Formulation and manufacturing optimization of lithium-ion graphite-based electrodes via machine learning

Understanding the formulation and manufacturing parameters that lead to higher energy density longevity is critical designing energy-dense graphite electrodes for battery applications. A limited dataset includes 27 different formulation, protocols, performance properties reported. Input from are varied: slurry composition, mixing protocol, electrode coating gap size, drying temperature, speed, calendering. Measurable outputs rheological characteristics, adhesion, electrochemical testing recorded. database with inputs output populated used train an artificial intelligence model. Validation of model performed upon test data optimized process predicted. The manufactured using shows excellent cycle life capacity agreement prediction. can be predict design produce thick, high-coat-weight, graphite-based electrodes.