Hosam Fathy Penn State University

Research Objectives: This proposal addresses the fundamental problem of optimally controlling battery systems for performance, efficiency, and health. The proposed research will use electrochemistry-based models for optimal battery management, with application to health-conscious battery charging, grid energy storage, and hybrid underwater vehicles. Scientific Merit: The proposed research will add three original contributions to the literature. First, it will furnish a novel framework for combined order and index reduction in electrochemistry-based battery models, thereby making these models more conducive to optimal control. Second, it will use boundary control theory to optimize battery charging/discharging for long-term health. Third, it will address the optimal power and thermal management of a battery pack as a single integrated problem. Together, these contributions have the transformative potential to bridge the current gap between the electrochemistry and control technologies.

Broader Impact: The proposed research make a major contribution to battery energy storage and power management. The research discoveries will furnish two new courses on energy system modeling and optimal power management, and support two students in a lab with a solid record of recruiting and training outstanding underrepresented minority students. The PI will disseminate this research to the electrochemistry and control communities ? as well as existing industrial collaborators ? through an edited textbook and web-based seminar series. Finally, the research will enable the PI to incorporate a stronger battery energy storage components into existing toy kits for K-12 student education and outreach.

Project Report

This project examined the problem of using electrochemistry-based models for health-conscious battery control, with particular focus on lithium-ion batteries. The project was motivated by the ubiquity of lithium-ion batteries: thanks to their high power and energy densities, these batteries are used heavily in consumer electronics, and are making inroads into the transportation and stationary energy storage sectors. The project was grounded in the key insight that eking out the best possible performance, efficiency, and longevity from a lithium-ion battery requires model-based control. Prior to this project, the literature already examined model-based lithium-ion battery control extensively. However, there was still a need for research and discovery in the areas of reduced-order, control-oriented battery modeling; battery state and parameter estimation; and computationally efficient online battery control. The intellectual merit of the project stems from the contributions it has made to the scientific literature. Many of these contributions have fed into other projects by the PI, including a multi-investigator NSF project on energy management in large-scale data centers and a multi-institution academic/industrial ARPA-E project on reconfigurable battery pack management. Specific contributions include the use of generalized polynomial chaos theory for combined battery state and parameter estimation, the use of trajectory optimization to improve the fidelity of the resulting gPC estimators, and the optimal design of experiments for thermo-electrochemical lithium-ion battery parameterization. Novel battery control algorithms have been explored, both in the automotive context and the data center context, in close coordination with other complementary research projects. This includes the use of pseudo-spectral optimization for optimal battery charging/discharging, as well as the use of electrochemistry-based models to examine the use of energy storage in large-scale data centers. The broader impact of the project stems from its educational value, both for students at Penn State University and for society at large. Six graduate students and one undergraduate have received substantial training through this project, and produced peer-reviewed publications covering lithium-ion battery modeling, estimation, and control. The students' publications include both research papers and an education/outreach paper disseminated broadly through the ASME Dynamic Systems and Control Magazine. Research produced through this project has been presented to both the control community (through the American Control Conference and ASME Dynamic Systems and Control Conference) and the electrochemistry community (through the Meeting of the Electrochemical Society). The project has provided training opportunities for outstanding students from diverse backgrounds. A paper produced through this project received a best presentation in session award at the 2014 American Control Conference, and the paper's first author was selected as a Mechanical Engineering Student Marshall upon graduation in 2014. Funding from this project has allowed the Principal Investigator to transform 30% of an undergraduate dynamic systems and controls course at Penn State to tutorial videos, freely available to students and scholars worldwide. A year after their initial creation, these videos received more than 24,000 hits online. Funding from this project has also allowed the PI to co-create a new course at Penn State on powertrain system modeling, simulation, and control - with significant emphasis on the modeling and control of electrochemical energy storage in advanced powertrains.

Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
Pennsylvania State University
University Park
United States
Zip Code