and spark plug are located and is connected to the main combustion chamber via one or multiple small orifices. Experiments were completed for turbulent jet ignition system orifice diameters of 2.0, 2.5 and 3.0 mm each at lean-to-stoichiometric equivalence ratios of 0.67, 0.8 and 1.0. The jet velocity at the orifice exit was calculated using mathematical correlations. The Mach number and Reynolds number were also computed. In addition, three-dimensional numerical simulation of the turbulent jet ignition combustion process of a premixed methane-air mixture in a RCM was performed using the Converge computational software. This research investigated the impacts of an auxiliary fueled prechamber on the burn rate and on the lean or dilute limit extension of the RCM. The numerical results are compared to data and optical images obtained from high speed imaging of combustion in the optically accessible RCM.

This work was supported in part by United States Department of Energy and National Science Foundation Partnership on Advanced Compustion Engines


Poster Number: ME-10

Title: Decoupling of Diameter and Pitch in Nanostructure Arrays Made by Colloidal Self Assembly

Authors: Xiaolu Huang; Matthew Bjork; Jack Jongwon Kim; Junghoon Yeom

Abstract: This paper reports the fabrication of ordered nanostructure array using colloidal self-assembly. Colloidal lithography, also known as nanosphere lithography (NSL), has been extensively and exhaustively utilized to create various nanostructures with the limitation in the resulting morphology and array spacing. Especially, independent control over the individual nanostructure size and array pitch remains a challenge and is the subject of this paper. Here, we show three different methods that expand the type of the nanostructure array produced from NSL. First, the combined technique of NSL and metal-assisted chemical etching (MACE) is shown to generate vertically-aligned Si nanowire (SiNW) array with the unprecedented dimensional control. Second, a stretchable elastomer with transfer printing is utilized to control the pitch of the original NS arrays, and with a custom-designed radial stretcher, a hexagonal symmetry of the resulting nanostructures is conserved. An array of sparsely ordered silicon or quartz nano pillars is obtained along with metallic nanostructures on NSs as etch masks. Finally, a double lift-off method is introduced to create an array of metallic nanodots that are not conventionally realized using the NSL template.


Poster Number: ME-11

Title: Nanorod Formation in a Gas Phase Plasma

Authors: Alborz Izadi; Rebecca J. Anthony

Abstract: Silicon (Si) nanocrystals have been the focus of much attention in recent years for their tunable optical properties, which arise due to quantum confinement. While spherical Si nano crystals have been explored in depth, Si nanorods (SiNRs) have the potential to exhibit different optoelectronic properties compared to their spatially isotropic counterparts, including polarized light emission and enhanced charge transport. There are several solution techniques including vapor-liquid-solid growth in which a gas flow contains silicon ions forms an optimized eutectic alloy with a metal nanodroplet in a batch process. In this study we present a method to streamline SiNR growth even further by combining hot-wire gold nano particle synthesis with plasma-based nanorod growth for freestanding SiNRs produced entirely in the gas-phase. Our preliminary work confirms formation of gold nano particles (AuNPs) in the hot-wire method. We supply a thin platinum wire, coated in gold, with electrical power at 10-15 W under argon flow and reduced pressure.


Poster Number: ME-12

Title: Probabilistic Collocation Method in Parameter Estimation Applied on an Abdominal Aortic Aneurysm Computational Model

Authors: Zhenxiang Jiang; Huan N. Do; Jungeun Choi;  Seungik Baek

Abstract: Abdominal aortic aneurysms (AAAs) is one of vascular diseases that could lead to a more than 90% of mortality rates. There is a crucial need for developing a patient-specific computational model of AAAs which can describe the growth and remodeling process for predicting the rupture. However, there are some parameters that can neither be measured in experiments nor estimated directly from the computational model. Therefore, the research for estimating those parameters indirectly based on large amount of data that are obtained from time dependent CT scan of AAAs are necessary. In the presented research, we employ bayesian calibration combined with a probabilistic collocation method to refine the damage function parameters. This work is expected to provide a better prediction of the growth and remodeling process and the probability of rupture.

This work was supported in part by NIH R01HL115185


Poster Number: ME-14

Title: Multi-Physics Modeling and Simulation of Anomalous Transport Using Distributed Order ODEs/PDEs

Authors: Ehsan Kharazmi; Mohsen Zayernouri

Abstract: Anomalous transport and nonlocal/history dependent effects in multi-physics systems are abundant in nature. Examples in many systems in science and engineering include: electrochemical processes, non-Brownian transport phenomena in porous and disordered materials, viscoelastic materials, bioengineering applications, non-Gaussian (Le'vy flights) processes in turbulent flows, non-Newtonian fluids and rheology, non-Markovian processes in multi-scale complex fluids and multi-phase applications. Fractional differential equation (FDEs) and, more generally, distributed order differential equations, open up new possibilities for robust modeling of such complex problems. In distributed order operators, the differential orders are distributed over a range of values rather than being just a fixed fraction as it is in standard/fractional ODEs/PDEs. In this work, we first provide an introduction to fractional calculus and distributed order models. Subsequently, we obtain the corresponding variational forms and set up the underlying function spaces and associated norms inorder to discretize the resulting weak form of the problem. Then, we develop a novel Petrov-Galerkin spectral method followed by the corresponding stability and error analyses. In anomalous physical processes, the distribution function can be obtained from experimental data, where the data uncertainty are incorporated through the distribution function, obtained from the observed data, hence, leading to a robust data-driven simulation framework for multi-physics problems.


