[Lecture Note] Li-ion Battery Ageing & Degradation : Health State Evalutaion and Lifetime Prediction

Latest Advances in the Global Energy Storage Industry

New energy storage has been developing rapidly.

Dominace : China, The United States, and Europe

Lastest ADvances in Energy Storage Technology:

Many choice depending on your humiliation of stereotype

A record high : 66.43GW / 189.48 GWh

Up to 2025, Renewable up

Accelerated ageing and Complex Life-time prediction of Li-ion Battery

Lifetime model of the Batteries :

1. Data-Drriven (Black - box Model)
2. Mechanism - BAsed (white-box model) : easy to understand, but not a pratical model. It is complex, how to distinguish the main factors ?
3. Gray Model ? Combine Black + White.

Degration in battergy performance

What kind of ? Quality?

Decided by the:

1. Intrisinc Factor : Material, components, Cell manufacturing
2. External factors

Critical factors : Physics(Mechanics), Chemistry

Materials and Components

• Initial status : LLI, LAM
• Polarization conditions: IR (Rct, Rsei, Rdiff)

Operando and Non-invasive Techniques

Spectro-Imaging Techniques

1. Sybnchtron Nano-CT/TXM
2. Solid NMR + MRI
3. Neutron Imaging Techniques

Two key issues during Bettery Degradation :

1. Electrolyte Degradation
2. Li Plating

Analysis of Electrolyte Inventory at Kneeing Points

• Components’ Changes of Ekectrolytes

At kneeing Point, the residual concentration decreased to about 8%.

Li-plating

1. Li plating exists during life-time of Li-ion Batteries
2. The Origins and “side effects” of Li-plating on battery : Accelerating DEgradation and Safety Concerns
3. How to characterize Li-plating : Qualitative, Quantitative and Operando Analysis

How

Methods :

1. Operando XRD : Formation of LiC6 -> Lithiation Uniformity
2. Operando NMR -> Detection

Interfacail Kinetics Affect Lithiation Uniformity

Batter interfacial kinetics -> More uniform Li plating

Long-Term Operando Quantification of Liplating on Graphite Anodes

Calendar Aging : Quantitative Analysis of Synergistic Effects of SEI Evolution and Electrolyte DEcomposition

Higher temperature, the higher transfer

LLI linearly increases with SOH

AI battery investigation

Different degradation cureve

How to se and unified

1. Mechanism-informed features ;
2. Soft - Modular of Expert System;

Continual & online learning

Unified BHM Framework

1. Feature engineering
2. SoftMoE SOH estimation
3. Cross-dataset continual learning
4. UKF online trajectory update

Conclusions

Batter Ageing / Degradation is a complex process

Li-plating & Electrolyte degradation are key factors

Results in large polarization and Loss of Lithium Inventory

The growth of SEI on graphite is an indicator to reflect cycling or calendaring degradation state of LFP-graphite batteries