Journal article
Diagnosis of wind turbine faults with transfer learning algorithms
Renewable energy, Vol.163, pp.2053-2067
01/2021
DOI: 10.1016/j.renene.2020.10.121
Abstract
A framework of using transfer learning algorithms, Inception V3 and TrAdaBoost, for fault diagnosis of two wind turbine faults is presented and verified. Two failure modes, blade icing accretion and gear cog belt fracture, are analyzed using SCADA data. A new index named ‘Comprehensive Index’ is defined to evaluate performance of different algorithms. Traditional machine learning algorithms do not perform well for data sets that are unbalanced and follow different distributions. The former causes bias in classification and the latter leads to poor adaptability of algorithms. A novel transfer learning algorithm studied in this paper, TrAdaBoost, has been proved to have superior performance on dealing with data imbalance and different distributions. A new approach to calibrate data labels using transfer learning algorithms is also proposed, which provides important insights into unsupervised learning for wind turbine fault diagnosis.
•A framework of using transfer learning algorithms on SCADA data for WT fault diagnosis is proposed.•A new index named ‘Comprehensive Index’ is defined to fairly evaluate performance of different algorithms.•TrAdaBoost shows its superior performance on dealing with data imbalance and different distributions.•A new framework to calibrate data labels is proposed, which can solve data improper labelling problem.•This paper provides important insights for unsupervised learning.
Details
- Title: Subtitle
- Diagnosis of wind turbine faults with transfer learning algorithms
- Creators
- Wanqiu Chen - University of Science and TechnologyYingning Qiu - University of Science and TechnologyYanhui Feng - University of Science and TechnologyYe Li - Shanghai Jiao Tong UniversityAndrew Kusiak - University of Iowa
- Resource Type
- Journal article
- Publication Details
- Renewable energy, Vol.163, pp.2053-2067
- Publisher
- Elsevier Ltd
- DOI
- 10.1016/j.renene.2020.10.121
- ISSN
- 0960-1481
- eISSN
- 1879-0682
- Grant note
- DOI: 10.13039/501100004608, name: Natural Science Foundation of Jiangsu Province
- Language
- English
- Date published
- 01/2021
- Academic Unit
- Nursing; Industrial and Systems Engineering
- Record Identifier
- 9984186969202771
Metrics
25 Record Views