Poster Number: ME-15

Title: Residual Limb Displacements within a Prosthetic Socket for Below Knee Amputees

Authors: Amy L. Lenz; Katie A. Johnson; Tamara Reid Bush

Abstract: In amputees, pressure ulcers are deep penetrating wounds that occur on the residual limb at the socket interface. Skin movement and residual limb loading coupled with deformation and strain on the skin plays a role on ulcer formation. Little work has been conducted to understand the circumferential and longitudinal displacements that occur on the residual limb during a walking activity. The current research assesses displacements occurring on the residual limb for individuals with a transtibial amputation while wearing a prosthetic. We hypothesize that larger localized displacements occur distally due to uneven limb motion within the socket. Four patients (4 M, age: 56.8 ± 9y) with unilateral transtibial amputation participated in the study (MSU IRB #14-089M). Data collection was accomplished by the development of a clear prosthetic for each patient, attaching 14 thin markers to the gel liner (which adheres to the residual limb) and tracking markers using a 12 camera Vicon system. Patient regions of high displacement vary greatly with different trends being observed. High displacement was considered as values above 3mm. Two patients demonstrate high displacement in the distal tibia region during stance whereas two other patients exhibited high displacement along the fibular head during stance. However, these regions of high displacement correspond respectively to patient reported regions of potential discomfort over prominent bony landmarks. Knowledge of this within socket displacement data will be valuable to clinicians for improving socket design and fit for the hope of reducing the occurrence of pressure ulcers.

This work was supported in part by None


Poster Number: ME-16

Title: Liquid Activated Textile Batteries for Wearable Biosensing Systems

Authors: Xiyuan Liu; Peter B. Lillehoj

Abstract: Wearable technology has become increasingly mainstream in recent years and offers great potential for many important applications including human health monitoring, environmental sensing and bio-agent detection. One of the main challenges with wearable sensors is the need for flexible, lightweight batteries that can be easily integrated with wearable materials and fabrics. In this paper, we demonstrate a unique liquid activated Ag-Al battery fabricated using textile. This battery is comprised of dry electrolyte layers sandwiched between Ag and Al electrodes. Upon application of an aqueous sample, the electrolyte layers become hydrated and generate an electrochemical reaction, thereby activating the battery. Using this scheme, we developed a single-cell battery that can produce an open-circuit voltage of 1.3 V. By connecting cells in series, higher voltages were achieved with minimal modification to the fabrication process. To demonstrate the functionality of this technology, we fabricated a dual-cell battery and used it to power a 1.6 V LED.

This work was supported in part by the National Science Foundation CAREER award (ECCS-135056)


Poster Number: ME-17

Title: Gas-Phase Synthesis of Gallium Nitride (GaN) Nanocrystals Using Non-Thermal Plasma

Authors: Rajib Mandal; Michael Bigelow; Branton Toback; Rebecca J. Anthony

Abstract: Bulk Gallium Nitride (GaN) is the standard light-emitting material, very efficient for Light Emitting Diodes (LEDs) and has been in use for many years. This material is very attractive due to its high-brightness and thermal stability. GaN is a direct band gap semiconductor with 3.4 eV band gap energy enabling its use in ultraviolet/blue light emission technologies. Relative non-toxicity of GaN compared to other popular semiconductors such as Cadmium selenide (CdSe) gives it distinct advantage. The applicability of GaN nanoparticles lies in high-brightness solid-state lighting devices. Here we present a study on synthesis of high-quality GaN nanocrystals using a fully gas-phase process. We have used a low-pressure nonthermal plasma reactor for the synthesis of GaN directly from gaseous precursors and deposited onto the glass substrate without any additional steps. The plasma reactor has some advantages over other available methods, namely, size monodispersity, easy control on nanocrystal size and the ability to deposit the NCs from the gas phase without removal from the reactor. It is also inexpensive and can be processed rapidly. Some studies have been performed with microwave plasma but it is very complex in nature. Thus, radiofrequency (RF) plasma could be an attractive alternative. An RF nonthermal plasma reactor is comprised of a borosilicate glass tube with dual ring electrodes encircling the tube externally. Vapor-phase precursors and carrier gases were flown through at relatively low pressure (typically 5-15 torr) with RF power ranging from 60W-100W. Our study has shown that small (3-5nm) crystalline GaN nanoparticles were produced.


