ARC Collaborative Research Seminar Series
Fall 2013

ARC seminars are free and open to the general public. Center members can download the presentation files on our password-access online portal iARC. Non-ARC members please email arc-event-inquiries@umich.edu with your requests.

If you wish to attend the seminar remotely, please contact William Lim (williamlim@umich.edu) for teleconference details. For parking information, contact Kathie Wolney (kathian@umich.edu).

Refreshments will be served 9:15-9:30am. The talks will begin at 9:30am sharp.
Please note new venue: Phoenix Memorial Laboratory, room 2000A (ground floor)


September 25, Wednesday (9:15 - 11:00am)
University of Michigan, UM North Campus, Phoenix Memorial Lab. 2000A

On Heavy Duty Vehicle Fuels & Combustion - Part 1

Reaction Pathway and Elementary Ignition Behavior of Surrogates for JP-8 and Alternative JP-8 Fuels
Dongil Kang, Jason Martz, Angela Violi, and André Boehman (PI), University of Michigan

        The objectives of this project are to provide validation experiments to compare the elementary ignition behavior, focusing on the chemical portion of the ignition delay, of the surrogate fuel formulations proposed by Violi and co-workers in comparison with practical JP-8 fuels. We will provide ignition data from a motored CFR engine apparatus for comparison to chemical kinetic modeling calculations. In this manner, there will be detailed experimental comparison of ignition behavior (critical compression ratio, critical equivalence ratio, % low temperature heat release) between the surrogate fuels and practical JP-8 fuels. The motored CFR engine has a compression ratio (CR) range from 4 to 14.5. This engine is equipped with a fuelling system that contains a heated intake manifold and a GDI fuel injector located upstream of the intake system to provide a premixed fuel-air mixture to the engine cylinder.

Validation of JP-8 Surrogates in an Optical Engine
PI: Prof. Marcis Jansons, Wayne State University

        The use of simulation-led design processes is a desirable means of reducing the time and cost of military engine development. An impediment limiting the fidelity of numerical engine combustion simulations is the uncertainty about the use of pure-compound surrogate mixtures to represent wide-ranging ignition and combustion characteristics of JP-8 fuels in the field. The behavior of iso-paraffinic kerosene (IPK) and a surrogate are examined over a range of charge densities and temperature in an optical engine. A means of surrogate validation under varying pressure-temperature conditions is developed by including excited-state species in a chemical reaction mechanism to simulate chemiluminescent light emission over the course of a combustion cycle in a 3-D CFD simulation. Experimental comparisons are made against synthetic images of HCHO*, OH* and CO2*, respectively marker species for low temperature combustion, ignition, and high temperature heat release, generated from the surrogate-mechanism-CFD model.


October 9, Wednesday (9:15 - 11:00am)
University of Michigan, UM North Campus, Phoenix Memorial Lab. 2000A

On Heavy Duty Vehicle Fuels & Combustion - Part 2

Combustion Behavior of JP8 fuels in Heavy-Duty Diesel Engine
Doohyun Kim, Jason Martz and Angela Violi, University of Michigan

        The U.S. Department of Defense directive 4140 mandated the aviation fuel JP-8 to be the universal fuel for all military applications. This logistical simplification has been field-tested and JP-8 has been successfully used in the U.S. Army’s compression ignition (CI) engines. However, the wide variations in cetane index, relative amount of classes of compounds and additives have shown significant impact on the CI engine operation. The purpose of this work is to gain insights on the fundamental processes that influence the combustion behaviors of various JP-8 fuels with the final goal to develop a predictive model that can be used to link fuel characteristics to combustion behaviors. Due to the complexity and variability of full-boiling range fuels, we have developed simplified blends, called surrogates, to study the combustion processes. The new surrogates for JP-8 and alternative JP-8s (IPK and S-8) are shown to successfully emulate the various liquid fuel properties that affect the diesel ignition process. The chemical matching of the surrogate with real JP-8 fuels was performed in shock tube reactors and rapid compression machines. In addition, computational fluid dynamics simulations were performed to validate the new surrogates with JP-8 spray data from Sandia National Laboratory’s constant volume chamber. The non-reacting set up was used to assess the effects of fuel liquid properties on spray characteristics. The result indicated that density and volatility of the fuel are keys to liquid penetration length, while viscosity and surface tension has little effect.

Simulation and Control of Combustion in Military Diesel Engines
N.A. Henein, A. Shrestha, Z. Zheng , U. Joshi, Wayne State University

        The overall goal of this project is to develop strategies, based on computer simulations, to control diesel engines to operate properly on different military fuels without loss in power and fuel economy, or increase in soot emissions. Current diesel engine cycle computer simulations use a combustion mechanism developed for pure n-heptane, which does not represent the distillate JP-8 fuel used in ground vehicles. This presentation describes the research conducted at WSU to develop a surrogate and to compare its autoignition and combustion characteristics with those of JP-8. This comparison is made in the constant volume combustor of the IQT (Ignition Quality Tester) and a in a high speed direct injection, single cylinder diesel engine. The activation energies for the global autoignition reactions of JP-8 and its surrogate were compared. Future research will include the use of the mechanismof the chosen surrogate in the diesel cycle simulation and a comparison of the computed results with experimental data obtained in tests on a military diesel engine fuelled with JP-8.


