open access publication

Article, 2024

State of Health Estimation for Lithium-Ion Battery Based on Sample Transfer Learning under Current Pulse Test

Batteries, ISSN 2313-0105, Volume 10, 5, Page 156, 10.3390/batteries10050156

Contributors

Li, Yuanyuan [1] Huang, Xinrong 0000-0001-7291-2613 [2] Meng, Jinhao 0000-0003-3490-5089 [3] Shi, Kai-Bo 0000-0002-9863-9229 (Corresponding author) [4] Teodorescu, Remus 0000-0002-2617-7168 [5] [6] Stroe, Daniel-Ioan 0000-0002-2938-8921 [5] [6]

Affiliations

  1. [1] Southwest Minzu University
    2. [NORA names: China; Asia, East];
  2. [2] Chang'an University
    3. [NORA names: China; Asia, East];
  3. [3] Xi'an Jiaotong University
    4. [NORA names: China; Asia, East];
  4. [4] Chengdu University
    5. [NORA names: China; Asia, East];
  5. [5] Aalborg University
    6. [NORA names: AAU Aalborg University; University; Denmark; Europe, EU; Nordic; OECD];

Abstract

Considering the diversity of battery data under dynamic test conditions, the stability of battery working data is affected due to the diversity of charge and discharge rates, variability of operating temperature, and randomness of the current state of charge, and the data types are multi-sourced, which increases the difficulty of estimating battery SOH based on data-driven methods. In this paper, a lithium-ion battery state of health estimation method with sample transfer learning under dynamic test conditions is proposed. Through the Tradaboost.R2 method, the weight of the source domain sample data is adjusted to complete the update of the sample data distribution. At the same time, considering the division methods of the six auxiliary and the source domain data set, aging features from different state of charge ranges are selected. It is verified that while the aging feature dimension and the demand for target domain label data are reduced, the estimation accuracy of the lithium-ion battery state of health is not affected by the initial value of the state of charge. By considering the mean absolute error, mean square error and root mean square error, the estimated error results do not exceed 1.2% on the experiment battery data, which highlights the advantages of the proposed methods.

Keywords

R2 method, SoH, TrAdaBoost, accuracy, age, aging features, auxiliary, battery, battery SOH, battery data, battery state of health, charge, conditions, current pulse test, data, data distribution, data types, data-driven method, difficulties, dimensions, discharge rate, distribution, diversity, division, division method, domain data, domain labeled data, dynamic test conditions, error, error results, estimation, estimation accuracy, estimation error results, estimation method, experiments, feature dimensions, features, health, health estimation, health estimation method, labeled data, learning, lithium-ion, lithium-ion batteries, lithium-ion battery state, lithium-ion battery state of health, mean square error, method, multi-source, pulse test, randomization, range, rate, results, root, root mean square error, sample data, sample data distribution, samples, source, source domain data, square error, stability, state, state of charge, state of health, state of health estimation, state-of-health estimation method, target, temperature, test, test conditions, transfer learning, type, update, variables, weight, work data

Funders

  • National Natural Science Foundation of China

Data Provider: Digital Science

LINKS
-

Matching Records in NORA

SUBJECTS
+

METRICS
+