Physics-Informed Machine Learning for Battery Degradation Diagnostics: A Comparison of State-of-the-Art Methods

TL;DR

This work addresses the challenge of diagnosing battery degradation at the component level without relying on long-term aging data. It develops and compares four PIML strategies—PINN, data augmentation, delta learning with co-kriging, and delta learning with elastic net—using early-life experimental data augmented with late-life, half-cell model–generated simulations. The study demonstrates that physics-informed approaches, particularly PINN and co-kriging, outperform purely data-driven baselines in predicting capacity and degradation modes ($Q$, $m_p$, $m_n$, $LII$) and exhibit superior extrapolation into heavily degraded regimes. The findings provide practical guidance on method selection and highlight the value of incorporating half-cell physics for robust, scalable degradation diagnostics in battery management systems.

Abstract

Monitoring the health of lithium-ion batteries' internal components as they age is crucial for optimizing cell design and usage control strategies. However, quantifying component-level degradation typically involves aging many cells and destructively analyzing them throughout the aging test, limiting the scope of quantifiable degradation to the test conditions and duration. Fortunately, recent advances in physics-informed machine learning (PIML) for modeling and predicting the battery state of health demonstrate the feasibility of building models to predict the long-term degradation of a lithium-ion battery cell's major components using only short-term aging test data by leveraging physics. In this paper, we present four approaches for building physics-informed machine learning models and comprehensively compare them, considering accuracy, complexity, ease-of-implementation, and their ability to extrapolate to untested conditions. We delve into the details of each physics-informed machine learning method, providing insights specific to implementing them on small battery aging datasets. Our study utilizes long-term cycle aging data from 24 implantable-grade lithium-ion cells subjected to varying temperatures and C-rates over four years. This paper aims to facilitate the selection of an appropriate physics-informed machine learning method for predicting long-term degradation in lithium-ion batteries, using short-term aging data while also providing insights about when to choose which method for general predictive purposes.

Figures (20)

• Figure 1: Overview of capacity fade curves for cells cycled under two distinct temperatures: a. 37; b. 55. The discharge rate for each group of cells appears in the legend in parentheses $(\cdot)$. The standard upper voltage cutoff ($V_{\mathrm{cutoff}}$) is set at 4.075V. The early-life region includes approximately 8 months of experimental data.
• Figure 2: Overview of the half-cell model parameters. a. remaining active masses; b. slippage parameters; c. derived degradation parameters.
• Figure 3: Simulated incremental capacity curves from the half-cell model with: a. one degradation mode; b. two degradation modes.
• Figure 4: An overview of degradation diagnostics using PIML techniques.
• Figure 5: Overview of physics-informed machine learning (PIML) techniques adopted in this comparative study.