[1]ZHAO Wenqing,LIU Liang,HU Jiawei,et al.Detection of transformer oil leakage based on deep separable atrous convolution pyramid[J].CAAI Transactions on Intelligent Systems,2023,18(5):966-974.[doi:10.11992/tis.202212016]
Copy

Detection of transformer oil leakage based on deep separable atrous convolution pyramid

CAAI Transactions on Intelligent Systems[ISSN 1673-4785/CN 23-1538/TP] Volume: 18 Number of periods: 2023 5 Page number: 966-974 Column: 学术论文—机器学习 Public date: 2023-09-05
Title:
Detection of transformer oil leakage based on deep separable atrous convolution pyramid
Author(s):
ZHAO Wenqing1; 2;  LIU Liang1;  HU Jiawei1;  ZHAI Yongjie1;  ZHAO Zhenbing3
1. School of Control and Computer Engineering, North China Electric Power University, Baoding 071003, China;
2. Engineering Research Center of the Ministry of Education for Intelligent Computing of Complex Energy System, Baoding 071003, China;
3. School of Electrical and Electronic Engineering, North China Electric Power University, Baoding 071003, China
Keywords:
transformer; oil spill detection; semantic information; deep separable atrous convolution pyramid; low-order features; high-order features; feature fusion; attention mechanism
CLC:
TP391
DOI:
10.11992/tis.202212016
Abstract:
To improve the detection efficiency of the transformer oil leakage patrol inspection image, a deep separable atrous convolution pyramid-based transformer oil leakage detection model is suggested. First, the ordinary convolution block in the atrous pyramid is modified into a deep separable convolution block for expansion of the pyramid receptive field and further enrichment of the semantic information of the feature graph extracted by the feature extraction network. Afterward, the fusion of low-order and high-order semantic features in the feature extraction stage is improved for further enhancement of the semantic information of the feature graph generated by the feature extraction network. Finally, to avoid semantic information loss in the feature graph after several convolution and pooling operations, spatial attention and channel attention mechanisms are introduced into the fusion process to further enhance the semantic information in the feature graph. It is found by comparing with algorithms such as traditional UNet (Convolutional Networks for Biomedical Image Segmentation), PSPNet (Pyramid Scene Parsing Network), DeepLabv3 + (Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation), and MCNN (Multiclass Convolutional Neural Network) via tests that the proposed network detection model is effective, with 76.85% precision, 64.63% average cross-merger ratio, 73.56% recall rate, and 30 frames per second. To confirm the effectiveness of the proposed method, an ablation experiment is designed. Compared with the basic network model, the precision, average intersection ratio, and recall rate are increased by 9.33%, 7.15%, and 5.66%, respectively.
References:
[1] RAJOTTE C. Guide for transformer maintenance[R]. Paris: Cigre Working Group A2.34, 2011.
[2] Power transformers - Part 7: Loading guide for oil-immersed power transformers: IEC 60076-7 Ed. 1.0 b: 2005[S]. International Electrotechnical Commission [iec], 2005.
[3] 翟永杰, 杨旭, 赵振兵, 等. 融合共现推理的Faster R-CNN输电线路金具检测[J]. 智能系统学报, 2021, 16(2): 237-246
ZHAI Yongjie, YANG Xu, ZHAO Zhenbing, et al. Integrating co-occurrence reasoning for Faster R-CNN transmission line fitting detection[J]. CAAI transactions on intelligent systems, 2021, 16(2): 237-246
[4] 苏丽, 孙雨鑫, 苑守正. 基于深度学习的实例分割研究综述[J]. 智能系统学报, 2022, 17(1): 16-31
SU Li, SUN Yuxin, YUAN Shouzheng. A survey of instance segmentation research based on deep learning[J]. CAAI transactions on intelligent systems, 2022, 17(1): 16-31
[5] 马岽奡, 唐娉, 赵理君, 等. 深度学习图像数据增广方法研究综述[J]. 中国图象图形学报, 2021, 26(3): 487-502
MA Dongao, TANG Ping, ZHAO Lijun, et al. Review of data augmentation for image in deep learning[J]. Journal of image and graphics, 2021, 26(3): 487-502
[6] 赵振兵, 张薇, 翟永杰, 等. 电力视觉技术的概念、研究现状与展望[J]. 电力科学与工程, 2020, 36(1): 1-8
