基于迁移学习卷积神经网络的电缆隧道锈蚀识别算法
|
摘要
随着电力行业的不断发展,高压电缆的铺排以及地下电缆隧道的建设与维护逐渐成为该领域中的热点问题之一。将迁移学习的核心思想与经典的卷积神经网络(LeNet5)相结合,提出了一种基于迁移学习卷积神经网络的电缆隧道锈蚀识别算法,实现了电缆隧道内部电源箱、风机等设备的锈蚀识别。该算法基于Tensorflow框架,能够有效地解决训练样本不足、训练时间冗长以及识别精度不高的问题。通过引入4种经典的目标识别算法进行对比实验,进一步验证了所提方案在网络训练时间以及识别精确度上的优势,为后续电缆隧道巡检机器人系统的构建提供了坚实的理论基础与实验支撑。
With the continuous development of the power industry, the laying of high-voltage cables and the construction and maintenance of underground cable tunnels have gradually become one of the hot issues in this field. This paper proposes a cable tunnel rust recognition method based on the integration of transfer learning and the classical convolutional neural network(LeNet5),which is able to realize the specific rust recognition of some electrical devices, such as the internal power box, fan and other equipment. The whole recognition process is based on the Tensorflow framework, and is able to solve such problems as insufficient training samples, long training time, and low recognition accuracy. Furthermore, by comparing with four classical target recognition algorithms, the proposed scheme is proved to be superior in terms of training time and recognition accuracy. The entire scheme provides a solid theoretical basis and experimental support for the realization of subsequent cable tunnel inspection robot system.
With the continuous development of the power industry, the laying of high-voltage cables and the construction and maintenance of underground cable tunnels have gradually become one of the hot issues in this field. This paper proposes a cable tunnel rust recognition method based on the integration of transfer learning and the classical convolutional neural network(LeNet5),which is able to realize the specific rust recognition of some electrical devices, such as the internal power box, fan and other equipment. The whole recognition process is based on the Tensorflow framework, and is able to solve such problems as insufficient training samples, long training time, and low recognition accuracy. Furthermore, by comparing with four classical target recognition algorithms, the proposed scheme is proved to be superior in terms of training time and recognition accuracy. The entire scheme provides a solid theoretical basis and experimental support for the realization of subsequent cable tunnel inspection robot system.
引文
[1]丁文其,袁森林,高小庆,等.电力隧道超大直径顶管施工扰动特性研究[J].岩土力学,2010, 31(9):2901-2906.DING Wenqi, YUAN Senlin, GAO Xiaoqing, et al. Research on construction disturbance characteristics caused by super large diameter pipe jacking in electric power tunnel[J]. Rock and Soil Mechanics, 2010, 31(9):2901-2906.
[2]李俊光,黄鑫,杨小礼.城市隧道下穿电缆隧道时的数值计算分析及变形沉降预测[J].铁道科学与工程学报,2008, 5(1):68-71.LI Junguang, HUANG Xin, YANG Xiaoli. Numerical simulation and settlement prediction of subway locating under cable tunnels[J].Journal of Railway Science and Engineering, 2008, 5(1):68-71.
[3]吴春荣,黄鑫,李海峰.细水雾灭火系统在电缆隧道中的应用研究[J].消防科学与技术,2008, 27(9):662-665.WU Chunrong, HUANG Xin, LI Haifeng. Study on application of mist extinguishing system in cable tunnel[J]. Fire Science and Technology,2008, 27(9):662-665.
[4] JULIER S J, UHIMANN J K. Unscented filtering and nonlinear estimation[J]. Proceedings of the IEEE, 2004, 92(3):401-422.
[5] JIANG B, SAMPLE A P, WISTORT R M, et al. Autonomous robotic monitoring of underground cable systems[C]//International Conference on Icar, IEEE, 2005:673-679.
[6]姜芸,付庄.一种小型电缆隧道检测机器人设计[J].华东电力,2009, 37(1):95-97.JIANG Yun, FU Zhuang. Design of pint-size cable tunnel inspecting robots[J]. East China Electric Power, 2009, 37(1):95-97.
[7]谢振宇,付庄,宋国强,等.基于行为的电缆隧道综合检测机器人控制系统研究[J].机械与电子,2008(4):47-51.XIE Zhenyu, FU Zhuang, SONG Guoqiang, et al. Control system research of behavior-based cable tunnel inspection robot[J].Machinery&Electronics, 2008(4):47-51.
