A novel data-fusion-model SOH estimation method of Li-ion battery

Source: Beijing Institute of Technology Press

Accurately determining the state of health (SOH) not only helps determine the need for battery replacement but also enhances the accuracy of estimating other important variables such as the State of Charge (SOC). A recent breakthrough study presented by researchers from Anhui University introduces a data-fusion-model method for SOH estimation of Li-ion battery packs based on partial charging curve. This advanced method can enable better predictive maintenance and prevent hazardous situations, such as overheating or thermal runaway.

The study focuses on battery "SOH", which is crucial for managing and optimizing battery performance, especially in applications like electric vehicles and renewable energy systems. The SOH estimation method utilizes dual Gaussian Process Regressions (GPRs) to construct a data-driven and non-parametric aging model based on charging-based Aging Features (AFs). To enhance the accuracy of the aging model, a noise model is established to replace the random noise. Subsequently, the state-space representation of the aging model is incorporated.

Key to the algorithm's success is a comprehensive noise model that adequately captures the diverse sources of uncertainty inherent in the measurement process. It enables researcher to effectively account for uncertainties and enhance the accuracy of SOH estimation for battery packs. Through the integration of the state-space representation with the noise models, a novel data-driven battery aging model has been formulated.

The degradation of batteries is a gradual process, characterized by nonlinear dynamics and non-Gaussian behavior. However, the presence of nonlinearity can result in a multimodal distribution of states, introducing potential bias and even causing filter divergence. As an approximate Bayesian filtering method, the PF can tackle these challenges with the advantage of simulating probability distributions of any shape.

The data-fusion method effectively tracks the degradation of battery capacity. Throughout the degradation process, different battery packs manifest varying rates of degradation. Notably, in the initial and late stages, there is a noticeable acceleration in degradation trends for capacity curves. The simulation results demonstrate that the data-fusion model approach achieves accurate SOH estimation, with maximum errors less than 1.5%.

In order to realize the maximum economic value of retired LIBs for EVs, the reuse of retired LIBs for EVs has become the most economical and environmentally friendly solution. The accurate health assessment is necessary to ensure safety and effectiveness. Besides, integrating battery health data into smart grid systems can enhance energy distribution, load balancing, and demand response capabilities.

In conclusion, a novel approach based on data fusion modeling will accurately estimate the state of SOH of batteries. Significantly, the proposed method capitalizes on charging data and has the potential to be integrated into charging infrastructure, thereby alleviating computational burden. However, the method depends on the standard charging process. Furthermore, future research will focus on investigating the applicability of the proposed method in fast charge states.

Reference

[1] Sun F. Green Energy and Intelligent Transportation—promoting green and intelligent mobility. Green Energy and Intelligent Transportation 2022;1(1):100017.

[2] Chen C, Xiong R, Yang R, Li H. A novel data-driven method for mining battery open-circuit voltage characterization. Green Energy and Intelligent Transportation 2022;1(1):100001.

[3] Li X, Yuan C, Wang Z, He J, Yu S. Lithium battery state-of-health estimation and remaining useful lifetime prediction based on non-parametric aging model and particle filter algorithm. Etransportation 2022; 11:100156.

Author: Xingzi Qiang, Wenting Liu, Zhiqiang Lyu, Haijun Ruan, Xiaoyu Li

Title of original paper: A data-fusion-model method for state of health estimation of Li-ion battery packs based on partial charging curve

Article link: https://doi.org/10.1016/j.geits.2024.100169

Journal: Green Energy and Intelligent Transportation

https://www.sciencedirect.com/science/article/pii/S2773153724000215