Model-based analysis and control of Li-ion batteries and related electrochemical devices
Abstract
Approved for Public ReleaseThe primary objective of this proposal is to develop physics-based models of Li-ion batteries and incorporate them into model-predictive control strategies. Control algorithms make actuation decisions based on sensor readings. However, important aspect of battery control can depend on attributes that cannot be measured directly by sensors. To make sensor inferences, the physics-based models must include predictions of attributes that are not measured directly by sensors. Thus, an important aspect of the proposed research is to develop new sub-models, such as for Li plating and chemo-mechanical degradation. Modeling chemo-mechanical behaviors requires a simultaneous representation of the electrochemistry and structural (stress-strain) behavior. Important constitutive properties must represent crystal-lattice-scale behaviors as functions of Li concentrations. Because physics-based models are computationally too expensive to be run in real time, reduced-order state-space models must be derived from the larger physics-based models. The proposed research will investigate alternative approaches to develop and validate the state-space models. Although the principal focus of the research is concerned with batteries, the underpinning theory and modeling approaches are much more general. Thus, where appropriate, theory and models will be considered for related electrochemical applications such as fuelcells or electrolyzers.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Jul 24, 2023
- Source ID
- N000142312694
Entities
People
- Robert J. Kee
Organizations
- Colorado School of Mines
- Office of Naval Research
- United States Navy