基于优化YOLOv4的主要电气设备智能检测及调参策略

doi:10.19595/j.cnki.1000-6753.tces.201595

Abstract
Figure/Table
References
Related Citation (1)

Download: PDF (7951 KB) HTML (1 KB)
Export: BibTeX | EndNote (RIS)

Abstract Multi-spectral imaging method based on UAV and robot is the development trend of high voltage equipment inspection, and the identification of main electrical equipment is the basis of intelligent diagnosis of its insulation state. In this paper, the train and test database of insulator, griding ring,vibration damper, transformer bushing and conductor was established. Based on YOLOv4, the Mosaic data augmentation method was optimized , which reduces the network loss by 0.7 and improves the precision to 84.3%. The influence of IoU algorithm based on bounding box regression on detection target of different scales was studied, and the training strategies of CIoU and GIoU were proposed for large and small targets respectivly. The influence of K-means and hierarchical algorithm on the clustering result of label width and high data of self-built database and on network performance is studied. Based on the training loss, precision and speed, the network parameters of YOLOv4 is optimized. The average loss of the improved model was reduced by 3% and the mean average precision increased by 0.8%. This study can be used to assess the operation state of electrical equipment in-site with multi-spectrum image.

Key words： Electric equipment detection YOLOv4 data augmente algrithm intersection ouer union K-means and hierarchical cluster

Received: 01 December 2020

PACS:

TM852

	Service

	E-mail this article
	Add to my bookshelf
	Add to citation manager
	E-mail Alert
	RSS
	Articles by authors

Cite this article:

URL:

https://dgjsxb.ces-transaction.com/EN/10.19595/j.cnki.1000-6753.tces.201595 OR https://dgjsxb.ces-transaction.com/EN/Y2021/V36/I22/4837

