ARC Researchers at ASME 2015 IMECE
International Mechanical Engineering Congress & Exposition
(November 13-19, 2015 at Houston, TX)

ARC researchers (principal investigators in bold) will be presenting their latest research developments. Below are their papers which may include non-ARC funded research (paper titles in bold are ARC funded).

Session: 7-9-3 Minisymposium: Multiphysics Coupling in Energy Storage - 3
Thursday, November 19, 2015 09:45 AM-11:15 AM
Invited Presentation
On the Accuracy and Estimation of Battery Behavior
Anna Stefanopoulou, Univ Of Michigan
Abstract: Many multi-scale physics-based models and their reduced-order renditions are used to estimate the lithium concentration and over-potential distributions throughout the battery electrode. This estimation can help determine power availability while avoiding locally damaging phenomena. We show how to evaluate the accuracy of various electrochemical battery models by using neutron imaging as in situ measurement of the lithium concentration along the anode and cathode electrode layers. The neutron imaging was performed at the National Institute for Standards and Technology (NIST) Center for Neutron Research. The collimated neutron beam originates from a 20MW reactor, which provides the high flux source of neutrons. The transmission image is captured by a micro-channel plate (MCP) neutron counting detector with 5 µm pixel pitch and 13.5 µm spatial resolution. The observations from an operating Lithium Iron Phosphate (LFP) pouch cell battery with typical commercial electrodes are used to define the limits of the validity of the electrode-averaged, also called as single particle, models used in observers. Even with validated electrochemical models and state of the art Kalman filtering techniques, state of charge (SOC) estimation still relies on the monotonicity of the open circuit voltage (OCV) as a function of SOC. For some chemistries and ranges of SOC the OCV-SOC relationship is very flat and renders SOC unobservable from voltage measurement. This talk will demonstrate the benefits of the fusion of bulk force and battery voltage measurements to estimate the average SOC in Li-ion battery cells. The force measured at the end plates of a constrained battery cell corresponds to the composite swelling and contraction of the anode and cathode layers during charging due to the Lithium intercalation. To characterize the relationship among bulk stress, battery state-of-charge (SOC) and operating temperature, carefully designed experiments were conducted for Lithium-Nickel-Manganese-Cobalt-Oxide/Graphite prismatic cells. Specifically, the bulk force as a function of SOC is first measured by using load sensors during low C-rate experiments. The bulk force is also measured during high current pulsing that causes thermal expansion in response to the elevated temperatures due to the high Joule’s heating. A thermo-mechanical battery model is then constructed and used to predict the bulk force during swelling from Li-intercalation and thermal expansion. The predicted force and measured force can form an error that augments the voltage error and leads to better SOC estimation during the SOC range where the force-SOC relation is not flat. The SOC region for which the force-based SOC estimation is beneficial depends on the cell chemistry and the phase-change of the electrode material during Li-intercalation.
Session: 4-8-2 Fluid-Structure Interaction II
Tuesday, November 17, 2015 15:30 PM-17:00 PM
IMECE2015-53105 Forecasting Subcritical and Supercritical Flutter using Gust Responses
Amin Ghadami, Bogdan I. Epureanu, University of Michigan, Ann Arbor
Abstract: Subcritical and superctitical flutter (Hopf bifurcations) have been observed in a variety of fluid-structural systems. Such phenomena lead to various types of stability issues and can cause dramatic changes in the dynamics. Therefore, forecasting such bifurcations, i.e. predicting bifurcations with measurements only from the pre-bifurcation regime is a significant challenge and an important need. This is especially important for complex large-dimensional systems when an accurate model of the system is not easily available or when the system properties/parameters are unknown or they change.
         This work presents a novel method to forecast flutter bifurcations (as well as the bifurcation diagram) based on observations of the gust response of the system only in the pre-bifurcation regime. The approach is mainly based on the phenomenon of critical slowing down (CSD) which accompanies such bifurcations. Although the CSD approach has been used already, those previous studies could only use very small perturbations. That represents a significant drawback which prohibits forecasting the entire bifurcation diagram. The method we propose is not limited to small perturbations and is not based on linearizations. Hence, it can use recoveries from large perturbations for bifurcation forecasting. Using larger perturbations has two main advantages. First, it enables better ways to alleviate issues created by measurement and process noise which can more detrimentally affect small perturbations and hence reduce the accuracy of predictions based on small perturbations. Second, the ability to use larger perturbation results in predictions of larger ranges of the bifurcation diagram which in turn provides a broader understanding of the system behavior in its post-bifurcation regime.
         The method does not require a model of the system. The forecast is based only on time series measurements of the way system response to gusts. To demonstrate the method and highlight its advantages, we have used surrogate time series obtained from a simulation of a nonlinear aeroelastic system which can experience different operating conditions and bifurcations. While the system has only two mechanical degrees of freedom, the aerodynamic part of the model used leads to a more complex 8-dimensional nonlinear aeroelastic system.
         Numerical simulations show that the method accurately predicts the bifurcation point and also the post-bifurcation regime in both supercritical and subcritical cases despite the fact that it uses only pre-bifurcation regime data and it does not use a model of the system. Since dramatic changes can occur in the system dynamics at bifurcations, predicting the bifurcation type (i.e., its supercritical or subcritical characteristics) without placing the system in the post-bifurcation regime is a great advantage. This type of forecasting opens the door to a variety of applications where knowledge of nearby bifurcations is important for safety and maximum system performance.