November 13, Wednesday (9:15 - 11:00am)
University of Michigan, UM North Campus, Phoenix Memorial Lab. 2000A

On High Performance Structures and Materials Modeling

Eliminating Hysteresis Losses in Tank Track Pads Using Meta-materials
Dr. Georges Fadel, Nitheesha Chennadi, Sampath V. Dangeti, Clemson University

        The highly dynamic nature of road wheel to track interaction and the inherent hysteretic property of elastomeric materials, cause dramatic temperature increase in the tank track pads. The search for an ideal material which provides high compliance as well as low energy loss is on-going. We propose alternatively to consider the design of specifically engineered meta-materials to address this problem and improve the lifetime of the tank track pad.
        The proposed method involves a two level optimization process for the design of the track pad. The top level optimization identifies target properties achievable using currently used elastomeric or similar materials. With these target properties, a meta-material topology optimization will, at the lower level, attempt to modify the topology of an elastic material to achieve the target properties identified in the top level while eliminating hysteretic losses.
        The presentation will focus on the FEA of a tank track pad to determine the specific target properties needed for the meta-material design and topology optimization method which is concurrently being adapted.

Sampling-based RBDO using Probabilistic Sensitivity Analysis and Virtual Support Vector Machine
Hyeongjin Song and Dr. K.K. Choi, University of Iowa
Dr. David Lamb and Dr. David Gorsich, U.S. Army TARDEC

        In this seminar, a sampling-based RBDO method using a classification method is presented. The score function is used to compute design sensitivities of probabilistic constraints. Since the probabilistic sensitivity analysis requires only the limit state function and not the response surface, an efficient classification method can be used for a sampling-based RBDO. The proposed virtual support vector machine (VSVM), which is a classification method, is a support vector machine (SVM) with virtual samples. By introducing virtual samples, VSVM overcomes the deficiency in existing SVM that uses only classification information as their input. In this paper, the universal Kriging method is used to obtain locations of virtual samples to improve the accuracy of the limit state function for highly nonlinear problems. A sequential sampling strategy effectively inserts new samples near the limit state function. Since SVM has simpler formulation than Kriging, computational cost for evaluating a large number of MCS samples can be significantly reduced.


December 18, Wednesday (9:15 - 11:00am)
University of Michigan, UM North Campus, Phoenix Memorial Lab. 2000A

On Advanced Component and Systems Modeling

Structural Modeling of Battery Packs using Parametric Reduced Order Models
Jau-Ching Lu, Dr. Kiran DSouza, Dr. Bogdan Epureanu, University of Michigan
Dr. Matthew Castanier, Dr. David Lamb, U.S. Army TARDEC

        In this seminar, a method for efficient numerical modeling of hybrid electric vehicle battery packs to enable probabilistic forced response (vibration) simulations that account for the effects of structural variations is presented. Two types of variations are considered in this study. The first is the level of prestress when joining the cells within a pack where this prestress level can be considered as a design variable. The second type of variations considered in this study is small, random structural discrepancies between the cells in the battery pack. These variations occur due to differences in the state of charge, temperature, manufacturing and other uncertainties. These variations are random, and can have a large impact on the response of the battery pack due to a phenomenon known as localization. Thus, these variations must be investigated in a probabilistic manner. The proposed efficient method for conducting probabilistic analyses is demonstrated for the case of an academic battery pack. Preliminary results for a more complex battery pack will be discussed also.

Reliability, Maintenance and Optimal Operation of Repairable Systems with Application to a Smart Charging Microgrid
Annette G. Skowronska (presenter & PhD student from TARDEC), Dr. Vijitashwa Pandey, Dr. Zissimos P. Mourelatos (PI), Oakland Univerisity
Dr. David Gorsich, Dr. Matt Castanier, U.S. Army TARDEC

        The definition of reliability may not be readily applicable for repairable systems. For the optimal design and maintenance of a repairable system, we use multiple metrics to define performance under uncertainty and obtain optimal tradeoffs among a minimal set of metrics. Critical installations such as a remote microgrid powering a military installation require a careful consideration of cost and repair strategies to address logistical challenges in performing repairs and supplying necessary spare parts. This presentation shows how a minimal set of metrics enhances decision making in such a scenario. It enables optimal tradeoffs between critical attributes, while guaranteeing that all important performance measures are satisfied. Our optimization problem considers factors such as cost to operate, maintenance, reliability, repair time and logistics. We demonstrate the value of the proposed approach using a US Army smart-charging microgrid installation and present preliminary results for vehicle to grid operation where a vehicle is a mobile power generation system.


ARC members can download the presentation files on our password-access online portal iARC.
Non-ARC members please email arcweb-info@umich.edu with your requests.