GROUP 1

Software Development

Computational infrastructure that underpins the Multi-Scale Modelling programme.

Group 1 focuses on software development, providing the computational infrastructure that underpins the Multi-Scale Modelling project rather than addressing research questions directly.

Its role is to develop, maintain, and contribute to software tools that support the other work packages and embed their scientific advances into a cohesive modelling and analysis ecosystem.

Group 1 contributes to five key tools spanning multiple scales of battery research: ONETEP for atomistic simulations, PyBaMM for continuum-scale cell modelling, PyBOP for parameter inference, PyProBE for experimental data processing, and PyECN for battery pack modelling.

Related publications
Multi-scale modelling software
Computational infrastructure · open-source tools · multi-scale modelling
Research focus
ONETEP Atomistic simulations and electronic structure modelling.
PyBaMM Continuum-scale electrochemical modelling.
PyBOP Parameter inference and model calibration.
Core capability · Parameterisation
GROUP 2

Parameterisation

Robust and standardised methods for accurate battery models.

Group 2 focuses on developing robust and standardised methods to parameterise battery models.

Parameterisation is critical for accurate battery models and is sensitive to variations in electrode, electrolyte composition, manufacturing processes, and the complexity of the underlying mechanisms, including spatial and temporal heterogeneity.

This can be achieved through parameter extraction from degradation experiments using PyBOP and developing an automated multiscale pipeline linking ONETEP and MLIP atomistic simulations to PyBaMM.

This will support better model selection, capture heterogeneity and nonlinear effects and enable faster, cheaper battery design and diagnosis.

Related publications
Battery parameterisation
Parameter extraction · multiscale pipeline · model selection
Parameter extraction Learning parameters from degradation experiments using data-driven inference.
Multiscale pipeline Connecting atomistic simulations with continuum battery models.
Heterogeneity Capturing spatial and temporal variations in battery behaviour.
Model selection Selecting models that capture the relevant underlying physics.
Better parameters mean better models — enabling faster, cheaper and more reliable battery design, diagnosis and prediction.