Sökning: onr:"swepub:oai:research.chalmers.se:d6e687fd-629d-46ac-a1c0-5b1858e5b4cd" >
Rapid online health...
Rapid online health estimation for lithium-ion batteries based on partial constant-voltage charging segment
-
- Guo, Wenchao (författare)
- Shanghai Jiao Tong University
-
- Yang, Lin (författare)
- Shanghai Jiao Tong University
-
- Deng, Zhongwei (författare)
- University of Electronic Science and Technology of China,Shanghai Jiao Tong University
-
visa fler...
-
- Li, Jilin (författare)
- Shanghai Jiao Tong University
-
- Bian, Xiaolei, 1990 (författare)
- Chalmers tekniska högskola,Chalmers University of Technology
-
visa färre...
-
(creator_code:org_t)
- 2023
- 2023
- Engelska.
-
Ingår i: Energy. - 0360-5442 .- 1873-6785. ; 281
- Relaterad länk:
-
https://doi.org/10.1...
-
visa fler...
-
https://research.cha...
-
visa färre...
Abstract
Ämnesord
Stäng
- Battery health evaluation is vital for ensuring the security and reliability of lithium-ion batteries. However, the currently proposed methods generally require high-quality input data for feature extraction in online applications. To overcome this obstacle, this paper proposes a rapid online health estimation method only based on partial constant-voltage (CV) charging segment. Firstly, through primary analysis of battery test data, the evolution of CV charging current is confirmed to be correlated with battery capacity. Subsequently, the current evolution constant of CV charging phase is mathematically formulated and quantitatively characterized using a novel health indicator (HI). Besides, charging time and charging capacity are also extracted as HIs to comprehensively capture the CV charging behavior and enhance the robustness of data-driven models. Considering the user's charging habits, an optimized CV segment is determined, enabling a significant reduction in data size and coverage. Finally, three data-driven methods are employed to construct health estimation models by using the extracted HIs, and the best performance is achieved by Gaussian process regression with MAE and RMSE lower than 0.8% and 1%, respectively. Remarkably, the proposed method demonstrates superiority in dealing with sparse sampling, and satisfactory results with 2.9% error under the sparsity of 10 s are obtained.
Ämnesord
- NATURVETENSKAP -- Matematik -- Sannolikhetsteori och statistik (hsv//swe)
- NATURAL SCIENCES -- Mathematics -- Probability Theory and Statistics (hsv//eng)
- TEKNIK OCH TEKNOLOGIER -- Elektroteknik och elektronik -- Signalbehandling (hsv//swe)
- ENGINEERING AND TECHNOLOGY -- Electrical Engineering, Electronic Engineering, Information Engineering -- Signal Processing (hsv//eng)
Nyckelord
- Lithium-ion battery
- Data-driven method
- Health indicator
- Health estimation
- Optimized segment
- Feature extraction
Publikations- och innehållstyp
- art (ämneskategori)
- ref (ämneskategori)
Hitta via bibliotek
-
Energy
(Sök värdpublikationen i LIBRIS)
Till lärosätets databas