Application of physics informed machine learning models and simulations tools in cell and battery design and performance predictions
Application of physics informed machine learning models and simulations tools in cell and battery design and performance predictions
Date delivered: 29th October 2026
Webinar description
Long-term battery cell cycling tests can take up to six months and cost as much as £5,000 per cell, creating significant challenges for developers seeking faster innovation.
Predictive modelling offers a way to accelerate development, but conventional machine learning and AI models often rely on extensive datasets from commercial cells and struggle when only limited early-stage data is available.
This webinar will explores the strengths and limitations of physics-based, machine learning and physics-informed machine learning approaches for forecasting long-term cell performance. Through real-world case studies, discover how UKBIC-generated data can improve prediction accuracy, identify degradation mechanisms, and help reduce development times, costs, and time-to-market.