Poster Number: ME-18

Title: Crystal Plasticity Finite Element Modeling of Multiphase Third Generation Advanced High Strength Steel (Q&P980) Undergoing Large Plastic Deformation

Authors: Bassam Mohammed; Taejoon Park; Farhang Pourboghrat

Abstract: Designing multiphase metals based on their constituent phase properties and using these metals for manufacturing automotive parts is a challenging process. A multiphase and multiscale model is strongly beneficial in achieving this goal since such a model can play an important role by connecting the material response at the macroscopic scale with the microstructural properties such as texture. In this study, a computationally efficient Crystal Plasticity Finite Element Model (CPFEM) was used to simulate the bulging and stamping of a three-phase (ferrite, martensite, and retained austenite) quenched and partitioned Q&P980 steel sheet, and to predict the corresponding crystallographic texture evolution. The CPFEM model was developed to capture the mechanical properties of steel phases based on their individual plastic deformation (hardening parameters), slip systems (BCC crystals for ferrite and martensite with 24 slip systems 12 {110} <111> and 12 {112}, FCC crystals for austenite with 12 slip systems {111} <110>), and the volume fraction of the constituents phases in the material. The macroscopic behavior of a polycrystalline aggregate is evaluated by a volume averaged response of the representative phases. The comparison between the multiphase CPFEM model, experimental and anisotropic Hill’48 yield criteria was applied for further validation of the proposed model. The results based on the multi-phase CPFEM model was in better agreement with the experimental results than phenomenological models in terms of capturing the applied punch force-displacement curve, strain distribution, and localized necking location relative to the sheet rolling direction.


Poster Number: ME-19

Title: On Confined Premixed Flames

Authors: Younis M. Najim; Norbert Mueller; Indrek S. Wichman

Abstract: The objective of this study is to perform an experimental and numerical investigation of a premixed flame propagating in a constant volume channel. The mixture was prepared from stoichiometric methane/air, and initially at rest, had temperature 296 K and pressure 102.65 kPa. In the experimental study, a high speed camera and a pressure sensor captured the evolution of flame structure and monitoring the pressure time-history during combustion. For two dimensional model, kinetic mechanisms is used to predict the reaction rate by direct integration using Reaction Design Scheme and by in situ adaptive tabulation. Since the kinetic mechanism is computationally expensive choice, the reaction progress variable is used for the three dimensional model in which the reaction rate is predicted using adaptive flame speed closure and algebraic flame surface density. The analysis of the numerical results revealed the effect of the interaction between flame front, pressure waves, and flame-induced flow on flame structure evolution. The numerical simulation conducted here for the three dimensional WDE channel uncovers an interesting behavior for the flame structure. We observed what may be defined as a “transverse tulip flame” which appeared in the direction perpendicular to the plane where the initial tulip flame is evolved after the latter underwent transition from cusped convex back to concave finger shape.

This work was supported in part by Higher Committee for Developing Education in Iraq.


Poster Number: ME-20

Title: Evolution of Solid Morphology Under Thermal Insult During Combustion

Authors: Thomas Pence; Indrek Wichman; Yen Nguyen

Abstract: Almost all flames interact strongly with nearby surfaces. For many materials, the surface undergoing pyrolysis (described as internal material thermochemical breakdown) and combustion is a growing, finite-thickness layer. This finite-thickness layer may exhibit voids, cracks, and other defects caused by heating and by material response to heating. One of these responses is extreme deformation. The surface layer structure of the degraded material varies enormously between materials. In addition, combustion alters surface morphology even as the surface morphology alters combustion. This occurs by (1) forming fissures that allow internal sample exposure to external heating; (2) allowing escaping volatiles into the gas to support further combustion; and (3) forming cracks and voids that weaken the material, making it susceptible to physical breakdown into smaller fragments consumed by the fire. Here, a model is examined that describes surface breakup for charring solids. This model incorporates stress development and stress relief as they occur during heating, pyrolysis and combustion.

This work was supported in part by CVRC


Poster Number: ME-21

Title: An Experimental Methodology for Creating Arbitrary Velocity Profiles in a Flow Facility

