Incremental capacity-based multi-feature fusion model for predicting state-of-health of lithium-ion batteries

2025-03-31 · Chenyu Jia, Guangshu Xia, Yuanhao Shi, Jianfang Jia, Jie Wen, Jianchao Zeng

Lithium-ion batteries have become an indispensable part of human industrial production and daily life. For the safe use, management and maintenance of lithium-ion batteries, the state of health (SOH) of lithium-ion batteries is an important indicator so that the SOH estimation is of significant practical value. In order to accurately predict SOH, this paper proposes a fusion prediction model which combines particle swarm optimization (PSO) algorithm, bi-directional long-short time memory network (BiLSTM) and adaptive boosting (AdaBoost) algorithm. In the proposed prediction model, indirect health indicators (HIs), which characterize battery degradation, are obtained with the help of incremental capacity analysis (ICA), and is fed into BiLSTM to extract time-series features, whose parameters are optimized by employing PSO algorithm. On this basis, the AdaBoost algorithm is applied to reduce the risk of overfitting the PSO-BiLSTM model. The study based on lithium-ion battery data from Center for Advanced Life Cycle Engineering (CALCE) shows that the PSO-BiLSTM-AdaBoost model has higher accuracy, better robustness, and generalization ability.

📄 PDF Abstract BibTeX arXiv:2503.23858

Code (0)

등록된 구현이 없습니다.

Methods 이 논문이 사용한 방법론

Sigmoid Activation 설명 없음

Tanh Activation 설명 없음

LSTM An LSTM is a type of recurrent neural network that addresses the vanishing gradient problem in vanilla…

Memory Network 설명 없음

BiLSTM A Bidirectional LSTM, or biLSTM, is a sequence processing model that consists of two LSTMs: one taking the input in a forward direction, and the other in a backwards…

Reducing Class-wise Confusion for Incremental Learning with Disentangled Manifolds

2025-03-22 · CVPR 2025 1 · Huitong Chen, Yu Wang, Yan Fan, Guosong Jiang 외

Class incremental learning (CIL) aims to enable models to continuously learn new classes without catastrophically forgetting old ones. A promising direction is to learn and use prototypes of classes during incremental up…

class-incremental learning Class Incremental Learning Incremental Learning

Diffusion Model Meets Non-Exemplar Class-Incremental Learning and Beyond

2024-08-06 · Jichuan Zhang, YaLi Li, Xin Liu, Shengjin Wang

Non-exemplar class-incremental learning (NECIL) is to resist catastrophic forgetting without saving old class samples. Prior methodologies generally employ simple rules to generate features for replaying, suffering from …

class-incremental learning Class Incremental Learning Exemplar-Free Incremental Learning+2

Leveraging Pre-Trained Models for Multimodal Class-Incremental Learning under Adaptive Fusion

2025-02-07 · Yukun Chen, Zihuan Qiu, Fanman Meng, Hongliang Li 외

Unlike traditional Multimodal Class-Incremental Learning (MCIL) methods that focus only on vision and text, this paper explores MCIL across vision, audio and text modalities, addressing challenges in integrating compleme…

class-incremental learning Class Incremental Learning cross-modal alignment Diversity+2

Crafting Your Evolving Dreams: Concept-Incremental Versatile Customization

2026-06-03 · Jiahua Dong, Wenqi Liang, Hongliu Li, Yang Cong 외 arxiv

Custom diffusion models (CDMs) have garnered significant interest owing to their remarkable capacity for generating personalized concepts. However, the majority of CDMs unrealistically presume that the user's collection …

Continual Learning

Class-Independent Increment: An Efficient Approach for Multi-label Class-Incremental Learning

2025-03-01 · Songlin Dong, Yuhang He, Zhengdong Zhou, Haoyu Luo 외

Current research on class-incremental learning primarily focuses on single-label classification tasks. However, real-world applications often involve multi-label scenarios, such as image retrieval and medical imaging. Th…

class-incremental learning Class Incremental Learning Image Retrieval Incremental Learning