Please use this identifier to cite or link to this item:
Title: Data-driven health modeling and assessment of Li-ion battery
Authors: Ma, Hongming
Keywords: Engineering::Electrical and electronic engineering
Issue Date: 2022
Publisher: Nanyang Technological University
Source: Ma, H. (2022). Data-driven health modeling and assessment of Li-ion battery. Master's thesis, Nanyang Technological University, Singapore.
Abstract: Lithium-ion battery is widely applied in numerous fields while machine learning (ML) algorithm is an effective method to model and monitor the aging and operation of battery. In this dissertation, certain popular algorithms and approaches of modeling battery are reviewed in the beginning. After that, the degradation mechanism of battery and methodology for estimating battery state-of-health (SOH) are analyzed. Then, a novel health indicator (HI) named voltage sequence (VS) extracted from V-I profile in charging process is proposed. To model lithium-ion batteries, an efficient machine learning algorithm namely broad learning system (BLS) is proposed, and datasets from Oxford are used to train and test the model. The datasets consist of eight cells, while Cell1 to Cell4 are selected to be the training datasets and Cell5 to Cell8 are selected to be the testing datasets. Root of mean squared error (RMSE) is used to denote estimation errors, and the testing results show that the proposed approach has a high accuracy (average RMSE is 0.0399) to model and monitor lithium-ion battery.
Fulltext Permission: restricted
Fulltext Availability: With Fulltext
Appears in Collections:EEE Theses

Files in This Item:
File Description SizeFormat 
Dissertation_Ma Hongming.pdf
  Restricted Access
3.88 MBAdobe PDFView/Open

Page view(s)

Updated on May 14, 2022


Updated on May 14, 2022

Google ScholarTM


Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.