The Incorporation of predictive knowledge into the design and control of advanced secondary battery technolgy
Abstract
ABSTRACT: The overarching objective of this proposal is to develop, extend, and apply predictive models that assist the development and deployment of electrochemical power systems. The primary focus is on secondary (rechargeable) batteries and supporting technologies such as battery management systems (BMS). The research begins with physics-based models that represent the underpinning electrochemical, thermal, and transport processes. Developing and validating thermodynamic, transport, and kinetics properties for new materials is an important aspect of the physics-based modeling. The models are typically manifested as large systems of partial differential equations that can be solved computationally. The physics-based models offer predictive capabilities that are highly valuable in the design, development, optimization, and control of new technologies.Because the physics models can be computationally expensive, they are usually not directly applicable to real-time simulation and control. The proposed approach exercises the fundamental physics models to derive lower dimensional models that can be utilized in system analysis tools that enable a variety of battery management functions. Perhaps surprisingly, very large physics models (e.g., order 100,000 states) can be reduced to local linear state-space models with order tens of states. Despite the extraordinary dimensional reduction, state-space models can represent accurately actuation-response dynamics. The system analysis tools are directly applicable to real-time model-predictive control (MPC), state-of-health (SOH) monitoring, and interpretation of electrochemical impedance spectra (EIS).
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- May 08, 2020
- Source ID
- N000142012492
Entities
People
- Robert J. Kee
Organizations
- Colorado School of Mines
- Office of Naval Research
- United States Navy