ZHAO Zhenbing, ZHANG Wei, ZHAI Yongjie, et al. Concept, research status and prospect of electric power vision technology[J]. Electric power science and engineering, 2020, 36(1): 1-8
[7] REDMON J, DIVVALA S, GIRSHICK R, et al. You only look once: unified, real-time object detection[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas: IEEE, 2016: 779?788.
[8] REDMON J, FARHADI A. YOLO9000: better, faster, stronger[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition. Honolulu: IEEE, 2017: 6517?6525.
[9] FARHADI A, REDMON J. YOLOv3: an incremental improvement[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Salt Lake City: IEEE, 2018: 1804?2767.
[10] BOCHKOVSKIY A, WANG C Y, LIAO H Y M. YOLOv4: optimal speed and accuracy of object detection[EB/OL]. (2020?04?23)[2023?03?29]. https://arxiv.org/abs/2004.10934.
[11] WANG C Y, BOCHKOVSKIY A, LIAO H Y M. YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors[EB/OL]. (2022?07?06)[2023?03?29]. https://arxiv.org/abs/2207.02696.
[12] LIU Wei, ANGUELOV D, ERHAN D, et al. SSD: single shot MultiBox detector[C]//European Conference on Computer Vision. Cham: Springer, 2016: 21?37.
[13] GIRSHICK R, DONAHUE J, DARRELL T, et al. Rich feature hierarchies for accurate object detection and semantic segmentation[C]//2014 IEEE Conference on Computer Vision and Pattern Recognition. Columbus: IEEE, 2014: 580?587.
[14] REN Shaoqing, HE Kaiming, GIRSHICK R, et al. Faster R-CNN: towards real-time object detection with region proposal networks[J]. IEEE transactions on pattern analysis and machine intelligence, 2017, 39(6): 1137-1149.
[15] LI Lu, ICHIMURA S, MORIYAMA T, et al. A system to detect small amounts of oil leakage with oil visualization for transformers using fluorescence recognition[J]. IEEE transactions on dielectrics and electrical insulation, 2017, 24(2): 1249-1255.
[16] LI Anqi, YE Dongxu, LYU Erli, et al. RGB-thermal fusion network for leakage detection of crude oil transmission pipes[C]//2019 IEEE International Conference on Robotics and Biomimetics. Dali: IEEE, 2020: 883?888.
[17] LI Lu, ICHIMURA S, YAMAGISHI A, et al. Oil film detection under solar irradiation and image processing[J]. IEEE sensors journal, 2020, 20(6): 3070-3077.
[18] 鲍伟超, 顾理, 何劲松, 等. 基于循环训练法的变压器漏油检测[J]. 计算机辅助设计与图形学学报, 2021, 33(3): 431-438
BAO Weichao, GU Li, HE Jinsong, et al. Transformer oil leakage detection based on loop training method[J]. Journal of computer-aided design & computer graphics, 2021, 33(3): 431-438
[19] GHORBANI Z, BEHZADAN A H. Monitoring offshore oil pollution using multi-class convolutional neural networks[J]. Environmental pollution, 2021, 289: 117884.
[20] WANG Feng, ZHONG Zhen, WANG Guang, et al. A penalized convolution model for oil leakage detection in electrohydraulic railway point systems[J]. IEEE transactions on instrumentation and measurement, 2021, 70: 1-9.
[21] CHEN Xu, LIU Lei, HUANG Wei. The detection and prediction for oil spill on the sea based on the infrared images[J]. Infrared physics & technology, 2016, 77: 391-404.
[22] LI Bing, WANG Tian, HU Zhedong, et al. Two-level model for detecting substation defects from infrared images[J]. Sensors, 2022, 22(18): 6861.
[23] CHEN L C, ZHU Yukun, PAPANDREOU G, et al. Encoder-decoder with atrous separable convolution for semantic image segmentation[C]//European Conference on Computer Vision. Cham: Springer, 2018: 833?851.
[24] CHOLLET F. Xception: deep learning with depthwise separable convolutions[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition. Honolulu: IEEE, 2017: 1800?1807.
[25] CHEN L C, GEORGE P, FLORIAN S, etal. Rethinking atrous convolution for semantic image segmentation[C]// Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition. Honolulu: IEEE, 2017: 1601?1614.
[26] WOO S, PARK J, LEE J Y, et al. CBAM: convolutional block attention module[C]//European Conference on Computer Vision. Cham: Springer, 2018: 3?19.
Similar References:

Memo

-

Last Update: 1900-01-01

Copyright © CAAI Transactions on Intelligent Systems