State of health and remaining useful life prediction of lithium-ion batteries based on a disturbance-free incremental capacity and differential voltage analysis method

Xia, Fei; Wang, Kangan; Chen, Jiajun

doi:10.1016/j.est.2023.107161

Cited by 39 publications

(5 citation statements)

References 51 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Model capability [1] GRNN Type 1 [2] Residual network with attention mechanism Type 2 [3] TCN Type 2 [4] CNN-LSTM Type 2 [5], [6] LSTM Type 3 [7] Bi-directional GRU Type 3 [8] Bi-directional LSTM Type 3 [9] LSTM with attention Type 3 [10] Bi-directional LSTM with attention Type 3 [11] Transformer Type 3 [12] FFNN Type 3…”

Section: Paper Primary Architecturesmentioning

confidence: 99%

See 1 more Smart Citation

A machine learning framework for remaining useful lifetime prediction of li-ion batteries using diverse neural networks

Lee,

Sun,

Liu

et al. 2023

Preprint

View full text Add to dashboard Cite

Accurate prediction of the remaining useful life (RUL) of li-ion batteries (LIBs) is essential for enhancing the operational efficiency and safety of LIB-powered applications. It also facilitates improvements in the cell design process and the evolution of fast charging methodologies, thereby minimizing cycle testing time. While artificial neural networks (ANNs) have emerged as promising tools for this task, identifying the optimal architecture across diverse datasets and optimization strategies is non-trivial. To address this challenge, a machine learning framework was developed for a systematic evaluation of diverse ANN architectures. Utilizing just 30% of the training dataset from 124 li-ion batteries cycled under various charging policies, hyperparameter optimization is conducted within this framework. This ensures that each model is evaluated at its optimal configuration, facilitating a balanced comparison for RUL prediction tasks. Furthermore, the study examines the influence of varied cycle windows on model efficacy. Employing a stratified partitioning method highlights the importance of uniform dataset representation across different subsets. Notably, the top-performing model, using cycle-by-cycle features from just 40 cycles, achieves a mean absolute percentage error of 10.7%.

show abstract

Section: Paper Primary Architecturesmentioning

confidence: 99%

“…Xia et al [7] employed a bidirectional gated recurrent unit (GRU) for the prediction of the RUL. They introduced an innovative approach by reconstructing the voltage from a second-order RC equivalent circuit model to obtain incremental capacity and differential voltage curves.…”

Section: Paper Primary Architecturesmentioning

confidence: 99%

A machine learning framework for remaining useful lifetime prediction of li-ion batteries using diverse neural networks

Lee,

Sun,

Liu

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Advanced BMS with wireless communication systems and advanced sensor technologies could improve battery monitoring [49,50]. Hence, next-generation models for battery performance and aging (SoH and RUL) monitoring can upgrade BMS functionalities [51][52][53][54][55].…”

Section: Battery Management Systemsmentioning

confidence: 99%

Selecting Suitable Battery Technologies for Untethered Robot

et al. 2023

View full text Add to dashboard Cite

Untethered robots carry their own power supply in the form of a battery pack, which has a crucial impact on the robot’s performance. Although battery technologies are richly studied and optimized for applications such as electric vehicles, computers and smartphones, they are often a mere afterthought in the design process of a robot system. This tutorial paper proposes criteria to evaluate the suitability of different battery technologies for robotic applications. Taking into consideration the requirements of different applications, the capabilities of relevant battery technologies are evaluated and compared. The tutorial also discusses current limitations and new technological developments, pointing out opportunities for interdisciplinary research between the battery technology and robotics communities.

show abstract

“…Commonly used data-driven methods include linear regression 25 , support vector machines 26 , Gaussian process regression 27 , deep neural networks 28 , 29 , etc. Xia et al 30 extracted features from incremental capacity (IC) curves and differential voltage (DV) curves to estimate SOH. Wang et al 31 extracted valuable health indicators from electrochemical impedance spectroscopy (EIS) as input for Gaussian process regression to estimate SOH.…”

Section: Introductionmentioning

confidence: 99%

“…The data-driven models have high accuracy and efficiency, but its generalizability depends on the extracted features and have poor stability 14 , 33 . For instance, due to the high usage variability, existing methods 30 , 34 , 35 need to extract specific features for different datasets or different working conditions, leading to the fact that models are dataset-specific, resulting in a waste of computing resources. The promising prospect of physics-informed neural network (PINN) 36 , 37 lies in amalgamating the strengths of physics-based and data-driven approaches, potentially addressing the aforementioned challenges.…”

Section: Introductionmentioning

confidence: 99%

Physics-informed neural network for lithium-ion battery degradation stable modeling and prognosis

Wang,

Zhai,

Zhao

et al. 2024

Nat Commun

View full text Add to dashboard Cite

Accurate state-of-health (SOH) estimation is critical for reliable and safe operation of lithium-ion batteries. However, reliable and stable battery SOH estimation remains challenging due to diverse battery types and operating conditions. In this paper, we propose a physics-informed neural network (PINN) for accurate and stable estimation of battery SOH. Specifically, we model the attributes that affect the battery degradation from the perspective of empirical degradation and state space equations, and utilize neural networks to capture battery degradation dynamics. A general feature extraction method is designed to extract statistical features from a short period of data before the battery is fully charged, enabling our method applicable to different battery types and charge/discharge protocols. Additionally, we generate a comprehensive dataset consisting of 55 lithium-nickel-cobalt-manganese-oxide (NCM) batteries. Combined with three other datasets from different manufacturers, we use a total of 387 batteries with 310,705 samples to validate our method. The mean absolute percentage error (MAPE) is 0.87%. Our proposed PINN has demonstrated remarkable performance in regular experiments, small sample experiments, and transfer experiments when compared to alternative neural networks. This study highlights the promise of physics-informed machine learning for battery degradation modeling and SOH estimation.

show abstract

State of health and remaining useful life prediction of lithium-ion batteries based on a disturbance-free incremental capacity and differential voltage analysis method

Cited by 39 publications

References 51 publications

A machine learning framework for remaining useful lifetime prediction of li-ion batteries using diverse neural networks

A machine learning framework for remaining useful lifetime prediction of li-ion batteries using diverse neural networks

Selecting Suitable Battery Technologies for Untethered Robot

Physics-informed neural network for lithium-ion battery degradation stable modeling and prognosis

Contact Info

Product

Resources

About