[8]冯凌,杨华夏,魏东.分体式电缆隧道检测机器人的移动结构研究[J].自动化仪表,2017, 38(6):46-50.FENG Ling, YANG Huaxia, WEI Dong. Research on the moving structure of split cable tunnel inspection robot[J]. Process Automation Instrumentation, 2017, 38(6):46-50.
[9] DONG Z, LAI C S, QI D, et al. A general memristor-based pulse coupled neural network with variable linking coefficient for multifocus image fusion[J]. Neurocomputing, 2018, 308:172-183.
[10]张晴晴,刘勇,潘接林,等.基于卷积神经网络的连续语音识别[J].工程科学学报,2015, 37(9):1212-1217.ZHANG Qingqing, LIU Yong, PAN Jielin, et al. Continuous speech recognition by convolutional neural net works[J]. Chinese Journal of Engineering,2015, 37(9):1212-1217.
[11] HUBEL D H, WIESEL T N. Receptive fields, binocular interaction and functional architecture in the cat's visual cortex[J]. The Journal of Physiology,1962, 160:106-154.
[12] LECUN Y, BOTTOU L, BENGIO Y, et al. Gradient-based learning applied to document recognition[J]. Proceedings of the IEEE, 1998,86(11):2278-2324.
[13] ZHAO Z H, YANG S P, MA Z Q. License plate character recognition based on convolutional neural network LeNet-5[J].Journal of System Simulation, 2010, 22(3):638-641.
[14] CIRESAN D C, MEIER U, GAMBARDELLA L M, et al.Convolutional neural network committees for handwritten character classification[C]//International Conference on Document Analysis and Recognition, IEEE, 2011:1135-1139.
[15] YUAN Z W, ZHANG J. Feature extraction and image retrieval based on AlexNet[C]//Eighth International Conference on Digital Image Processing, 2016.
[16] TSOUPOS A, KHADKIKAR V M. A novel SVM technique with enhanced output voltage quality for indirect matrix converters[J].IEEE Transactions on Industrial Electronics, 2019, 66(2):832-841.
[17] XIAO Z, YE S J, ZHONG B, et al. BP neural network with rough set for short term load forecasting[J]. Expert Systems with Applications,2009, 36(1):273-279.
[18] HUANG G B, ZHU Q Y, SIEW C K. Extreme learning machine:theory and applications[J]. Neurocomputing, 2006, 70(1-3):489-501.
[19]逯彦秋,安关峰,程进.基于主动导波的钢筋锈蚀识别技术[J].北京工业大学学报,2014, 40(6):865-871.LU Yanqiu, AN Guanfeng, CHENG Jin. Technology of steel corrosion detection based on the guided wave testing[J]. Journal of Beijing University of Technology, 2014, 40(6):865-871.
[20]赵新,易伟建,徐圣.钢筋混凝土梁锈蚀损伤定位与识别[J].振动与冲击,2007, 26(8):35-38.ZHAO Xin, YI Weijian, XU Sheng. Location and identification of corrosion of reinforced concrete beams[J]. Journal of Vibration and Shock,2007, 26(8):35-38.
[21] PAN S J, YANG Q. A survey on transfer learning[J]. IEEE Transactions on Knowledge and Data Engineering, 2010, 22(10):1345-1359.
[22] ELFWING S, UCHIBE E, DOYA K. Sigmoid-weighted linear units for neural network function approximation in reinforcement learning[J]. Neural Networks, 2018, 107:3-11.
[2]李俊光,黄鑫,杨小礼.城市隧道下穿电缆隧道时的数值计算分析及变形沉降预测[J].铁道科学与工程学报,2008, 5(1):68-71.LI Junguang, HUANG Xin, YANG Xiaoli. Numerical simulation and settlement prediction of subway locating under cable tunnels[J].Journal of Railway Science and Engineering, 2008, 5(1):68-71.
[3]吴春荣,黄鑫,李海峰.细水雾灭火系统在电缆隧道中的应用研究[J].消防科学与技术,2008, 27(9):662-665.WU Chunrong, HUANG Xin, LI Haifeng. Study on application of mist extinguishing system in cable tunnel[J]. Fire Science and Technology,2008, 27(9):662-665.