Authors: Alireza Safaripour; David Olson; Ahmed Naguib; Manoochehr Koochesfahani


This work was supported in part by AFOSR


Poster Number: ME-22

Title: Windkessel Approach for Venous Ulcer Risk Assessment

Authors: Wu Pan; Seungik Baek; Tamara Reid Bush

Abstract: Venous ulcers are sever skin wounds that affect 2.5 million people in US, however, no preventive measures had been developed to effectively identify and treat ulcer-prone population. Experimental work has shown that blood flow response to locally induced external load varied between patients with venous ulcer and healthy population. Therefore, the goal of this study was to develop a model that represents the blood flow response to external loading and to compare model parameters across samples of healthy individuals and those with venous ulcers. A Windkessel based circuit model was proposed where the resistor and capacitor represented vascular resistance and vessel compliance respectively. The model parameters were then iterated and optimized to match with the experimental data, and the values of the parameters for each participant were obtained and analyzed statistically. Significant differences were found in localized vessel resistance and compliance between ulcerated legs and healthy legs. The model demonstrated its capability for identifying patient-specific parameters. The values of the parameters and the statistical trend between different populations suggested that there are thresholds for these values which could be identified and utilized to determine when a person moves to a high risk category for wound formation. Then, a “just-in-time” prevention strategy would be applied. Future work will be increase the sample size so that ranges of model parameter values can be established and identified for patient sub-groups. The modeling of skin blood flow will help us better understand the blood perfusion behavior pre- and post- venous ulcers.


Poster Number: ME-23

Title: Material Characterization of Soft Tissue in Human Buttocks and Thigh Regions

Authors: Wu Pan; Joshua P. Drost; Zachary Sadler; Tamara Reid Bush

Abstract: Accurate characterization of material properties of soft tissue is imperative to model human body and design devices that interact properly with people. The mechanical properties of buttocks and thighs are particularly important in medical seating (wheelchair design), automotive seating and prosthetic design for the lower limb. However, there is no data defining the properties of the multi-layer tissue system in vivo, especially for the buttocks and thigh regions which are sensitive and difficult to test. Hence, the goal of this study was to determine the in vivo material properties of the tissue in human buttocks and thigh regions. A special designed chair was built to allow the test to be performed in a seated position, which is critical to guarantee same tissue tension in seating applications. External force was applied via a load-cell implemented indenter to six different regions along the buttocks and thigh. The deformation of the soft tissue was measured by a 3D motion capture system. Based on the obtained force-deformation relationship, Mooney-Rivlin material model was adopted to characterize the material properties in the thigh region and the material parameters were reported. This material model was successfully able to characterize the soft tissue in thigh region. The material parameters are a must to improve models of the human body and have great potential in medical seating and automotive occupant safety and prosthetic design. Future work will include analysis of buttocks regions and tests of more diversed group with different age and Body Mass Index.


Poster Number: ME-24

Title: Influence of Unsteady Effects on the Torque Generation of Rotors with Curved Channels

Authors: Raul Quispe-Abad; Norbert Mueller

Abstract: The Wave Disc Engine (WDE) is a novel idea among the wave rotor technology. This new engine concept is a radial rotor in which the typical processes of an Internal Combustion Engine (Compression, Combustion, and Expansion) are realized. For the torque production, the unsteady expansion process of outflowing combusted gases is used. The analysis of torque generation for conventional turbines typically considers only steady effects. For the unsteady expansion process, an unsteady component is added to this conventional analysis. This research is focused on the influence of the unsteady effect component on the torque generation. Using Computational Fluid Dynamic analysis and the integral form of the Angular Momentum equation, the contributions of steady and unsteady effects are quantified and presented. Two findings will be pointed out: the percentage of the contributions respect to the full amount of torque generated and some potential options to improve these contributions based on the channel geometry of the rotor. The outcome of this research contributes to maximizing the work extraction of Wave Disc Engine technology. This will be a step forward to a near future of the fabrication of the WDE for portable and residential scale power generation.


Poster Number: ME-25

Title: Multi-Objective Optimization Using Variable-Length Genomes

Authors: Matt Ryerkerk; Ron Averill; Kalyanmoy Deb; Erik Goodman

Abstract: Many optimization problems use solutions that consist of a number of analogous components. Examples include sensor coverage, wind farm, and laminate stacking problems. Using standard algorithms to solve these problems requires assuming a fixed number of sensors, turbines, or plies. However, if the optimal number is not known a priori this will lead to a sub-optimal solution. A better method is to allow the number of components to vary. As the number of components varies so does the dimensionality of the search space, making the use of gradient-based methods difficult. Genetic Algorithms (GAs) that utilize a variable-length genome to represent the solution are viable candidates. The traditional GA operators, designed to work with fixed-length genomes, are no longer valid. Notably, the recombination operators need to be modified such that they can operate on parents of varying lengths in a respectful manner. Additionally, it is observed that adding or removing components from a solution tends to become a destructive process as the algorithm progresses. In order to reach the optimal number of components the selection operator must be modified to include length-based niching of solutions. When considering multi-objective problems variable-length algorithms are required in order to fully characterize the Pareto front.

