针对目前钢丝绳断丝定量检测的问题，利用深度卷积神经网络强大的特征提取能力，提出一种基于迁移学习的钢丝绳断丝定量识别方法。通过连续小波变换将原始断丝漏磁信号转换成时频图。将预训练网络GoogLeNet的低层参数直接迁移，使用标记好的时频图对网络高层进行参数调整，得到最终的目标模型。通过内外部断丝实验验证了所提出的定量识别模型的效果，将传统的BP(back propagation)神经网络与所提出的方法进行对比。结果表明：基于迁移学习的断丝定量识别方法能准确区分钢丝绳的内外部断丝故障，分类准确率达到了97.2%；与传统BP神经网络相比，所提出的方法对各种断丝具有更好的识别性能。
Abstract
To address the current problems of quantitative testing of broken wires for steel wire ropes, using the powerful feature extraction ability of deep convolution neural network, a quantitative identification method based on transfer learning is proposed. The original magnetic flux leakage signals are converted into time-frequency images by continuous wavelet transform. The low-level parameters of the pre-trained network GoogLeNet are directly transferred, and the labeled time-frequency images are used to adjust the parameters in the high level of the network to obtain the final target model. The effectiveness of the proposed quantitative recognition model is verified by the experiments of internal and external broken wires. The traditional BP(back propagation) neural network is used as a comparison. The results show that the quantitative identification method based on transfer learning can accurately classify the internal and external broken wires of the wire rope with an accuracy rate of 97.2%. Compared with the traditional BP neural network, the proposed method has better recognition performance for various broken wires.
关键词
钢丝绳 /
断丝 /
迁移学习 /
定量识别
{{custom_keyword}} /
Key words
steel wire rope /
broken wire /
transfer learning /
quantitative identification
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1]谭继文．钢丝绳安全检测原理与技术[M]．北京：科学出版社，2009．
TAN Jiwen. Safety detection principle and technology of wire rope[M]. Beijing: Science Press, 2009.
[2]杨叔子．钢丝绳电磁无损检测[M]．北京：机械工业出版社，2017．
YANG Shuzi. Electromagnetic nondestructive testing of steel wire rope[M]. Beijing: China Machine Press, 2017.
[3]田志勇，张耀，谭继文．基于BP神经网络的钢丝绳断丝定量检测[J]．煤炭学报，2006(02)：245-249．
TIAN Zhiyong, ZHANG Yao, TAN Jiwen. Quantative test of broken wire for steel rope based on the back-propagation artificial neural networks[J]. Journal of China Coal Society, 2006(02): 245-249.
[4] Kim J W, Park S. Magnetic flux leakage sensing and artificial neural network pattern recognition-based automated damage detection and quantification for wire rope non-destructive evaluation[J]. Sensors, 2018, 18(1): 109.
[5]刘志怀，秦芳，刘娜，等．基于主成分分析和BP神经网络的钢丝绳断丝定量检测方法[J]．振动与冲击，2018，37(18)：276-281．
LIU Zhihuai, QIN Fang, LIU Na, et al. Quantitative testing method for broken wire in steel rope based on principal component analysis and BP artificial neural network model[J]. Journal of Vibration and Shock, 2018, 37(18): 276-281.
[6]朱良，谭继文，张义清．钢丝绳断丝损伤信号特征量提取方法研究[J]．煤矿机械，2019，40(02)：45-47．
ZHU Liang, TAN Jiwen, ZHANG Yiqing. Study on extracting method of damage signal characteristic of wire rope broken wire[J]. Coal Mine Machinery, 2019, 40(02): 45-47.
[7]王应晨，段修生．深度学习融合模型在机械故障诊断中的应用[J]．振动，测试与诊断，2019，039(006)：1271-1276．
WANG Yingchen, DUAN Xiusheng. Improved Deep Learning Fusion Model in Fault Diagnosis[J]. Journal of Vibration,Measurement & Diagnosis, 2019, 039(006): 1271-1276.
[8] Jing L., Zhao M., Li P., et al. A convolutional neural network based feature learning and fault diagnosis method for the condition monitoring of gearbox[J]. Measurement, 2017, 111: 1-10.
[9] Guo X., Chen L., Shen C. Hierarchical adaptive deep convolution neural network and its application to bearing fault diagnosis[J]. Measurement, 2016, 93: 490-502.
[10]鄢仁武，林穿，高硕勋，等．基于小波时频图和卷积神经网络的断路器故障诊断分析[J]．振动与冲击，2020，39(10)：198-205．
YAN Renwu, LIN Chuan, GAO Shuoxun, et al. Fault diagnosis and analysis of circuit breaker based on wavelet time-frequency representations and convolution neural network[J]. Journal of Vibration and Shock, 2020, 39(10): 198-205.
[11]曹继平，王赛，岳小丹，等．基于自适应深度卷积神经网络的发射车滚动轴承故障诊断研究[J]．振动与冲击，2020，39(5)：97-104+149．
CAO Jiping, WANG Sai, YUE Xiaodan, et al. Rolling bearing fault diagnosis of launch vehicle based on adaptive deep CNN[J]. Journal of Vibration and Shock, 2020, 39(5): 97-104+149.
[12] SHI G. Visualization and pruning of SSD with the base network VGG16[C]∥International Conference on Deep Learning Technologies, 2017: 90-94.
[13] Rezende E., Ruppert G., Carvalho T., et al. Malicious Software Classification Using Transfer Learning of ResNet-50 Deep Neural Network[C]// 2017 16th IEEE International Conference on Machine Learning and Applications (ICMLA). IEEE, 2018.
[14] S. J. Pan and Q. Yang. A survey on transfer learning[J]. IEEE Transactions on Knowledge and Data Engineering, 2010, 22(10): 1345–1359.
[15] Siyu S., Stephen M.A., Ruqiang Y., et al. Highly Accurate Machine Fault Diagnosis Using Deep Transfer Learning[J]. IEEE Transactions on Industrial Informatics, 2019:1-1.
[16] Yan R., Gao R.X., Chen X. Wavelets for fault diagnosis of rotary machines: A review with applications[J]. Signal Processing, 2014, 96: 1–15.
[17]康守强，胡明武，王玉静，等．基于特征迁移学习的变工况下滚动轴承故障诊断方法[J]．中国电机工程学报，2019，39(3)：764-772．
KANG Shouqiang, HU Mingwu, WANG Yujing, et al. Fault Diagnosis Method of a Rolling Bearing Under Variable Working Conditions Based on Feature Transfer Learning[J]. Proceedings of the CSEE, 2019, 39(3): 764-772.
[18] Szegedy C., Liu W., Jia Y., et al. Going Deeper with Convolutions[C]// IEEE Conference on Computer Vision and Pattern Recognition. Boston:IEEE, 2014.
[19] Zhang Y, Jing L, Xu W, et al. A Sensor for Broken Wire Detection of Steel Wire Ropes Based on the Magnetic Concentrating Principle[J]. Sensors, 2019, 19(17): 3763.
[20] Chen Z. Q., Li C., Sanchez R.V. Gearbox Fault Identification and Classification with Convolutional Neural Networks[J]. Shock and Vibration, 2015, 2015(PT.5): 1-10.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
PDF(1506 KB)
