Application of Deep Learning in Abnormal Insulator Infrared Image Diagnosis
-
摘要: 绝缘子的红外图像分析一般采用图像处理的方法,易受背景环境和数据量的影响,准确率和效率均较低,本文提出一种深度学习的异常诊断方法,基于改进的Faster R-CNN方法搭建检测网络,开展不同类型的绝缘子测试。研究结果表明:相对于神经网络(Back Propagation,BP)、Faster R-CNN方法,本文方法可高效地诊断出绝缘子的异常缺陷,平均检测精度达到90.2%;单Ⅰ型和Ⅴ型绝缘子的异常诊断准确率高于双Ⅰ型绝缘子。研究结果可为输电线路绝缘子异常诊断提供一定的参考。Abstract: Because of the effects of the background environment and data volume, the accuracy and efficiency of abnormal defects in traditional infrared images of insulators are generally low. In this study, a deep-learning anomaly diagnosis method is proposed. Based on the improved faster region-based convolutional neural network (R-CNN) method, a detection network is built to test different types of insulators. Results show that compared with the back propagation neural network and faster R-CNN methods, the proposed method can diagnose abnormal defects of insulators efficiently with a mean average precision of 90.2%. In addition, the diagnostic accuracy of single type Ⅰ and type Ⅴ insulators is higher than that of double type Ⅰ insulators. The results can provide a reference for insulator defect identification in transmission lines.
-
Key words:
- insulator /
- abnormal diagnosis /
- deep learning /
- Faster R-CNN /
- mAP /
- infrared image
-
表 1 软硬件配置
Table 1. Hardware and software configuration
Name Model Operating system Ubuntu 16.04.1 Database mysql 5.5.20 CPU Intel Xeon Silver 4114T 12C GPU NVIDIA GTX1080Ti Memory 32 G Hard disk 1 T Frame Detectron 表 2 样本配置信息
Table 2. Information of sample configuration
sample type training set verification set test set total positive 500 250 750 1500 negative 500 250 125 875 total 1000 500 875 2375 表 3 不同方法的实验结果统计
Table 3. Statistics of experimental results by different methods
Name Precision Recall mAP Time/s BP 93.5% 90.4% 80.3% 2.3 Faster R-CNN 98.7% 95.3% 88.7% 1.2 BFEM 99.2% 97.6% 90.2% 0.9 表 4 绝缘子异常诊断的准确率
Table 4. Accuracy of insulator anomaly diagnosis
Insulator type Abnormal total Detected number Accuracy Single Ⅰ 62 61 98.4% Double Ⅰ 47 44 93.6% Ⅴ 31 31 100.0% -
[1] 陈俊佑, 金立军, 段绍辉, 等.基于Hu不变矩的红外图像电力设备识别[J].机电工程, 2013, 30(1): 5-8. https://www.cnki.com.cn/Article/CJFDTOTAL-JDGC201301003.htmCHEN Junyou, JIN Lijun, DUAN Shaohui, et al. Power equipment identification in infrared image based on Hu invariant moments[J]. Journal of Mechanical & Electrical Engineering, 2013, 30(1): 5-8. https://www.cnki.com.cn/Article/CJFDTOTAL-JDGC201301003.htm [2] 邹辉, 黄福珍.基于改进Fast-Match算法的电力设备红外图像多目标定位[J].中国电机工程学报, 2017, 37(2): 591-598. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGDC201702027.htmZOU Hui, HU Fuzhen. Multi-target localization for infrared images of electrical equipment based on improved fast-match algorithm[J]. Proceedings of the CSEE, 2017, 37(2): 591-598. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGDC201702027.htm [3] 魏秀深.解析深度学习:卷积神经网络原理与视觉实践[M].北京:电子工业出版社, 2018.WEI Xiushen. Analytic Deep Learning: Convolutional Neural Network Theory And Visual Practice[M]. Beijing: Electronic Industry Press, 2018. [4] 罗舜.电力变压器套管将军帽发热故障的红外诊断分析[J].变压器, 2018, 55(1): 50-53. https://www.cnki.com.cn/Article/CJFDTOTAL-BYQZ201801018.htmLUO Sun. Infrared diagnosis analysis of power transformer bushing coupler heating[J]. Transformer, 2018, 55(1): 50-53. https://www.cnki.com.cn/Article/CJFDTOTAL-BYQZ201801018.htm [5] 张杰, 付泉泳, 袁野.变压器局部放电带电检测技术应用研究[J].变压器, 2018, 55(8): 66-71. https://www.cnki.com.cn/Article/CJFDTOTAL-BYQZ201808023.htmZHANG Jie, FU Quanyong, YUAN Ye. Application research of electric detection technology of partial discharge for transformer[J]. Transformer, 2018, 55(8): 66-71. https://www.cnki.com.cn/Article/CJFDTOTAL-BYQZ201808023.htm [6] 梁天明, 袁焯锋, 石延辉.高压交流滤波电容器局部过热诱因分析及预防[J].