This work was supported in part by Funded through BEACON center (NSF grant DBI-0939454)


Poster Number: ME-26

Title: Efficient Spectral Methods for Anomalous Transport

Authors: Mehdi Samiee; Mohsen Zayernouri

Abstract: Anomalous transport refers to nonequilibrium thermo-fluid processes that cannot be described and predicted by the old fashion methods of mathematical and statistical physics. Non-local models and fractional PDEs (FPDEs) provide a proper modeling framework, in which interesting applications of anomalous transport are investigated. FPDEs are emerging as a powerful tool for modeling multiscale phenomena including overlapping microscopic/macroscopic scales, and long-range time memory or spatial interactions. Such phenomena occur in reacting turbulent flows, rheology, anomalous transport phenomena in porous/disordered materials, complex fluids, and multi-phase applications. We develop a novel Petrov-Galerkin (PG) spectral method for a unified class of linear FPDEs subject to homogeneous Dirichlet initial/boundary conditions. Our general mathematical model seemlessly includes time- and space-fractional diffusion, advection, advection-dispersion, and wave equations. We prove the stability of oue scheme, furthermore, we formulate a fast linear solver for the resulting high-dimensional system, hence, lowering the computational complexity of the problem. to demonstrate the efficiency and spectral rate of convergence in our PG spectral method, we perform several computational test cases.


Poster Number: ME-27

Title: A Closed-Loop Circulatory Model Coupled with Cellular-Based LV Electromechanics: Assessment of Ventricular Mechanics Under Different Loading Conditions

Authors: Sheikh Mohammad Shavik; Lik Chuan Lee

Abstract: Cellular-based high-resolution LV finite element (FE) models are useful in understanding ventricular mechanics associated with normal and abnormal heart functions. These models are usually constructed using active contraction models that are developed based on experimental data at the cellular level. As such, it is unknown if these models are able to reproduce physiological behavior at the organ level. The goal of this work is to assess some fundamental physiological behaviors under different loading conditions in a cellular-based LV FE model that is coupled with a closed loop circulatory model. Specifically, we analyze the model predictions of (1) pressure-volume (PV) loops, (2) myocardial oxygen consumption (mVO2) and mechanical work relationship and (3) three-dimensional strain under different loading conditions. By consolidating the PV loops under different loading conditions, we show that the model is able to reproduce a linear end-systolic pressure-volume relationship (ESPVR) and a curvilinear end-diastolic pressure-volume relationship (EDPVR). Furthermore, by assuming that mVO2 is proportional to the hydrolysis of ATP (Adenosine triphosphate) required to uncouple the actin-myosin bond, we show that the model is able to reproduce a linear relationship between mVO2 and the pressure-volume area. These findings are consistent with the physiological measurements in the LV across many species. Finally, our model predicts that the change of loading conditions affects the longitudinal, circumferential and radial strain – time behavior under different loading conditions. However, interestingly, the peak longitudinal strain is insensitive to the loading conditions. In future the circulatory model will be extended by incorporating the effect of LA and eventually including the RV to establish a bi-ventricular closed-loop circulatory model.


Poster Number: ME-28

Title: Shock-Turbulence Interactions in High Speed Multi-Fluid Flows

Authors: Yifeng Tian; Farhad Jaberi; Zhaorui Li; Daniel Livescu

Abstract: High-order numerical simulations of an isotropic multi-fluid turbulence interacting with a planar shock wave are performed using a hybrid numerical method, which combines a monotonicity-preserving scheme with a compact scheme. The main objective of this study is to investigate the effects of density variations due to compositional changes on the shock-turbulence interactions and mixing in very high speed flows. Convergence tests are conducted to establish the accuracy of results using different meshes with a wide range of grid sizes inside and outside the shock zone. The computed statistics are found to be independent of the grid when the turbulence after the shock is well resolved and the scale separation between numerical shock thickness and turbulent scales is adequate. A simulation of single-fluid turbulence is also conducted with similar conditions. Compared to the single-fluid case, turbulence amplification by the normal shock wave is found to be much higher and the reduction in turbulent length scales is much more significant in the presence of density variations due to compositional changes. Turbulent mixing enhancement by the shock wave is also more important in the multi-fluid case. The mechanisms behind multi-fluid shock-turbulence interaction and scalar mixing are identified by analyzing the transport equations for the Reynolds stress, vorticity and scalar variance.

This work was supported in part by Los Alamos National Laboratory


Poster Number: ME-29

Title: IC Engines for a Low CO2 World

Authors: Sedigheh Tolou; Harold Schock; Matthew Brusstar; Thomas Veling; Ray Kondel; Greg Davis

