A Simple and Effective KAN-Based Architecture for Accurate Battery RUL Prediction
Accurately estimating a lithium-ion battery’s Remaining Useful Life (RUL) is crucial for ensuring the safety and reliability of battery management systems. However, the performance of emerging architectures, such as Kolmogorov-Arnold Networks (KANs), is often hindered by the significant n...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/11084810/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849713169309630464 |
|---|---|
| author | Guangzai Ye Li Feng Jianlan Guo Yuqiang Chen Shufei Li |
| author_facet | Guangzai Ye Li Feng Jianlan Guo Yuqiang Chen Shufei Li |
| author_sort | Guangzai Ye |
| collection | DOAJ |
| description | Accurately estimating a lithium-ion battery’s Remaining Useful Life (RUL) is crucial for ensuring the safety and reliability of battery management systems. However, the performance of emerging architectures, such as Kolmogorov-Arnold Networks (KANs), is often hindered by the significant noise and complex temporal dynamics present in real-world battery data. To overcome this, we propose a novel Knowledge Distillation-Based Denoising and Channel-Independent Kolmogorov-Arnold Networks (DCI-KANs) architecture, specifically designed to enhance robustness and accuracy for multivariate time series RUL prediction. Our approach integrates a VMD-based denoising mechanism, compressed into a more compact KAN model via knowledge distillation, to mitigate noise efficiently. It also incorporates a channel-independent KANs structure with a regularized weighted loss function to handle variable-specific degradation patterns. Experimental results on two public battery datasets show that DCI-KANs substantially outperform existing state-of-the-art methods in RUL prediction accuracy, achieving a 43 % and 22 % reduction in Absolute Relative Error (ARE) on the NASA and CALCE datasets, respectively. This work presents a simple yet effective framework that makes KANs a practical, robust, and computationally efficient solution for the challenging task of real-world battery RUL prediction. |
| format | Article |
| id | doaj-art-ab6fd4aa2c9d41abb48371fba70fb216 |
| institution | DOAJ |
| issn | 2169-3536 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj-art-ab6fd4aa2c9d41abb48371fba70fb2162025-08-20T03:14:02ZengIEEEIEEE Access2169-35362025-01-011312870012871510.1109/ACCESS.2025.359048011084810A Simple and Effective KAN-Based Architecture for Accurate Battery RUL PredictionGuangzai Ye0https://orcid.org/0000-0002-0075-9602Li Feng1https://orcid.org/0000-0001-6172-8650Jianlan Guo2Yuqiang Chen3Shufei Li4School of Computer Science and Engineering, Macau University of Science and Technology, Macau, SAR, ChinaSchool of Computer Science and Engineering, Macau University of Science and Technology, Macau, SAR, ChinaDongguan Polytechnic, Dongguan, ChinaDongguan Polytechnic, Dongguan, ChinaDongguan Polytechnic, Dongguan, ChinaAccurately estimating a lithium-ion battery’s Remaining Useful Life (RUL) is crucial for ensuring the safety and reliability of battery management systems. However, the performance of emerging architectures, such as Kolmogorov-Arnold Networks (KANs), is often hindered by the significant noise and complex temporal dynamics present in real-world battery data. To overcome this, we propose a novel Knowledge Distillation-Based Denoising and Channel-Independent Kolmogorov-Arnold Networks (DCI-KANs) architecture, specifically designed to enhance robustness and accuracy for multivariate time series RUL prediction. Our approach integrates a VMD-based denoising mechanism, compressed into a more compact KAN model via knowledge distillation, to mitigate noise efficiently. It also incorporates a channel-independent KANs structure with a regularized weighted loss function to handle variable-specific degradation patterns. Experimental results on two public battery datasets show that DCI-KANs substantially outperform existing state-of-the-art methods in RUL prediction accuracy, achieving a 43 % and 22 % reduction in Absolute Relative Error (ARE) on the NASA and CALCE datasets, respectively. This work presents a simple yet effective framework that makes KANs a practical, robust, and computationally efficient solution for the challenging task of real-world battery RUL prediction.https://ieeexplore.ieee.org/document/11084810/Remaining useful life predictionmultivariate time seriesKolmogorov-Arnold networkschannel independentvariational mode decompositionlithium-ion batteries |
| spellingShingle | Guangzai Ye Li Feng Jianlan Guo Yuqiang Chen Shufei Li A Simple and Effective KAN-Based Architecture for Accurate Battery RUL Prediction IEEE Access Remaining useful life prediction multivariate time series Kolmogorov-Arnold networks channel independent variational mode decomposition lithium-ion batteries |
| title | A Simple and Effective KAN-Based Architecture for Accurate Battery RUL Prediction |
| title_full | A Simple and Effective KAN-Based Architecture for Accurate Battery RUL Prediction |
| title_fullStr | A Simple and Effective KAN-Based Architecture for Accurate Battery RUL Prediction |
| title_full_unstemmed | A Simple and Effective KAN-Based Architecture for Accurate Battery RUL Prediction |
| title_short | A Simple and Effective KAN-Based Architecture for Accurate Battery RUL Prediction |
| title_sort | simple and effective kan based architecture for accurate battery rul prediction |
| topic | Remaining useful life prediction multivariate time series Kolmogorov-Arnold networks channel independent variational mode decomposition lithium-ion batteries |
| url | https://ieeexplore.ieee.org/document/11084810/ |
| work_keys_str_mv | AT guangzaiye asimpleandeffectivekanbasedarchitectureforaccuratebatteryrulprediction AT lifeng asimpleandeffectivekanbasedarchitectureforaccuratebatteryrulprediction AT jianlanguo asimpleandeffectivekanbasedarchitectureforaccuratebatteryrulprediction AT yuqiangchen asimpleandeffectivekanbasedarchitectureforaccuratebatteryrulprediction AT shufeili asimpleandeffectivekanbasedarchitectureforaccuratebatteryrulprediction AT guangzaiye simpleandeffectivekanbasedarchitectureforaccuratebatteryrulprediction AT lifeng simpleandeffectivekanbasedarchitectureforaccuratebatteryrulprediction AT jianlanguo simpleandeffectivekanbasedarchitectureforaccuratebatteryrulprediction AT yuqiangchen simpleandeffectivekanbasedarchitectureforaccuratebatteryrulprediction AT shufeili simpleandeffectivekanbasedarchitectureforaccuratebatteryrulprediction |