Cloud-Edge-End Cooperative Detection of Wind Turbine Blade Surface Damage Based on Lightweight Deep Learning Network

被引：8

作者：

Liu, Yajuan ^{[1
]}

Wang, Zhen ^{[1
]}

Wu, Xiaolun ^{[2
]}

Fang, Fang ^{[1
]}

Saqlain, Ali Syed ^{[1
]}

机构：

[1] North China Elect Power Univ, Sch Control & Comp Engn, Beijing 102206, Peoples R China

[2] NR Elect Co Ltd, Res Inst, Nanjing 211102, Peoples R China

来源：

IEEE INTERNET COMPUTING | 2023年 / 27卷 / 01期

关键词：

Blades; Image edge detection; Cloud computing; Servers; Deep learning; Autonomous aerial vehicles; Computational modeling;

D O I：

10.1109/MIC.2022.3175935

中图分类号：

TP31 [计算机软件];

学科分类号：

081202 ; 0835 ;

摘要：

Blade health is directly related to the safety and efficiency of wind turbine (WT) operation. In this article, a cloud-edge-end collaborative detection method for WT blade surface damage is proposed based on lightweight deep learning network. The blade images are obtained by unmanned aerial vehicle. The YOLOv3 is optimized on the cloud server, including backbone network replacement, filter pruning, and knowledge distillation. After model training, the lightweight deep learning model YOLOv3-Mobilenet-PK is obtained and deployed on edge device to detect the surface damage of the WT blades, then the detection results can be viewed through the portable mobile device. The results show that the mean average precision (mAP) of the detection method proposed in this article is over 90%, the detection speed is about two times that of the YOLOv3-DarkNet53. This method has the advantages of fast detection speed, high accuracy, and less occupation of bandwidth.

引用

页码：43 / 51

页数：9

共 15 条

[1] SMD LED chips defect detection using a YOLOV3-dense model
Chen, Ssu-Han
Tsai, Chia-Chun
[J]. ADVANCED ENGINEERING INFORMATICS, 2021, 47
[2] Damage detection techniques for wind turbine blades: A review
Du, Ying
Zhou, Shengxi
Jing, Xingjian
Peng, Yeping
Wu, Hongkun
Kwok, Ngaiming
[J]. MECHANICAL SYSTEMS AND SIGNAL PROCESSING, 2020, 141 (141)
[3] Howard AG, 2017, Arxiv, DOI [arXiv:1704.04861, DOI 10.48550/ARXIV.1704.04861, 10.48550/arXiv.1704.04861]
[4] Ice detection using thermal infrared radiometry on wind turbine blades
Gomez Munoz, Carlos Quiterio
Garcia Marquez, Fausto Pedro
Sanchez Tomas, Juan Manuel
[J]. MEASUREMENT, 2016, 93 : 157 - 163
[5] Deep Residual Learning for Image Recognition
He, Kaiming
Zhang, Xiangyu
Ren, Shaoqing
Sun, Jian
[J]. 2016 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2016, : 770 - 778
[6] Filter Pruning via Geometric Median for Deep Convolutional Neural Networks Acceleration
He, Yang
Liu, Ping
Wang, Ziwei
Hu, Zhilan
Yang, Yi
[J]. 2019 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2019), 2019, : 4335 - 4344
[7] Hinton G., 2015, COMPUT SCI
[8] The Time Complexity Analysis of Neural Network Model Configurations
Lee, Rich
Chen, Ing-Yi
[J]. 2ND INTERNATIONAL CONFERENCE ON MATHEMATICS AND COMPUTERS IN SCIENCE AND ENGINEERING (MACISE 2020), 2020, : 178 - 183
[9] Toward Compact ConvNets via Structure-Sparsity Regularized Filter Pruning
Lin, Shaohui
Ji, Rongrong
Li, Yuchao
Deng, Cheng
Li, Xuelong
[J]. IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2020, 31 (02) : 574 - 588
[10] Emotion Recognition for Cognitive Edge Computing Using Deep Learning
Muhammad, Ghulam
Hossain, M. Shamim
[J]. IEEE INTERNET OF THINGS JOURNAL, 2021, 8 (23) : 16894 - 16901

← 1 2 →