Abstract: The importance of geopolitical and environmental issues associated with energy consumption are inevitable in today’s global society. Power plants used in transportation must economically provide good power density, low exhaust emissions, and high fuel efficiency over a wide range of operation. Achieving one of these benefits does not guarantee optimum gain on others and in fact they are often contravening goals. The highly successful conventional spark ignition (SI) systems have limitations in efficiency potential due to the need for their stoichiometric operation. A viable option is lean burn engines which can provide significant improvements in fuel efficiency. The intent of this research is evaluating two of the most promising lean burn engine designs; Direct Injection Stratified Charge (DISC) and Dual-Mode Turbulent Jet Ignition (DM-TJI). The goal behind the study is comparing the DISC and DM-TJI engines based on engines’ performance, exhaust emission characterization, combustion stability analysis and cycle-to-cycle variation, cost and manufacturability of the engines and their emission control system, deposit formation and the influence on long term performance, and multi-fuel tolerance (especially low carbon biofuels). Analysis involves both the experiments and numerical simulations. A single-cylinder metal engine is available at US Environmental Protection Agency (EPA) National Vehicle & Fuel Emissions Laboratory (NVFEL); cell#10. The engine has the Ford Eco-Boost 1.6L head. A single-cylinder metal engine will be constructed by MSU/EPA and be moved to one of EPA test cells. Complimentary optical engine work is being conducted at MSU. Additionally, a 1D GT-POWER model and 3D simulations will be used to simulate both DISC and DM-TJI engines as required.

This work was supported in part by US Environmental Protection Agency, Tenneco Inc., Michigan Economic Development Corporation, General Motors


Poster Number: ME-30

Title: High Fidelity Numerical Study of Turbulent Jet Ignition and Combustion in Advanced Combustion Engines

Authors: AbdoulAhad Validi; Farhad Jaberi

Abstract: Turbulent Jet Ignition (TJI) is an efficient and novel method for initiating and controlling combustion in advanced combustion systems, e.g. internal combustion engines. It enables combustion in ultra-lean mixtures by utilizing hot product turbulent jets emerging from a pre-chamber combustor as the ignition source for the main combustion chamber. Here, we study the TJI-assisted ignition and combustion of lean methane-air mixtures in a Rapid Compression Machine (RCM) for various flow/combustion conditions with the hybrid large eddy simulation/filtered mass density function (LES/FMDF) computational model. In the LES/FMDF model, the filtered form of compressible Navier-Stokes equations are solved with a high-order finite difference scheme for the turbulent velocity, while the FMDF transport equation is solved with a Lagrangian stochastic method to obtain the scalar (species mass fraction and temperature) field. The LES/FMDF data are used to study the physics of TJI and combustion in RCM. The results show the very complex behavior of the flow and flame structure in the pre-chamber and RCM and the ability of TJI to drastically change the engine behavior.

This work was supported in part by National Science Foundation


Poster Number: ME-31

Title: Impact Testing and Analysis of Composite Fan Case Structures

Authors: Andy VanderKlok; Jim Dorer; Andy Stamm; Ryan Dutour; Eryi Hu; Xinran Xiao

Abstract: High speed fans play a vital role in the aeronautical field, specifically in aero-engines for producing thrust in hi-bypass turbo-jet engines commonly seen in most commercial airliners. One of the challenges in engine design is the possible fan blade out event (FBO), which can occur from bird strike or fatigue. When this happens a blade is released at ballistic speeds which can cause catastrophic damage to nearby aircraft components. to prevent such damage, a fan case is constructed to contain the loose blade. Traditionally the fan case is made from a well characterized metallic material. An alternative material is composites. The major advantages of composites are corrosive resistant, light weight, tailorable, and high strength. However, little is known about how composites respond to high speed impact in FBO. Assessing the dynamic response and failure modes of composites in FBO is crucial in the design of composite fan case. The proposed work is focused on the development of gas gun testing method that can closely replicate the impact and damage caused by FBO on composite fan cases. to investigate FBO, a spin pit testing facility has been developed. A new, energy based method for the design of projectiles in gas gun test to replicate FBO has been proposed. The proposed method will be examined by comparison of the gas gun test with spin pit FBO test. This work will contribute to the development of a combined numerical and experimental approach for the design of a composite fan case.

This work was supported in part by NASA Glenn Research Center


Poster Number: ME-32

Title: Investigation of the Chemo-Mechanical Coupling in Lithiation of Amorphous Si Using Finite Element Analysis

Authors: Miao Wang; Xinran Xiao

Abstract: Si experiences large deformations during battery cycling, which has been cited as a primary cause for the electrode fracture and thus capacity fading. The problems caused by the electrode deformations can be remedied through electrode design, and the numerical models will greatly facilitate this design process. Understanding the stress evolution and its effect on Si lithiation is of great importance for Si electrode design. In this study, the chemo-mechanical coupling of Si lithiation was investigated by simulating two geometries, amorphous thin film and nanospheres, using finite element analysis. An asymmetric rate response between lithiation and delithiation has been observed in a-Si thin films, but not in Si nanospheres. The simulations reveal that, in a-Si thin film, the lithiation induced stress field is almost uniform and opposite in sign during lithiation and delithiation. During lithiation, the stress is compressive which leads to stress suppressed diffusion. During delithiation, it is vice versa. The stress state is an important factor for the asymmetric rate response observed in a-Si thin film. In Si nanospheres, on the other hand, the stress varies and its sign changes across the particles during lithiation and delithiation. The stress suppressed and enhanced diffusion happens simultaneously in a particle. The asymmetric rate response, if any, was not obvious.


