Journal of Modern Power Systems and Clean Energy

ISSN 2196-5625 CN 32-1884/TK

+Advanced Search
Home > Archive>Volume 9, Issue 1, 2021 >210-218. DOI:10.35833/MPCE.2018.000672
PDF HTML Export
Automatic Discontinuity Classification of Wind-turbine Blades Using A-scan-based Convolutional Neural Network
Author:
Affiliation:

1.Korea Electronics Technology Institute (KETI), 401-402, Bucheon Technopark, 655, Pyeongcheon-ro, Wonmi-gu, Bucheon-si, Gyeonggi-do 14502, Korea;2.Sam Yong Inspection Engineering Co. Ltd., New Black Seok Building, 19, Dasan-ro 11-gil, Jung-gu, Seoul 04598, Korea

Fund Project:

This work was supported by the Korea Institute of Energy Technology Evaluation and Planning (KETEP) “Development of System for Damage Detection on the Outer of Fibrous Composite Blade for Wind Power Plants In-process and In-service Inspection” (No. 20153030024070), funded by the Ministry of Trade, Industry, and Energy (MOTIE), Korea.

  • Article
    • |
  • Figures
    • |
  • Metrics
    • |
  • Reference
    • |
  • Related
    • |
  • Cited by
    • |
    Abstract:

    Recent development trends in wind power generation have increased the importance of the safe operation of wind-turbine blades (WTBs). To realize this objective, it is essential to inspect WTBs for any defects before they are placed into operation. However, conventional methods of fault inspection in WTBs can be rather difficult to implement, since complex curvatures that characterize the WTB structures must ensure accurate and reliable inspection. Moreover, it is considered useful if inspection results can be objectively and consistently classified and analyzed by an automated system and not by the subjective judgment of an inspector. To address this concern, the construction of a pressure- and shape-adaptive phased-array ultrasonic testing platform, which is controlled by a nanoengine operation system to inspect WTBs for internal defects, has been presented in this paper. An automatic classifier has been designed to detect discontinuities in WTBs by using an A-scan-imaging-based convolutional neural network (CNN). The proposed CNN classifier design demonstrates a classification accuracy of nearly 99%. Results of the study demonstrate that the proposed CNN classifier is capable of automatically classifying the discontinuities of WTB with high accuracy, all of which could be considered as defect candidates.

    表 2 Table 2
    表 1 Table 1
    表 4 Table 4
    图1 Testbed with sizes and locations of artificial defects. (a) Overview. (b) Side view of testbed and delamination.Fig.1
    图2 Defect inspection system with PAUT platform.Fig.2
    图3 Discontinuity classification indication levels.Fig.3
    图4 Example of PAUT result image. (a) Example of observed defects. (b) Example of defect-free detections.Fig.4
    图5 Architecture of a discontinuity classifier.Fig.5
    图6 Actual defect-free detection results predicted as discontinuities. (a) Result image 1. (b) Result image 2. (c) Result image 3. (d) Result image 4. (e) Result image 5. (f) Result image 6.Fig.6
    图7 Actual discontinuity results predicted as defect-free detections. (a) With small-amplitude defects predicted as defect-free detections. (b) With large-amplitude defects predicted as defect-free detections.Fig.7
    表 3 Table 3
    Reference
    Related
    Cited by
Get Citation

Copy
Share
Article Metrics
  • Abstract:
  • PDF:
  • HTML:
  • Cited by:
History
  • Received:October 10,2018
  • Online: January 22,2021

Address:No.19 Chengxin Avenue, Jiangning District, Nanjing 211106, China

E-mail: mpce@alljournals.cn

Tel:86-25-81093060

Fax:86-25-81093040

Copyright：Journal of Modern Power Systems and Clean Energy ® 2025 All Rights Reserved
Supported by：Beijing E-Tiller Technology Development Co., Ltd.

ICP：ICP备09008660号