Current Projects

Overview

Batteries are increasingly relied upon to provide multiple services during applications (e.g. traction in an electric vehicle, vehicle-to-grid, ancillary services) and to act as the ultimate resiliency element (e.g. electric vehicles used as power units during Hurricane Sandy). However, the ability to perform these diverse services is compromised by battery aging phenomena that eventually lead to failure. Understanding of how service conditions and context affect battery aging is limited due to a) battery high context dependency on generation and load dynamics, and environmental conditions; b) the multi-scale cell and module nature of battery systems; and c) the fact that a battery itself varies with age, as batteries are repurposed after a first life (e.g. electric vehicle) into a second life (e.g. grid or residential).

This CAREER project aims to understand battery aging dynamics as context-dependent, and to provide a unified theory that links application-level events and conditions with cell- and module-level aging events. The Pl hypothesizes that a battery electrochemical nature and aging, multi-scale system, observability challenges, and its context-dependency can all be modeled using ecological tools, with ecology defined as a branch of biology that explores organism relationships to one another and to their environment. Therefore, methods proven useful to study ecological relationships are well suited to study battery life, and can provide new knowledge, testing and estimation techniques. This project draws from two pertinent areas in ecology: 1) multi-scale field testing and 2) modeling of interrelationships among ecosystem elements to understand coupled effects and improve remaining life predictions. Hence, the research objectives are: 1 ) Identify a battery context and its observability through sensors and data in real deployment conditions for two lives (electric vehicle and grid); 2) Optimize a methodology to translate real-life conditions into the laboratory; 3) Design a large multi-scale testing platform in the laboratory for new and aged cells and modules that mimics real-life conditions; 4) Explore multi-scale battery dynamics and aging by developing reasoning networks that capture the whole battery context variations throughout its scales, reaching the application level; develop theories that link these networks across lives; design battery management systems that can learn to construct and apply these networks to improve their decision making and prediction.

Intellectual Merit

This novel project will provide knowledge and perspectives to two fields by capitalizing upon the similarities between battery context-dependencies, battery life, and ecological systems. This new outlook will provide a unified theory for testing, estimation and management of batteries across cell, module, pack, and application scales and life scales in a research field that up to this point has been disconnected between scales. Testing approaches, interrelationship models, and estimation methods used in ecology are predicted to improve upon present, state-of-the-art battery research methods to provide economic, resiliency and environmental benefits by better understanding and leveraging the unique, time-dependent relationships each battery has with its context.

Broader Impacts

This work will benefit all battery portable, transportation, and grid applications as well as multiple sectors. It will include the emerging battery repurposing sector, by providing tangible methods to improve testing, estimation and management techniques. The result will be longer battery life, better performance, and less environmental waste. Educational impacts include active learning opportunities for undergraduate and graduate students via research and educational interactions with individualized testing boards linked to the newly created large multi-scale testing platform. This strategy will enable low cost, highly distributed testing environments. The Pl will disseminate tools via national education conferences to improve the nearly nonexistent battery testing training of students. This project will facilitate new paths in multi-disciplinary graduate courses. The Pl has a passion to increase representation of Hispanic females in STEM. Outreach will include hosting 4 diverse Community College students for summer research through the Michigan College and University Partnership, and participating in Society for Hispanic Professional Engineers conferences, specifically in the female Hispanic track.

The current approach for designing wave energy converters is to use a floating-body tuned to the wave climate, which results in a very large device that is expensive to build, service and deploy. Additionally, because the device is designed to be tuned to a specific climate, it will not work effectively in a different location ·with a different climate. Therefore, the current approach for designing wave energy converters is not conducive to long-term economic application.

Economically significant size reduction and year-round power increases are only possible through operation near theoretical efficiency limits in constantly changing wave conditions, which requires active hydrodynamic control. However, the wave-by-wave control necessary for best conversion is not possible without wave-elevation information up to some duration into the future (this in large part is because of the force due to the waves generated by body oscillation in response to the incident wave field). By incorporating wave-elevation prediction based on a deterministic propagation model that accounts for a realistic range of wave-group velocities in conjunction with wave measurements in the up-wave directions, we have been able to confirm, through simulations, a 10-fold increase in power conversion under a swept-volume oscillation constraint for an omni-directional heaving buoy type device.

Availability of instantaneous wave profile ("wave surface elevation" or "wave elevation") measurements and wave surface elevation predictions is important to the success of the control approach being pursued in this work. Equally important is the near-optimal wave-by-wave control approach itself.

Proposed Research:

Overview:

MTU will analyze and simulate the power capture from arrays of wave energy converters (WECs) with and without the presence of an object. Nonlinear WECs will be analyzed and exploited for more energy capture. For object detection, MTU will develop an estimator. In addition to having a model that detects the presence of an object, the estimator will use that model and account for uncertainties that we have in the model and also measurement errors; in any case we need to know statistical characteristics about these uncertainties and errors. MTU will participate in the WEC array overall design, analysis, modeling and simulations; control design for Design 2, nonlinear modeling and control, and topology optimization.