Enhanced vision-transformer integrating with semi-supervised transfer learning for state of health and remaining useful life estimation of lithium-ion batteries

Energy and AI, Год журнала: 2024, Номер 17, С. 100405 - 100405

Опубликована: Авг. 3, 2024

The state of health (SOH) and remaining useful life (RUL) lithium-ion batteries are crucial for management diagnosis. However, most data-driven estimation methods heavily rely on scarce labeled data, while traditional transfer learning faces challenges in handling domain shifts across various battery types. This paper proposes an enhanced vision-transformer integrating with semi-supervised SOH RUL batteries. A depth-wise separable convolutional is developed to extract local aging details convolutions establishes global dependencies between information using multi-head attention. Maximum mean discrepancy employed initially reduce the distribution difference source target domains, providing a superior starting point fine-tuning model. Subsequently, abundant data same type as through learning, compensating model's limitations capturing characteristics. Consistency regularization incorporates cross-entropy predictions without adversarial perturbations into gradient backpropagation overall In particular, experimental groups 13–15 different types batteries, root square error was less than 0.66 %, relative 3.86 %. Leveraging extensive unlabeled proposed method could achieve accurate RUL.

Язык: Английский

---

A Review of Non-Destructive Testing for Lithium Batteries

Energies, Год журнала: 2024, Номер 17(16), С. 4030 - 4030

Опубликована: Авг. 14, 2024

With the rapid development of mobile devices, electronic products, and electric vehicles, lithium batteries have shown great potential for energy storage, attributed to their long endurance high density. In order ensure safety batteries, it is essential monitor state health charge/discharge. There are commonly two methods measuring batteries: destructive testing non-destructive testing. Destructive not suitable in situ or analysis as can cause irreversible deformation damage battery. Herein, this review focuses on three including ultrasonic testing, computer tomography, nuclear magnetic resonance. Ultrasonic widely used crack fatigue detection. X-ray tomography neutron gained increasing attention monitoring status batteries. Nuclear resonance be conduct ex review, summarized, current status, achievements, perspectives technology.

Язык: Английский

---

A procedure for evaluating the SOH of Li-ion batteries from data during the constant voltage charge phase and the use of an ECM with internal resistance

Journal of Energy Storage, Год журнала: 2024, Номер 108, С. 115074 - 115074

Опубликована: Дек. 24, 2024

Язык: Английский

---

Long‐Term Estimation of SoH Using Cascaded LSTM‐RNN for Lithium Batteries Subjected to Aging and Accelerated Degradation

Energy Storage, Год журнала: 2024, Номер 6(8)

Опубликована: Ноя. 5, 2024

ABSTRACT Accurate estimation of state health (SoH) the battery over long‐term is a critical challenge for management systems in electric vehicles. This due to challenges accurately modeling accelerated aging and degradation phenomena caused by diverse operating conditions battery. paper presents cascaded recurrent neural networks (RNN) with long short‐term memory (LSTM) estimate internal resistance SoH, taking account various abnormal A datasheet‐based model developed using fade equations. The training validation data set LSTM‐RNN are generated subjecting factors that cause degradation, such as fast charging, varying temperatures, overutilization, cell imbalance. trained SoH only once after completion every charge–discharge cycle. error index parameters proposed estimator well within 1%, demonstrating reliability robustness

Язык: Английский

---

The end-of-life power battery recycling & remanufacturing center location-adjustment problem considering battery capacity and quantity uncertainty

Journal of Environmental Management, Год журнала: 2024, Номер 357, С. 120774 - 120774

Опубликована: Апрель 1, 2024

Язык: Английский

---