To enable the development of reliable methods for identifying electrochemical model parameters, it is important to have realistic simulations for method developments. The overall objective is to develop a thermal control framework that can be integrated into PyBaMM for closed loop simulations. This will support development of methods for electrochemical parameter identification (WP T3.2).

Identification of time-varying model parameters is essential for maintaining model fidelity. An example of such a system is lithium-ion batteries, whose behaviour changes as they age. When the battery ages, its response during operation differs from that of a battery fresh from the factory.

During battery operation, we want to constrain variables such as current and cell temperature within a predefined safe region. The current can be directly constrained while the cell temperature can be regulated via the surface temperature. Enforcing these limits is essential for safe operation. To be able to trust an identification method developed for a given application, it is therefore important to include a realistic implementation of a controller in simulations.

The work during this research stay has focused on implementing a realistic controller in PyBaMM to support the development of identification methods for electrochemical battery models, such as the Doyle–Fuller–Newman
(DFN) model.

Daniel Jakobsson

Chalmers University of Technology

Electrical engineering/Systems and Control/Automatic control