ARC Collaborative Research Seminar Series
Winter 2014

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)


February 21, Friday (9:15 - 11:00am)
6501 E. 11 Mile Road, Warren, MI 48397-5000, USA.
Building 200B, TARDEC University

On Vehicle Systems Design and Optimization

1. A Model-Based Feasibility Study of an Adaptive Modular Vehicle Fleet with Full Mission Capability (link)

        We study the feasibility of deploying a vehicle fleet that can adapt to changing mission requirements through modular architectures. We model a fleet comprising vehicles with diverse functionality and specifications, assigned to complete a mission modeled stochastically. The modeling environment includes powertrain models to compute vehicle performance and fuel use on specific duty cycles, stochastic simulation of fleet operations, and transportation, operating, acquisition and retirement cost models. We assume that the technological feasibility of modularization is feasible. We examine and compare scenarios where missions are completed with conventional and modular fleets, and we estimate the corresponding costs and flexibility to mission adaptation in order to evaluate advantages and disadvantages of the two fleet forms.

2. Optimal Crowdsourcing Framework for Engineering Design (link)

        Preferences of stakeholders in the decisions made while designing a complex system are essentially evaluations of design performance attributes and can be elicited directly or extracted from collected data. A crowd-generated evaluation is not guaranteed to benefit from larger, more diverse crowds; rather, evaluator ability often determines the accuracy of extracted performance estimates. Our research is motivated by previous experiments with both simulated and human crowds that reveal the difficulty of statistically locating “high ability” in a crowd dominated by “low ability” evaluators. We present recent progress with two “feature learning” algorithms (parse coding and sparse restricted Boltzmann machines) that allow the use of design (and associated data) features to predict evaluator ability more effectively. Preliminary results show large increases in predictive accuracy on a market dataset (analogous to a crowdsourced evaluation) making future controlled crowdsourced evaluation experiments promising.


March 14, Friday (9:15 - 11:00am)
University of Michigan, UM North Campus, Phoenix Memorial Lab. 2000A

On Vehicle Structures and Human Modeling for Survivability

1. Improving Soldier Safety and Accommodation through Improved Vehicle Design
Matthew P. Reed, Jingwen Hu, Jonathan D. Rupp, University of Michigan
Thrust Area II

        We will give an overview of Soldier-centered research in Thrust Area 2. Data from the ARC Seated Soldier Study are being used to develop new driver and squad accommodation models for vehicle layout. A study is underway to quantify the effects of body armor and body borne gear on seated reach difficulty and capability. We are extending the Seated Soldier Study to focus on unusual driver workstation configurations, and a new seat measurement tool for squad conditions is being developed. These projects are being conducted in close collaboration with TARDEC to ensure that the results can be immediately implemented. The presentation will conclude with a broader view of work at U-M to address Soldier safety and accommodation.

2. Light weight vehicle structures that absorb and direct destructive energy away from the occupant (link)
Presenter: Weiran Jiang, GSRA University of Michigan
Quad Members: Dr. Syed Mohammad (M-ATV (Army), SFAE-CS-MRAP/MS 298); Dr. Ravi Thyagarajan (US Army RDECOM-TARDEC); Dr. Nam Purush (Modeling & Simulation, BAE Systems); Nick Vlahopoulos (UM).

        Pursuing occupant centric vehicle structures that provide safety from explosive threats while at the same time make the operation of the vehicle comfortable and safe for the soldiers, comprises one of the main thrusts in the Army S&T activities. In this research simulation algorithms for developing vehicle concepts that direct the blast energy which enters the vehicle structure away from the occupant space are pursued. Such vehicle architectures will be combined with energy absorbing material and user/vehicle interfaces that maximize the isolation between the vehicle surfaces and the occupant in the case of an explosion. It has been previously demonstrated that delaying and controlling the contact between a vehicle and an occupant reduces the loads developed in the occupant’s members. The characterization of properties associated with a “softer” steel material that can be used for absorbing the destructive energy through deformation is considered. This “softer” steel material will be able to reduce the shock loads which are transferred in the interior of the vehicle by absorbing energy through higher deformation compared to high strength steel. Utilization of anisotropic material properties will be investigated for guiding the destructive energy into sacrificial parts of the vehicle structure made out of “soft” steel and away from the occupants. A summary of the research which has been conducted during the first year of this project will be presented. First the use of anisotropic material and energy dissipation concepts were investigated for a flat plate. Based on the lessons learned from such a simple configuration a model of the TARDEC V-Hull structure was used for increasing the occupant survivability by reducing the dynamic response index for injury. The direction of the research after completion of this initial effort will also be discussed.


March 28, Friday (9:15 - 11:00am)
University of Michigan, UM North Campus, Phoenix Memorial Lab. 2000A

On Hybrid Electric Vehicle Systems

Projects presenting:
1. Benefit Analysis Via Simulation and Experiments on Range Extension of a Hybridized 12 volt lead-Acid Battery Using double Layer Ultracapacitors (project link)

        We present some recent results on hybridization of a 12 volt lead-acid battery using double layer ultracapacitors. The goal is to investigate the circumstances under which the integration of the ultracapacitors will result in improved range. In the hybrid topology the battery and ultracapacitor are connected in parallel while the battery is decoupled from the load and the ultracapacitor is responsible for providing the power during traction and capturing the regenerative energy during braking. The battery functions as the on-board charger for the ultracapacitor module. The custom built full-bridge DC/DC converter is located between the battery and ultracapacitor in order to regulate the charging current to the ultracapacitor according to the control strategy. Simulations and hardware in the loop experiments using a city driving cycle are used to evaluate the HES performance.

2. Optimization of Series Hybrid Vehicle Powertrain Considering Auxiliary Cooling Loads, Battery Health, and Electric Motor Design (project link)
Xueyu Zhang, Andrej Ivanco (presenter), and Zoran Filipi, Clemson University

        A unified, multi-physics hybrid electric vehicle simulation tool for the next-generation military medium trucks has been enhanced with the addition of the finite element electric machine model, and integration of the high-fidelity battery cooling model. The new capabilities are a result of cross-cutting collaborations with the ARC members, and they enabled progress towards achieving our main objectives, namely: (i) development of a framework for multi-variable, multi-objective optimization of the Vehicle Power System (VPS), (ii) optimization of the component design for system level goals.
        We will highlight development of the VPS control strategy using Dynamic Programming, capable of optimizing power flow with consideration of both propulsion power and parasitic power consumption from cooling auxiliaries. Extraction of implementable rules includes scheduling of threshold power based on battery temperature and instantaneous SOC values. On-going effort is focused on integration of the battery aging model into the VPS, and the impact of battery fading on the fuel consumption.
        In-wheel electric motors are considered in this application because their duty cycles differ significantly from commercial HEV or Integrated Starter Generator (ISG) configurations. Optimization framework is setup by coupling the vehicle simulation with the constrained optimization solver, and results provide clear guidance for e-motor design, as well as assessment of efficiency benefits and reduction of cooling requirement.


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.