[4] JULIER S J, UHIMANN J K. Unscented filtering and nonlinear estimation[J]. Proceedings of the IEEE, 2004, 92(3):401-422.
[5] JIANG B, SAMPLE A P, WISTORT R M, et al. Autonomous robotic monitoring of underground cable systems[C]//International Conference on Icar, IEEE, 2005:673-679.
[6]姜芸,付庄.一种小型电缆隧道检测机器人设计[J].华东电力,2009, 37(1):95-97.JIANG Yun, FU Zhuang. Design of pint-size cable tunnel inspecting robots[J]. East China Electric Power, 2009, 37(1):95-97.
[7]谢振宇,付庄,宋国强,等.基于行为的电缆隧道综合检测机器人控制系统研究[J].机械与电子,2008(4):47-51.XIE Zhenyu, FU Zhuang, SONG Guoqiang, et al. Control system research of behavior-based cable tunnel inspection robot[J].Machinery&Electronics, 2008(4):47-51.
[8]冯凌,杨华夏,魏东.分体式电缆隧道检测机器人的移动结构研究[J].自动化仪表,2017, 38(6):46-50.FENG Ling, YANG Huaxia, WEI Dong. Research on the moving structure of split cable tunnel inspection robot[J]. Process Automation Instrumentation, 2017, 38(6):46-50.
[9] DONG Z, LAI C S, QI D, et al. A general memristor-based pulse coupled neural network with variable linking coefficient for multifocus image fusion[J]. Neurocomputing, 2018, 308:172-183.
[10]张晴晴,刘勇,潘接林,等.基于卷积神经网络的连续语音识别[J].工程科学学报,2015, 37(9):1212-1217.ZHANG Qingqing, LIU Yong, PAN Jielin, et al. Continuous speech recognition by convolutional neural net works[J]. Chinese Journal of Engineering,2015, 37(9):1212-1217.
[11] HUBEL D H, WIESEL T N. Receptive fields, binocular interaction and functional architecture in the cat's visual cortex[J]. The Journal of Physiology,1962, 160:106-154.
[12] LECUN Y, BOTTOU L, BENGIO Y, et al. Gradient-based learning applied to document recognition[J]. Proceedings of the IEEE, 1998,86(11):2278-2324.
[13] ZHAO Z H, YANG S P, MA Z Q. License plate character recognition based on convolutional neural network LeNet-5[J].Journal of System Simulation, 2010, 22(3):638-641.
[14] CIRESAN D C, MEIER U, GAMBARDELLA L M, et al.Convolutional neural network committees for handwritten character classification[C]//International Conference on Document Analysis and Recognition, IEEE, 2011:1135-1139.
[15] YUAN Z W, ZHANG J. Feature extraction and image retrieval based on AlexNet[C]//Eighth International Conference on Digital Image Processing, 2016.
[16] TSOUPOS A, KHADKIKAR V M. A novel SVM technique with enhanced output voltage quality for indirect matrix converters[J].IEEE Transactions on Industrial Electronics, 2019, 66(2):832-841.
[17] XIAO Z, YE S J, ZHONG B, et al. BP neural network with rough set for short term load forecasting[J]. Expert Systems with Applications,2009, 36(1):273-279.
[18] HUANG G B, ZHU Q Y, SIEW C K. Extreme learning machine:theory and applications[J]. Neurocomputing, 2006, 70(1-3):489-501.
[19]逯彦秋,安关峰,程进.基于主动导波的钢筋锈蚀识别技术[J].北京工业大学学报,2014, 40(6):865-871.LU Yanqiu, AN Guanfeng, CHENG Jin. Technology of steel corrosion detection based on the guided wave testing[J]. Journal of Beijing University of Technology, 2014, 40(6):865-871.
[20]赵新,易伟建,徐圣.钢筋混凝土梁锈蚀损伤定位与识别[J].振动与冲击,2007, 26(8):35-38.ZHAO Xin, YI Weijian, XU Sheng. Location and identification of corrosion of reinforced concrete beams[J]. Journal of Vibration and Shock,2007, 26(8):35-38.
[21] PAN S J, YANG Q. A survey on transfer learning[J]. IEEE Transactions on Knowledge and Data Engineering, 2010, 22(10):1345-1359.
[22] ELFWING S, UCHIBE E, DOYA K. Sigmoid-weighted linear units for neural network function approximation in reinforcement learning[J]. Neural Networks, 2018, 107:3-11.