Poster Number: ME-33

Title: Image-Based Computational Modeling of the Ventricular Mechanics in Pulmonary Hypertension Patients

Authors: Ce Xi; Lik Chuan Lee

Abstract: Pulmonary hypertension (PH) is a disease resulting from restricted flow in the pulmonary arterial circulation. This results in an increase in pulmonary vascular resistance. Left untreated, the disease leads to right ventricular (RV) failure, which is the most common cause of death in PH patients.to the best of our knowledge, there are currently no in vivo studies that determine the impact of PH on the regional ventricular mechanics and mechanical properties in the human heart. In our study, We have used a cardiac electromechanics model to quantify the regional ventricular mechanics between a PH patient and a normal human subject. By adjusting a few parameters in the model, we are able to match the experimentally measured pressure-volume (PV) loops of RV and LV for the PH patient and the normal subject.The differences of all quantified variables and parameters (e.g., regional ventricular strain and passive stiffness) were compared between the PH patient and normal subject. Regional differences between quantities obtained from LV and RV were also compared. Regional ventricular wall stresses and strains were computed for the PH patient and normal subject. Our findings indicate that the diastolic passive stiffness and the contractility of the PH patient are larger than that of the normal subject in both the LV and RV.


Poster Number: ME-34

Title: Simulation of the 3 Omega Method for Measuring the Thermal Conductivity of Superconducting Niobium

Authors: Peng Xu; Neil Wright; Thomas Bieler; Chris Compton

Abstract: Enhancing performance of Superconducting Radio Frequency (SRF) particle accelerators relies on improvements in metallurgy and manufacturing of their Nb walls. Informing improvements in Nb manufacture is understanding its thermal conductivity at about 2 K, where many accelerators operate. Measuring the thermal conductivity of single crystals or thin films of Nb is challenging. The 3 omega technique has proved to be versatile for measuring the thermal conductivity of thin films and small samples, especially of semiconductors and dielectrics. Here, extension of the technique to superconductors is studied. Simulation of temperature-dependent thermal conductivity showed that the 3omega method can be applied to measure the thermal conductivity of Nb at low temperature. Finite element simulation of the 3omega method was used to parameterize the roles of heater geometry, anisotropic thermal conductivity of thin films, and temperature dependent thermal conductivity on the reported values of Nb. Variation in heater thickness resulted almost the same slope and axis intercept, indicative of thermal conductivity and diffusivity, respectively. Variation in the width of the heater resulted in the same slope, but different axis intercepts. Cross-plane thermal conductivity of thin film is always more important, however, in-plane thermal conductivity cannot be neglected. From the 3 omega simulation, the conditions (e.g. sample geometry, frequency range) can be predicted when conventional analytical formulas are suited to determine the thermal conductivity of bulk and thin film materials from the data measured.

This work was supported in part by This research is supported by the US DOE Cooperative Agreement through grant number DE­SC0000661 and Michigan State University.


Poster Number: ME-35

Title: Characterization of the Through Thickness Mechanical Property of Thin Polymer Films

Authors: Shutian Yan; Xinran Xiao

Abstract: Thin polymer films have a wide range of applications. The mechanical properties of the films play an important role to their performances and functions. For example, porous thin polymer films are often the choice of materials for separators in Li-ion batteries. The mechanical integrity of the separator is critical to the durability and the safety of the batteries. to improve the separator design, it is desired to know the stresses experienced by a separator in a battery under different charging-discharging cycles and in the events of impact. Such information can be estimated through numerical simulations using multiphysics models. The stress-strain response of the separator is a required input. The in-plane stress-strain behavior of a thin polymer film may be measured using common lab equipment. The characterization of the through thickness mechanical properties is far more challenging. The battery separator is usually 20~30 micron in thickness. to obtain the stress-strain behavior, the displacement measurement must have a submicron resolution. In this work, we investigate a capacitance based displacement measurement method for through thickness behavior characterization. Our preliminary study shows that this method can achieve a resolution of 0.5 micron. The measured nominal compressive modulus is comparable to the reported values in literature. As shown in previous studies, both the in-plane and the through thickness mechanical properties are lower when tested in electrolyte solutions. In this work, to test samples in solutions, a special testing fixture with capacitance based displacement measurement has been designed to characterize the through thickness mechanical behavior.


Poster Number: ME-36

Title: Swelling Induced Burst in Hyperelastic Spheres and Cylinders

Authors: Vahid Zamani; Thomas J. Pence

Abstract: The conventional theory of hyperelasticity is generalized to incorporate a swelling effect. We consider the spherical symmetric deformation in which swelling has been taken into account and then the theory is used to study an abrupt inflation instability in pressurized shells. At fixed pressure, a slowly changing amount of swelling can suddenly cause a finite jump in the shell radius.

This work was supported in part by This work is made possible by NPRP grant from the Qatar National Research Fund (a member of Qatar Foundation).


