Decay rate for single state hysteresis models

[tommaull]

Description

Single state hysteresis models typically feature a decay rate. So far, this has not been implemented in pybamm as far as I can see.

Previous hysteresis models implemented in pybamm:
#2256 and #3194

Motivation

Voltage due to hysteresis can have a long timescale, which is left unmodeled in pybamm. This prevents non-instantaneous hysteresis dynamics from being modeled in pybamm.

Possible Implementation

Decay parameter, dependent on change in lithiation. Personally, I also think it would make more sense to have a single OCP with a hysteresis term than switch between two OCPs.
Latz_Plett_hysteresis (1).pdf

Additional context

No response

[ejfdickinson]

Is the recommendation that this applies to SPMe / DFN models with reference to the open-circuit potential and electrode reaction overpotential, for each electrode independently, and locally in the x-dimension?

The Latz and co-workers approach appears to be speaking of the full cell in terms of OCV and SOC. As a DFN / SPMe submodel alternative to current sigmoid, this would presumably need to work with OCP and lithiation extent respectively.

[tommaull]

@ejfdickinson - yes. I hadn't thought about the x-dim as I usually work with SPMes, but that makes sense. I've opened a discussion here, as there's a number of improvements to OCP that could be made. #3376