Instrumentation & Measurement Magazine 25-2 - 76

Method
Table 3 - Performance comparison of different data augmentation methods
Training time (s)
Time domain data-DCGANs
Time frequency graph-DCGANs
Time frequency graph-ACGANs
signal diagnosis. The experiment shows the following: data
augmentation methods based on time-frequency diagrams
have a greater improvement in generation accuracy and
training speed than data augmentation methods based on
time-series signals; and the data augmentation method based
on ACGANs can greatly simplify the model training process
and generate new data with higher quality. The proposed
method also has good practicability in actual pipeline magnetic
flux leakage signal diagnosis and identification. After
applying this method, the accuracy rate of MFL recognition
has been improved significantly, from 95% to 99.5%. Additionally,
compared with traditional data augmentation methods,
this method can better improve model training efficiency and
MFL recognition accuracy.
Acknowledgment
This work was supported in part by the National Natural Science
Foundation of China under Grant 62073158, Natural
Science Foundation of Liaoning Province under Grant 2019BS-158,
the Scientific Research Funds of Liaoning Provincial
Department of Education under Grant L2020017, in part by
Funded by China Postdoctoral Science Foundation under
Grant 2020M670796, and in part by the Talent Scientific Research
under Grant 2019XJJL-008 into Supported by Talent
Scientific Research Fund of LSHU (No. 2019XJJL-008) of Liaoning
Petrochemical University.
References
[1] X. Lang et al., " Multiple leaks detection in fluid pipeline based on
improved VMD and TWSVM, " in Proc. 2020 IEEE 16th Int. Conf.
Control and Automation (ICCA), 2020.
[2] X. Lang and W. Yuan, " Localization method of multiple leaks
based on time-frequency analysis and improved differential
evolution, " IEEE Sensors J., vol. 20, no. 23, pp. 14383-14390, Dec.
2020.
[3] X. Lang, P. Li, Y. Guo, J. Cao and S. Lu, " A multiple leaks'
localization method in a pipeline based on change in the sound
velocity, " IEEE Trans. Instrum. Meas., vol. 69, no. 7, pp. 5010-5017,
Jul. 2020.
[4] Z. Wu, Y. Deng, J. Liu and L. Wang, " A reinforcement learningbased
reconstruction method for complex defect profiles in MFL
inspection, " IEEE Trans. Instrum. Meas., vol. 70, pp. 1-10, 2021.
[5] Y. Ren, J. Liu, J. Zhang, L. Jiang and Y. Luo, " A data reconstruction
method based on adversarial conditional variational
autoencoder, " in Proc. 2020 IEEE 9th Data Driven Control and
Learning Systems Conf. (DDCLS), pp. 622-626, 2020.
[6] M. Fu, J. Liu, H. Zhang and S. Lu, " Multisensor fusion for
magnetic flux leakage defect characterization under information
76
7890
2380
3210
Accuracy (%)
96
97.8
99.6
incompletion, " IEEE Trans. Ind. Electron., vol. 68, no. 5, pp.
4382-4392, May 2021.
[7] X. Hu, H. Zhang, D. Ma and R. Wang, " A tnGAN-based leak
detection method for pipeline network considering incomplete
sensor data, " IEEE Trans. Instrum. Meas., vol. 70, pp. 1-10, 2021.
[8] Y. Guo, Z. Pang, J. Du, F. Jiang and Q. Hu, " An improved AlexNet
for power edge transmission line anomaly detection, " IEEE
Access, vol. 8, pp. 97830-97838, 2020.
[9] A. Navarro, O. Begovich, J. A. Delgado-Aguiñaga and J. Sánchez,
" Real time leak isolation in pipelines based on a time delay neural
network, " in Proc. 2019 IEEE Int. Autumn Mtg. Power, Electronics
and Computing (ROPEC), pp. 1-6, 2019.
[10] M. Zhou, Z. Pan, Y. Liu, Q. Zhang, Y. Cai and H. Pan, " Leak
detection and location based on ISLMD and CNN in a pipeline, "
IEEE Access, vol. 7, pp. 30457-30464, 2019.
[11] P. Zhang, X. Wang, W. Zhang and J. Chen, " Learning spatial-
spectral-temporal EEG features with recurrent 3D convolutional
neural networks for cross-task mental workload assessment, "
IEEE Trans. Neural Syst. Rehabilitation Eng., vol. 27, no. 1, pp.
31-42, Jan. 2019.
[12] H. Tang, S. Lu, G. Qian, J. Ding, Y. Liu and Q. Wang, " IoT-based
signal augmentation and compression method for efficient
motor bearing fault diagnosis, " IEEE Sensors J., vol. 21, no. 2, pp.
1820-1828, Jan. 2021.
[13] A. S. Iliyasu and H. Deng, " Semi-supervised encrypted traffic
classification with deep convolutional generative adversarial
networks, " IEEE Access, vol. 8, pp. 118-126, 2020.
[14] H. Zhang, V. Sindagi and V. M. Patel, " Image de-raining using a
conditional generative adversarial network, " IEEE Trans. Circuits
and Syst. Video Technol., vol. 30, no. 11, pp. 3943-3956, Nov. 2020.
[15] H. Liu, J. Zhou, Y. Xu, Y. Zheng, X. Peng, and Wei Jiang,
" Unsupervised fault diagnosis of rolling bearings using a deep
neural network based on generative adversarial networks, "
Neurocomputing, vol. 315, pp. 412-424, 2018.
[16] M. Xu and Y. Wang, " An imbalanced fault diagnosis method for
rolling bearing based on semi-supervised conditional generative
adversarial network with spectral normalization, " IEEE Access,
vol. 9, pp. 27736-27747, 2021.
[17] H. Zhang, R. Wang, R. Pan and H. Pan, " Imbalanced fault
diagnosis of rolling bearing using enhanced generative
adversarial networks, " IEEE Access, vol. 8, pp. 185950-185963,
2020.
[18] H. Zhang, X. Hu, D. Ma, R. Wang and X. Xie, " Insufficient data
generative model for pipeline network leak detection using
generative adversarial networks, " IEEE Trans. Cybernetics, early
access, 2020.
[19] M. Önder and Y. S. Akgül, " Automatic generation of matching
clothes design using generative adversarial networks, " in Proc.
IEEE Instrumentation & Measurement Magazine
April 2022
Instrumentation & Measurement Magazine 25-2

Table of Contents for the Digital Edition of Instrumentation & Measurement Magazine 25-2
