A defect prediction method based on distributed optical fiber strain sensing is proposed for the health monitoring of reinforced concrete structures. By constructing a general artificial neural network and performing deep learning training on defective samples, feature extraction and classification and recognition can be automatically realized, avoiding the complexity of manual modeling methods. By carrying out defect simulation experiments, the accuracy of the defect prediction method was verified. Experiments show that the classification and recognition model after deep learning can achieve accurate prediction of defect samples, with an accuracy rate of over 99%.
Wu, Y.H., Zhu, S., Xu, W. and Zhang, H.M. (2018) Progress in Distributed Optical Fiber Crack Sensing Engineering. Laser & Optoelectronics Progress, 55, 090002. https://doi.org/10.3788/LOP55.090002
[3]
Yao, Y., Tung, S.T.E. and Glisic, B. (2014) Crack Detection and Characterization Techniques: An Overview. Structural Control and Health Monitoring, 21, 1387-1413. https://doi.org/10.1002/stc.1655
[4]
Das, S. and Saha, P. (2018) A Review of Some Advanced Sensors Used for Health Diagnosis of Civil Engineering Structures. Measurement, 129, 68-90. https://doi.org/10.1016/j.measurement.2018.07.008
Yan, G.P., Chen, Y., Li, Y.C. and Yan, Z.F. (2020) Distributed Strain Crack Detection Based on One-Dimensional Stacked Convolutional Autoencoder. Computer Systems and Applications, 29, 144-150.
[7]
Hinton, G.E., Srivastava, N., Krizhevsky, A., Sutskever, I. and Salakhutdinov, R.R. (2012) Improving Neural Networks by Preventing Co-Adaptation of Feature Detectors. Computer Science.