Poster Number: ME-38

Title: Physics-Based Turbine Power Models for a Variable Geometry Turbocharger

Authors: Tao Zeng; Devesh Upadhyay; Harold Sun; Guoming G. Zhu

Abstract: Control-oriented models for Variable Geometry Turbochargers (VGT) typically calculate the turbine power based on isentropic assumptions with a fixed or a map based value for the turbocharger mechanical efficiency. While the fixed efficiency assumption is an obvious over simplification, the map based approach, on the other hand, may not be globally accurate due to the need for interpolation between varying vane positions and extrapolation when the turbocharger is operating outside the mapped region. In this paper physics-based models of turbine power as well as the power loss are developed, utilizing the VGT vane position and the shaft speed. This makes it possible to define the mechanical efficiency as a function of the vane position thereby eliminating the above mentioned uncertainties as well as allowing a smooth extension over the entire operating range. The proposed model is validated against a few sets of test data from both steady state and transient operations.

This work was supported in part by Ford Motor Company


Poster Number: ME-39

Title: Soft Lithographic Printing and Transfer of Photosensitive Polymers: Facile Fabrication of Free-Standing Structures and Patterning Fragile and Unconventional Substrates

Authors: Yaozhong Zhang; Jea-Hyeoung Han; Likun Zhu; Mark A. Shannon; Junghoon Yeom

Abstract: Dry film photoresists (DF PRs) are widely used to perform photolithography on nontraditional substrates such as printing circuit boards, plastic sheets, or non-planar surfaces. Commercially available DF PRs are usually in a negative tone and rather thick, limiting lithographic resolution and versatility. The relatively large pressure required for lamination also prevents the technology from being used for delicate substrates. Here we present a modified soft-lithographic process, namely photoresist blanket transfer (PR BT), transferring a spin-coated PR film from a flat elastomeric stamp to a substrate. The elastomeric stamp is highly compliant, bringing the PR film into intimate contact with the substrate and eliminating the need for a large lamination pressure. Photolithography on unconventional substrates such as etched, fragile, and porous ones is demonstrated. Single or multiple transfers of PRs by BT are utilized to fabricate multilayer, free-standing, and re-entrant polymeric microstructures. A fragile and porous substrate such as an anodized aluminum oxide membrane can also be patterned using PR BT. Moreover, a reliable method to create metal electrodes and high surface area catalysts inside microchannels is discussed for novel microfluidic applications.


Poster Number: ME-40

Title: Peridynamic Simulation of Crack Propagation in Orthotropic Materials

Authors: Wu Zhou; Dahsin Liu

Abstract: Peridynamic is a nonlocal numerical technique. With only two material constants, peridynamic has been employed for investigating damage processes of orthotropic fiber composite materials with Cf as stiffness along the fiber direction and Cm as that for all other directions. However, according to orthotropic theory, the stiffness of unidirectional fiber composites should change continuously with the fiber orientation. In this poster, bond-based peridynamic is modeled with a continuously changing modulus C for orthotropic materials. That is, C changes continuously from along the fiber direction to transverse to the fiber direction with a very similar manner as the change of Young’s modulus in a lamina from along the fiber direction to transverse to the fiber direction. Various ratios of Young’s modulus in fiber orientation to that in matrix orientation, i.e. E11/E22, ranging from 4 to 50, have been investigated. The modified peridynamic model is then employed to study the dynamic behavior of an orthotropic beam under three point bending. The beam has a length of 200mm and a thickness of 50mm. It has a fiber orientation of 135-degree with respect to the beam length. A notch is initially assigned at the bottom of the mid-span of the beam. Impact loading is added to the center of the top surface. Crack initiation and propagation are simulated by the continuous orthotropic peridynamic model. The numerical results match well with the experimental observations published in literature.


Poster Number: ME-41

Title: Evaluation of Thumb Carpometacarpal Joint Laxity and Muscle Strength in Osteoarthritis: A Pilot Study

Authors: A. R. Cussen; G. Shafer-Crane; E.E. Hornbach; T. R. Bush

Abstract: Osteoarthritis (OA) of the thumb carpometacarpal (CMC) joint affects approximately half of U.S. adults over 55, and often causes pain and functional deficits like difficulty opening jars and medications. In this pilot study, we explore the potential of three dimensional (3D) kinematics to identify factors associated with thumb CMC joint instability in OA. We hypothesize that 3D motion capture and force measures can be utilized to detect a) reduced muscle strength during abduction/adduction and flexion/extension motions, b) reduced active range of motion (AROM), and c) increased time to complete a task in OA subjects compared to asymptomatic controls. The results of this proof on concept study suggest that motion capture is sufficient to detect decreases in AROM and increased duration to complete a task in OA subjects. Force application differences were not identified in this sample. Future studies will utilize larger subject populations to determine statistical significant of these pilot findings.