Speaker: Dr. Samuel Cooper, Imperial College, UK
Battery manufacturers want to know the relationship between their manufacturing parameters and the performance of the resulting cell, so that they can optimise their product for particular applications. The literature contains many examples of physics-based models of the various manufacturing processes, including mixing, coating, heat and compression, each of which are hugely complex; expensive to simulate and hard to validate.
Recent advances in generative machine learning (ML) methods have allowed the relationship from manufacturing parameters to microstructure to be directly learned from data.
In this talk, Sam will present a modular approach to the cell optimisation cycle that makes use of these ML methods, in combination with GPU accelerated metric extraction (TauFactor 2), electrochemical cell simulation (PyBaMM), and Bayesian optimisation. In addition, Sam will be introducing a new kintsugi SEM imaging method for accurately observing the nanostructure of the carbon binder domain; “VoxCel” an open-source, voxel-based, GPU accelerated, multi-physics cell simulation; MLs methods for generating 3D data from 2D images, as well as, inpainting artefacts in image data; and a data fusion method for combining multi-modal datasets using GANs. Lastly, Sam will present a webapp that normalises the data obtained from testing cells in a lab for easy comparison to commercial cells: cell-normaliser.
- Join the Teams meeting on the 12 May 2023