[1] 张汉花, 邹军, 田蓓, 等. 高空核爆电磁脉冲激励下架空输配电线路绝缘子闪络情况分析[J]. 电工技术学报, 2020, 35(2): 435-443.
Zhang Hanhua, Zou Jun, Tian Bei, et al.Flashover possibility analysis of overhead power transmission and distribution line insulators with the excitation of high altitude electromagnetic pulse[J]. Transactions of China Electrotechnical Society, 2020, 35(2): 435-443.
[2] 张倩, 王建平, 李帷韬. 基于反馈机制的卷积神经网络绝缘子状态检测方法[J]. 电工技术学报, 2019, 34(16): 3311-3321.
Zhang Qian, Wang Jianping, Li Weitao.Insulator state detection of convolutional neural networks based on feedback mechanism[J]. Transactions of China Electrotechnical Society, 2019, 34(16): 3311-3321.
[3] 徐建军, 黄立达, 闫丽梅, 等. 基于层次多任务深度学习的绝缘子自爆缺陷检测[J]. 电工技术学报, 2021, 36(7): 1407-1415.
Xu Jianjun, Huang Lida, Yan Limei, et al.Insulator self-explosion defect detection based on hierarchical multi-task deep learning[J]. Transactions of China Electrotechnical Society, 2021, 36(7): 1407-1415.
[4] Goodfellow I, Bengio Y, Courville A.Deep learning[M]. Cambridge: The MIT Press, 2016.
[5] Hinton G E, Salakhutdinov R R.Reducing the dimensionality of data with neural networks[J]. Science, 2006, 313(5786): 504-507.
[6] 李超然, 肖飞, 樊亚翔, 等. 基于门控循环单元神经网络和Huber-M估计鲁棒卡尔曼滤波融合方法的锂离子电池荷电状态估算方法[J]. 电工技术学报, 2020, 35(9): 2051-2062.
Li Chaoran, Xiao Fei, Fan Yaxiang, et al.A hybrid approach to lithium-ion battery soc estimation based on recurrent neural network with gated recurrent unit and Huber-M robust Kalman filter[J]. Transactions of China Electrotechnical Society, 2020, 35(9): 2051-2062.
[7] 李涛, 胡维昊, 李坚, 等. 基于深度强化学习算法的光伏-抽蓄互补系统智能调度[J]. 电工技术学报, 2020, 35(13): 2757-2768.
Li Tao, Hu Weihao, Li Jian, et al.Intelligent economic dispatch for PV-PHS integrated system: a deep reinforcement learning-based approach[J]. Transactions of China Electrotechnical Society, 2020, 35(13): 2757-2768.
[8] 丁禹, 尚学伟, 米为民. 基于深度学习的电网调控文本知识抽取方法[J]. 电力系统自动化, 2020, 44(24): 161-168.
Ding Yu, Shang Xuewei, Mi Weimin.Deep learning based knowledge extraction method for text of power grid dispatch and control[J]. Automation of Electric Power Systems, 2020, 44(24): 161-168.
[9] 杨维全, 朱元振, 刘玉田. 基于卷积神经网络的暂态电压稳定快速评估[J]. 电力系统自动化, 2019, 43(22): 46-52.
Yang Weiquan, Zhu Yuanzhen, Liu Yutian.Fastassessment of transient voltage stability based on convolutionalneural network[J]. Automation of Electric Power Systems, 2019, 43(22): 46-52.
[10] 田芳, 周孝信, 史东宇, 等. 基于卷积神经网络综合模型和稳态特征量的电力系统暂态稳定评估[J]. 中国电机工程学报, 2019, 39(14): 4025-4032.
Tian Fang, Zhou Xiaoxin, Shi Dongyu, et al.Power system transient stability assessment based on comprehensive convolutional neural network model and steady-state features[J]. Proceedings of the CSEE, 2019, 39(14): 4025-4032.
[11] 李文璞, 谢可, 廖逍, 等. 基于Faster RCNN变电设备红外图像缺陷识别方法[J]. 南方电网技术, 2019, 13(12): 79-84.
Li Wenpu, Xie Ke, Liao Xiao, et al.Intelligent diagnosis method of infrared image for transformer equipment based on improved Faster RCNN[J]. Southern power system technology, 2019, 13(12): 79-84.
[12] 蒋逸雯, 李黎, 李智威, 等.基于深度语义学习的电力变压器运维文本信息挖掘方法[J]. 中国电机工程学报, 2019, 39(14): 4162-4172.
Jiang Yiwen, Li Li, Li Zhiwei, et al.An information mining method of power transformer operation and maintenance texts based on deep semantic learning[J]. Proceedings of the CSEE, 2019, 39(14): 4162-4172.
[13] 张洋, 姬波, 卢红星, 等. 基于LSTM 神经网络的短期高压负荷电流预测方法[J]. 计算机科学, 2019, 46(6): 49-54.
Zhang Yang, Ji Bo, Lu Hongxing, et al.Short-term high voltage load current prediction method based on LSTM neural network[J]. Computer Science, 2019, 46(6): 49-54.
[14] Bochkovskiy A, Wang C Y, Liao H Y M. YOLOv4: optimal speed and accuracy of object detection[J] arXiv:2004.10934v1, 2020.
[15] 武建华, 梁利辉, 纪欣欣, 等. 基于YOLOv3算法的绝缘子红外图像故障检测方法[J]. 广东电力, 2020, 33(9): 77-84.
Wu Jianhua, Liang Lihui, Ji Xinxin, et al.Infrared image fault detection method for insulator based on YOLOv3 algorithm[J]. Guangdong Electric power, 2020, 33(9): 77-84.
[16] Liu Yunpeng, Ji Xinxin, Pei Shaotong, et al.Research on automatic location and recognition of insulators in substation based on YOLO-v3[J]. High Voltage, 2019, 5(1): DOI:10.1049/hve.2019.0091.
[17] Misra D.Mish: a self regularized non-monotonic neural activation function[J]. arXiv:1908.08681, 2019.
[18] 阿斯顿·张, 李沐. 动手学深度学习[M]. 北京:人民邮电出版社, 2019.
[19] Rezatofighi H, Tsoi N, Gwak J Y, et al.Generalized intersection over union: a metric and a loss for bounding box regression[C]//2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Long Beach, CA, USA, 2019: 658-666.
[20] Zheng Zhaohui, Wang Ping, Liu Wei, et al.Distance-IoU loss: faster and better learning for bounding box regression[J]. arXiv preprint arXiv: 1911.08287, 2019.
[21] Sangdoo Yun, Dongyoon Han, Seong Joon Oh, et al.CutMix: regularization strategy to train strong classifiers with localizable features[J]. arXiv:1905. 04899. 2019.
[22] DeVries T, Taylor G. Improved regularization of convolutional neural networks with cutout[J]. arXiv:1708.04552.2017.
[23] Zhang H, Cisse M, Dauphin Y N, et al. Mixup: beyond empirical risk minimization[J]. arxiv:1710. 09412v2. 2017.
[24] 李航. 统计学习方法[M]. 2版. 北京: 清华大学出版社, 2019.