Advanced search
1 file | 740.07 KB Add to list
Author
Organization
Abstract
Structural Health Monitoring is an important field that involves the continuous measuring of the structural status of infrastructures. In order to be able to detect the damage status, data collected from sensors have to be processed to identify the difference between the damaged and the undamaged states. There exist machine learning techniques attempting to extract features from vibration data, however, they require prior knowledge about the factors affecting the structure. In this paper, we propose a novel method of damage detection using a convolution neural network and a handcrafted feature extraction. This method uses a convolution neural network to extract deep features in time series and uses handcrafted features to find a statistically significant correlation of each lagged features in time series data. These two types of features are combined to increase discrimination ability compared to deep features only. Finally, the neural network will be used to classify the time series into normal and damaged states. The accuracy of damaged detection was tested on a benchmark dataset from Los Alamos National Laboratory and the result shows that hybrid features provided a highly accurate damage identification.

Downloads

  • (...).pdf
    • full text (Published version)
    • |
    • UGent only
    • |
    • PDF
    • |
    • 740.07 KB

Citation

Please use this url to cite or link to this publication:

MLA
Bui-Ngoc, Dung, et al. “Structural Health Monitoring Using Handcrafted Features and Convolution Neural Network.” Proceedings of 1st International Conference on Structural Damage Modelling and Assessment (SDMA 2020), edited by Magd Abdel Wahab, vol. 110, Springer, 2020, pp. 103–12, doi:10.1007/978-981-15-9121-1_8.
APA
Bui-Ngoc, D., Bui-Tien, T., Nguyen-Tran, H., Abdel Wahab, M., & De Roeck, G. (2020). Structural health monitoring using handcrafted features and convolution neural network. In M. Abdel Wahab (Ed.), Proceedings of 1st International Conference on Structural Damage Modelling and Assessment (SDMA 2020) (Vol. 110, pp. 103–112). Singapore: Springer. https://doi.org/10.1007/978-981-15-9121-1_8
Chicago author-date
Bui-Ngoc, Dung, Thanh Bui-Tien, Hieu Nguyen-Tran, Magd Abdel Wahab, and Guido De Roeck. 2020. “Structural Health Monitoring Using Handcrafted Features and Convolution Neural Network.” In Proceedings of 1st International Conference on Structural Damage Modelling and Assessment (SDMA 2020), edited by Magd Abdel Wahab, 110:103–12. Singapore: Springer. https://doi.org/10.1007/978-981-15-9121-1_8.
Chicago author-date (all authors)
Bui-Ngoc, Dung, Thanh Bui-Tien, Hieu Nguyen-Tran, Magd Abdel Wahab, and Guido De Roeck. 2020. “Structural Health Monitoring Using Handcrafted Features and Convolution Neural Network.” In Proceedings of 1st International Conference on Structural Damage Modelling and Assessment (SDMA 2020), ed by. Magd Abdel Wahab, 110:103–112. Singapore: Springer. doi:10.1007/978-981-15-9121-1_8.
Vancouver
1.
Bui-Ngoc D, Bui-Tien T, Nguyen-Tran H, Abdel Wahab M, De Roeck G. Structural health monitoring using handcrafted features and convolution neural network. In: Abdel Wahab M, editor. Proceedings of 1st International Conference on Structural Damage Modelling and Assessment (SDMA 2020). Singapore: Springer; 2020. p. 103–12.
IEEE
[1]
D. Bui-Ngoc, T. Bui-Tien, H. Nguyen-Tran, M. Abdel Wahab, and G. De Roeck, “Structural health monitoring using handcrafted features and convolution neural network,” in Proceedings of 1st International Conference on Structural Damage Modelling and Assessment (SDMA 2020), Ghent University, Belgium, 2020, vol. 110, pp. 103–112.
@inproceedings{8686744,
  abstract     = {Structural Health Monitoring is an important field that involves the continuous measuring of the structural status of infrastructures. In order to be able to detect the damage status, data collected from sensors have to be processed to identify the difference between the damaged and the undamaged states. There exist machine learning techniques attempting to extract features from vibration data, however, they require prior knowledge about the factors affecting the structure. In this paper, we propose a novel method of damage detection using a convolution neural network and a handcrafted feature extraction. This method uses a convolution neural network to extract deep features in time series and uses handcrafted features to find a statistically significant correlation of each lagged features in time series data. These two types of features are combined to increase discrimination ability compared to deep features only. Finally, the neural network will be used to classify the time series into normal and damaged states. The accuracy of damaged detection was tested on a benchmark dataset from Los Alamos National Laboratory and the result shows that hybrid features provided a highly accurate damage identification.},
  author       = {Bui-Ngoc, Dung and Bui-Tien, Thanh and Nguyen-Tran, Hieu and Abdel Wahab, Magd and De Roeck, Guido},
  booktitle    = {Proceedings of 1st International Conference on Structural Damage Modelling and Assessment (SDMA 2020)},
  editor       = {Abdel Wahab, Magd},
  isbn         = {9789811591204},
  issn         = {2366-2557},
  language     = {eng},
  location     = {Ghent University, Belgium},
  pages        = {103--112},
  publisher    = {Springer},
  title        = {Structural health monitoring using handcrafted features and convolution neural network},
  url          = {http://dx.doi.org/10.1007/978-981-15-9121-1_8},
  volume       = {110},
  year         = {2020},
}

Altmetric
View in Altmetric