Accelerated Multiscale Modelling of Batteries (AMMBa)

13e7aeae-db8a-48ba-b617-c288a9c5e32d

Closed

Innovate UK
Innovate UK
Innovate UK
Innovate UK
Innovate UK

£698,100

Sept. 30, 2020

March 31, 2021

The aim of the AMMBA project is to improve our ability to simulate the performance of batteries quickly and accurately, thereby enabling better designs to be delivered in less time and for lower costs.

The need to reduce carbon emissions from transport is clear and pressing, as part of a strategy to keep global temperature rises below 2degC and minimise the effects of climate change. The key contributor to this within the mobility sector is the electrification of vehicles, enabling the motive power to come directly from low-carbon, grid-generated sources. The most fundamental part of new vehicles is the battery, the storage medium of this renewable energy -- but the current performance of batteries in terms of energy density and cost makes ZEVs lower range and more expensive than the vehicles we wish them to replace.

To increase the uptake of ZEVs we need to make them cheaper and with greater range -- both issues which largely stem from the capability of the battery. Simulation and modelling can play a pivotal role in this by providing the engineers and designers with the tools required to understand the performance of the battery pack in a virtual environment, which allows faster, cheaper development to take place. Currently we rely on significant amounts of physical testing which, whilst delivering accurate results, is costly and only provides data for the cell and battery configuration in question. Simulation tools are available and heavily used, but the inbuilt models do not typically link the cell level electro-chemistry to the thermodynamic responses which leads to lower accuracy than is required. Improving simulation and modelling capabilities by coupling these phenomena will allow determination of pack level performance from a simple cell characterisation -- a more accurate method than is currently available. In addition, through the use of machine learning techniques, the AMMBa project aims to develop simulation tools that run far faster than is currently possible, enabling designers and engineers to compare far more designs, in less time than they do now. In doing so this will allow better battery designs to be delivered faster and for lower costs than is currently possible, leading to improved, cheaper ZEVs.