A Primer on Bayesian Parameter Estimation and Model Selection for Battery Simulators

2025-12-10

AI Summary

Physical battery models face challenges in experimental calibration—including parameter unidentifiability, structural non-identifiability, and model selection ambiguity. To address these, this work introduces SOBER and BASQ, two novel Bayesian inference algorithms—first systematically applied to battery modeling. Both methods formulate a unified joint Bayesian framework that simultaneously performs parameter estimation and model selection, while integrating sequential experimental design and rigorous uncertainty quantification. Compared to conventional Markov Chain Monte Carlo (MCMC) approaches, they achieve substantially improved inference efficiency and numerical stability. Extensive validation across diverse electrochemical models demonstrates strong discriminative power and robustness under varying experimental conditions. This paradigm enables automated, interpretable modeling of novel battery materials, offering a generalizable statistical methodology to accelerate battery material discovery and performance assessment.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Relational Probabilistic Models
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs
• User Modeling, Personalization and Recommendation: Psychology-informed user models and recommender systems
• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics

Abstract

Physics-based battery modelling has emerged to accelerate battery materials discovery and performance assessment. Its success, however, is still hindered by difficulties in aligning models to experimental data. Bayesian approaches are a valuable tool to overcome these challenges, since they enable prior assumptions and observations to be combined in a principled manner that improves numerical conditioning. Here we introduce two new algorithms to the battery community, SOBER and BASQ, that greatly speed up Bayesian inference for parameterisation and model comparison. We showcase how Bayesian model selection allows us to tackle data observability, model identifiability, and data-informed model development together. We propose this approach for the search for battery models of novel materials.

Problem

• Bayesian methods improve battery model alignment with data
• SOBER and BASQ accelerate Bayesian parameter estimation and selection
• Approach addresses observability, identifiability, and data-informed model development

Innovation

• Bayesian inference algorithms accelerate parameter estimation
• Model selection tackles observability and identifiability issues
• Combines prior assumptions with observations for improved conditioning