当前位置>主页 > 期刊在线 > 通信工程 >

通信工程2019年2期

一种基于卷积神经网络和纵横交叉优化算法的电缆隧道温度异常识别方法
孟安波,杨跞,王伟,殷豪,曾云,黄圣权
(广东工业大学 自动化学院,广东 广州 510006)

摘  要:电力系统的安全稳定运行一直是我国电力行业的重中之重,而电缆隧道的安全也是电网安全运行的重要一环。本文使用电缆隧道巡检系统拍摄的图片,基于卷积神经网络(R-CNN)算法,在图像中定位异常状况点,并映射到红外图片。对电缆以及接头温度进行分析来及时对异常情况做出报警,可以维护供电安全并延长电缆使用寿命。针对电缆隧道巡检图像的时效性需求,采用纵横交叉(CSO)算法对图像分割的阈值优化,便于快速定位异常位置。


关键词:电缆隧道;温度异常;卷积神经网络;纵横交叉优化算法



中图分类号:TP391.41;TP183         文献标识码:A         文章编号:2096-4706(2019)02-0046-05


Recognition Method for Temperature Anomaly of Cable Tunnel Based on Convolution Neural Network and Cross Optimization Algorithm
MENG Anbo,YANG Luo,WANG Wei,YIN Hao,ZENG Yun,HUANG Shengquan
(Guangdong University of Technology,School of Automation,Guangzhou 510006,China)

Abstract:The safe and stable operation of the power system has always been the top priority of China’s power industry,and the safety of cable tunnels is also an important part of the safe operation of the power grid. In this paper,the picture taken by the cable tunnel inspection system is based on the convolutional neural network method,positioning the cable connector in the image and mapping to the infrared image. Analysis of the temperature of the cable connector in time to make an alarm on the abnormal  situation,you can maintain the safety of power supply and extend the service life of the cable. Aiming at the small number of sample images collected,a migration learning method is adopted,which reduces the training intensity and ensures a better positioning and recognition effect.

Keywords:cable tunnel;temperature anomaly;convolutional neural network;crossbar optimization algorithm


参考文献:

[1] 刘凯. 大连地区电缆隧道在线监测系统的设计与实现 [D].大连:大连理工大学,2016.

[2] 黄岩. 电力电缆接头的温度监测与预警研究 [J]. 时代农机,2015,42(10):35-36.

[3] 王龙阁,郭宏燕,陈磊,等. 一种基于分布式光纤光栅传感器的电缆温度监测系统 [J]. 电器工业,2016(12):74-76.

[4] LAWRENCE S,GILES C L,TSOI A C,et al. Face recognition:A convolutional neural-network approach [J]. IEEE Transactions on Neural Networks,1997,8(1):98-113.

[5] TURAGA C S,MURRAY F J,JAIN V,et al. Convolutional Networks Can Learn to Generate Affinity Graphs for Image Segmentation [J]. Neural Computation,2010,22(2):511-538.

[6] Dong C,Loy C C,He K M,et al. Image super-resolution using deep convolutional networks [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2016,38(2):295-307.

[7] Xu B,Wang N,Chen T,et al. Empirical evaluation of rectified activations in convolutional network [J]. arXiv preprint arXiv:1505.00853,2015.

[8] Hinton G E,Srivastava N,Krizhevsky A,et al. Improving neural networks by preventing co-adaptation of feature detectors [J]. arXiv preprint arXiv:1207.0580,2012.

[9] 陈庆,闫斌,叶润,等. 航拍绝缘子卷积神经网络检测及自爆识别研究 [J]. 电子测量与仪器学报,2017,31(6):942-953.

[10] 孟安波,胡函武,刘向东. 基于纵横交叉算法优化神经网络的负荷预测模型 [J]. 电力系统保护与控制,2016,44(7):102-106.

[11] 黄心汉,苏豪,彭刚,等. 基于卷积神经网络的目标识别及姿态检测 [J]. 华中科技大学学报(自然科学版),2017,45(10):7-11.

[12] Ren S,He K,Girshick R B,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.

[13] Szegedy C,Liu W,Jia Y,et al. & Rabinovich,A. Going deeper with convolutions [C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2015:1-9.


作者简介:

孟安波(1971-),男,汉族,重庆人,博士,教授,主要研究方向:人工智能在电力市场与电力系统中的应用;

讯作者:

杨跞(1993-),男,汉族,河南洛阳人,硕士研究生,主要研究方向:人工智能在电力系统中的应用;

王伟(1988-),男,汉族,河南洛阳人,硕士研究生,主要研究方向:电力系统负荷预测;

殷豪(1972-),女,汉族,重庆人,副教授,主要研究方向:人工智能在电力系统中的应用;

曾云(1994-),女,汉族,湖北荆门人,硕士研究生,主要研究方向:电力市场电价预测;

黄圣权(1990-),男,汉族,广东阳江人,硕士研究生,主要研究方向:风力发电风速预测。