Online estimation of lithium‐ion batteries state of health during discharge

Abstract

Summary In order to solve the problem that the unsatisfactory accuracy of SOH estimation method, which seeks the relationship between battery life and external characteristics through experiments, is restricted by battery consistency in a large number of battery applications, this paper proposes an SOH estimation framework which can automatically correct the errors caused by the battery consistency problem online. The SOH framework realizes the automatic online fast correction of SOH estimation error through the designed closed‐loop feedback framework. Another advantage of this framework is that it can achieve accurate estimation for the batteries state of health (SOH) during the irregular charging and discharging process of electric vehicles. And in this framework, a new equivalent circuit based on the autoregressive (AR) model is proposed to reduce the complexity of the battery method while ensuring the accuracy of the estimation, which has better robustness in practical applications. Finally, it is proved that the online estimation of lithium‐ion batteries SOH during discharge proposed in this paper has better practicability and higher estimation accuracy by comparing with the traditional SOH method of external feature relationship.

State of health estimation for lithium-ion batteries based on incremental capacity analysis under slight overcharge voltage

“…We can find that the estimation performances of the proposed methods are all below 1%. In summary, the GPR method-based ICA can achieve reliable and accurate SOH estimation under a small sample of slight overcharge voltage cycling for the LFP battery Liu F. et al, 2021.…”

“…The SOH is an important indicator to evaluate the aging state. Currently, the SOH estimation methods have been mainly divided into two categories: model-based methods and data-driven methods (Xiong et al, 2018;Liu F. et al, 2021;Khaleghi et al, 2022;Wu et al, 2022). Among them, the model-based methods mainly include electrochemical methods, empirical methods, and equivalent circuit methods.…”

Fusion estimation strategy based on dual adaptive Kalman filtering algorithm for the state of charge and state of health of hybrid electric vehicle Li‐ion batteries

“…33 Yang SJ et al 34 presented a voltage reconstruction model, which takes into consideration the limited battery operation range in practice and the over-potential caused by the large current rate, and not only accurately estimates the SOH but also quantitatively identifies the aging models. Liu Fang et al 35 proposed a SOH estimation framework that can automatically correct the errors caused by the battery consistency problem online, and a new equivalent circuit based on the autoregressive (AR) model is proposed to reduce the complexity of the battery method while ensuring the accuracy of the SOH estimation.…”

An Improved Capacity-Loss Diagnostic Model Based on Long Short-Term Memory Network

“…Based on the electrochemical model, many researchers have proposed physical and semi-empirical models that account for diverse capacityloss mechanisms such as loss of lithium inventory (LLI), loss of active material (LAM), etc. [12][13][14] The application of these methods in realworld is a challenge because LIB is a non-linear and time-variable system, where capacity-loss is the coupling result of multiple aging mechanisms and its computational process is composed of several differential and partial differential equations. Based on the above studies, 15 the capacity-loss can be diagnosed by extracting the characteristic parameters associated with the capacity-loss mechanism, which greatly simplifies the computational load and achieves the decoupling of different capacity-loss mechanisms to some extent.…”