电力电容器与无功补偿, 2015, 36(6): 49-53. https://www.cnki.com.cn/Article/CJFDTOTAL-DLDY201506011.htmLIANG Tianming, YUAN Daofeng, SHI Yanhui. Cause analysis and preventions on local overheating of high voltage ac filter capacitor[J]. Power Capacitor & Reactive Power Compensation, 2015, 36(6): 49-53. https://www.cnki.com.cn/Article/CJFDTOTAL-DLDY201506011.htm [7] 潘臻, 安立.一起35 kV并联电容器组事故爆炸原因分析[J].电力电容器与无功补偿, 2015, 36(3): 17-20. https://www.cnki.com.cn/Article/CJFDTOTAL-DLDY201503005.htmPAN Zhen, AN Li. Analysis of 35 kV shunt capacitor banks explosion accident[J]. Power Capacitor & Reactive Power Compensation, 2015, 36(3): 17-20. https://www.cnki.com.cn/Article/CJFDTOTAL-DLDY201503005.htm [8] 黄斌, 李昊, 徐姗姗, 等.一起35 kV并联电容器组爆炸原因分析及防范措施[J].电力电容器与无功补偿, 2018, 39(1): 23-27. https://www.cnki.com.cn/Article/CJFDTOTAL-DLDY201801005.htmHUANG Bin, LI Hao, XU Sansan, et al. Reason analysis and precautionary measures for a 35kv shunt capacitor bank explosion[J]. Power Capacitor & Reactive Power Compensation, 2018, 39(1): 23-27. https://www.cnki.com.cn/Article/CJFDTOTAL-DLDY201801005.htm [9] 商俊平, 李储欣, 陈亮.基于视觉的绝缘子定位与自爆缺陷检测[J].电子测量与仪器学报, 2017, 31(6): 844-849. https://www.cnki.com.cn/Article/CJFDTOTAL-DZIY201706007.htmSHANG Junping, LI Chuxin, CHEN Liang. Location and detectionfor self-explode insulator based on vision[J]. Journal of Electronic Measurement and Instrumentation, 2017, 31(6): 844-849. https://www.cnki.com.cn/Article/CJFDTOTAL-DZIY201706007.htm [10] 沈新平, 彭刚, 袁志强.基于霍夫变换和RANSAC算法的绝缘子定位方法[J].电子测量技术, 2017, 40(6): 132-137. https://www.cnki.com.cn/Article/CJFDTOTAL-DZCL201706031.htmSHEN Xinping, PENG Gang, YUAN Zhiqiang. Insulator location method based on hough transformation and RANSAC algorithm[J]. Electronic Measurement Technology, 2017, 40(6): 132-137. https://www.cnki.com.cn/Article/CJFDTOTAL-DZCL201706031.htm [11] 李军锋, 王钦若, 李敏.结合深度学习和随机森林的电力设备图像识别[J].高电压技术, 2017, 43(11): 3705-3711. https://www.cnki.com.cn/Article/CJFDTOTAL-GDYJ201711028.htmLI Junfeng, WANG Qinruo, LI Min, et al. Electric Equipment Image Recognition Based on Deep Learning and Random Forest[J]. High Voltage Engineering, 2017, 43(11): 3705-3711. https://www.cnki.com.cn/Article/CJFDTOTAL-GDYJ201711028.htm [12] 侯春萍, 章衡光, 张巍, 等.输电线路绝缘子自爆缺陷识别方法[J].电力系统及其自动化学报, 2019, 31(6): 1-6.HOU Chunping, ZHANG Hengguang, ZHANG Wei, et al. Recognition method for faults of insulators on transmission lines[C]//Proceedings of the CSU-EPSA, 2019, 31(6): 1-6. [13] 左川.基于图像识别的输电线路绝缘子检测方法研究[D].北京: 华北电力大学, 2019.ZUO Chuang. Research on detection method of transmission line insulator based on image recognition[D]. Beijing: North China Electric Power University, 2019. [14] 杨光俊.卷积神经网络在电力设备红外图像识别中的应用研究[D].广州: 华南理工大学, 2019.YANG Guangjun. Research on the application of convolutional neural network in infrared image recognition of power equipment[D]. Guangzhou: South CHINA University of Technology, 2019. [15] 周可慧, 廖志伟, 肖异瑶, 等.基于改进CNN的电力设备红外图像分类模型构建研究[J].红外技术, 2019, 41(11): 1033-1038. https://www.cnki.com.cn/Article/CJFDTOTAL-HWJS201911007.htmZHOU Kehui, LIAO Zhiwei, XIAO Yiyao, et al. Construction of infrared image classification model for power equipments based on improved CNN[J]. Infrared Technology, 2019, 41(11): 1033-1038. https://www.cnki.com.cn/Article/CJFDTOTAL-HWJS201911007.htm [16] 许必宵.基于多尺度特征融合与上下文分析的目标检测技术研究[D].南京: 南京邮电大学, 2019.XU Bixiao. Research on object detection technology based on multi-scale feature fusion and context analysis[D]. Nanjing: Nanjing University of Posts and Telecommunications, 2019. [17] 张丹丹.基于航拍图像的绝缘子自爆位置的检测[D].成都: 西华大学, 2018.ZHANG Dandan. Detection of self-exploding position of insulator based on aerial image[D]. Chengdu: Xihua University, 2018. [18] 王梦.基于绝缘子图像的缺陷检测方法研究[D].武汉: 华中科技大学, 2019.WANG Meng. A thesis submitted in partial fulfillment of the requirements[D]. Wuhan: Huazhong University of Science & Technology, 2019. [19] 国家能源局.带电设备红外诊断应用规范: DL/T 664-2008[S].北京: 中国标准出版社, 2008.National Energy Administration. Application rules of infrared diagnosis for live electrical equipment: DL/T 664-2008[S]. Beijing: China Electric Power Press, 2008.