An integrated deep neural network model combining 1D CNN and LSTM for structural health monitoring utilizing multisensor time-series data

Mohammadreza Ahmadzadeh; Seyed Mehdi Zahrai; Maryam Bitaraf

doi:10.1177/14759217241239041

journal article Mar 26, 2024

An integrated deep neural network model combining 1D CNN and LSTM for structural health monitoring utilizing multisensor time-series data

Mohammadreza Ahmadzadeh

Seyed Mehdi Zahrai

Maryam Bitaraf

Structural Health Monitoring Vol. 24 No. 1 pp. 447-465 · SAGE Publications

View at Publisher Save 10.1177/14759217241239041

Abstract

Introducing deep learning algorithms into the field of structural health monitoring (SHM) has contributed to the automatic extraction of damage-sensitive features, but the type and architecture of these algorithms are still in dispute. This paper proposes a hybrid deep learning framework entitled time-distributed one-dimensional convolutional neural network (1D CNN) long short-term memory (LSTM) model, which utilizes raw multisensor time histories to detect structural damages. Using a sliding window that moves along the temporal dimension, the multisensor data are first segmented into subsequences. The 1D CNN layers are simultaneously applied to each subsequence to extract damage-sensitive features from row data samples. These features are then fed into the LSTM layers to extract temporal features between subsequences. As the final step, these extracted features are classified using fully connected layers. In order to assess the performance of this model, a numerical model of a high-rise frame with nonlinear members is used. This hybrid model is assumed to identify the location of damages to this frame. In order to assess the proposed model with a real-world structure, a well-known benchmark building is employed to identify damage patterns by this deep hybrid neural network. A set of metrics related to the performance of the model is measured and evaluated. It is found that the model has an average accuracy of above 96.6% in localizing damage in the numerical structure and above 99.6% in detecting each damage pattern in the experimental building. The results indicate that the proposed model can be applied effectively to the SHM of different structural systems with different damage patterns.

Topics

No keywords indexed for this article. Browse by subject →

References

68

[1]

10.1088/0964-1726/10/3/304

[2]

10.1002/9781118443118

[3]

10.1002/9780470612071

[4]

Machine learning algorithms for damage detection under operational and environmental variability

Eloi Figueiredo, Gyuhae Park, Charles R Farrar et al.

Structural Health Monitoring 10.1177/1475921710388971

[5]

10.1243/03093247v142049

[6]

10.1016/0022-460x(91)90595-b

[7]

10.2172/249299

[8]

10.1016/s0168-874x(02)00227-5

[9]

10.1006/jsvi.1998.1878

[10]

10.1061/ajrua6.0000872

[11]

10.3390/app7050510

[12]

10.1016/s0141-0296(96)00149-6

[13]

10.1007/s11831-012-9069-x

[14]

10.1016/j.engstruct.2014.03.010

[15]

10.1111/mice.12447

[16]

10.1061/(asce)0733-9445(2004)130:7(1055)

[17]

10.1061/(asce)0733-9399(2004)130:1(85)

[18]

10.1002/tal.1312

[19]

10.1016/j.engstruct.2017.05.054

[20]

10.1016/j.ymssp.2013.02.019

[21]

10.1111/mice.12377

[22]

10.1111/mice.12313

[23]

10.1016/j.jsv.2004.01.003

[24]

10.11113/jt.v61.1624

[25]

10.7158/s12-036.2013.14.1

[26]

10.1016/j.jsv.2015.11.008

[27]

10.1260/136943306778812787

[28]

10.1111/j.1467-8667.2005.00399.x

[29]

10.1002/stc.116

[30]

10.1061/(asce)cp.1943-5487.0000517

[31]

Deep learning

Yann LeCun, Yoshua Bengio, Geoffrey Hinton

Nature 10.1038/nature14539

[32]

Computer vision and deep learning–based data anomaly detection method for structural health monitoring

Yuequan Bao, Zhiyi Tang, Hui Li et al.

Structural Health Monitoring 10.1177/1475921718757405

[33]

Real-Time Patient-Specific ECG Classification by 1-D Convolutional Neural Networks

Serkan Kiranyaz, Turker Ince, Moncef Gabbouj

IEEE Transactions on Biomedical Engineering 10.1109/tbme.2015.2468589

[34]

Advances in Computer Vision-Based Civil Infrastructure Inspection and Monitoring

Billie F. Spencer, Vedhus Hoskere, Yasutaka Narazaki

Engineering 10.1016/j.eng.2018.11.030

[35]

Khodabandehlou H Struct Control Health Monit (2019)

[36]

Deep Learning‐Based Crack Damage Detection Using Convolutional Neural Networks

Young‐Jin Cha, Wooram Choi, Oral Büyüköztürk

Computer-Aided Civil and Infrastructure Engineerin... 10.1111/mice.12263

[37]

10.1177/1475921718764873

[38]

10.1177/1475921718768747

[39]

10.48550/arxiv.2111.09117

[40]

10.1016/j.jsv.2016.10.043

[41]

Avci O (2017)

[42]

10.1016/j.neucom.2017.09.069

[43]

10.1177/14759217221137931

[44]

10.1177/14759217231158786

[45]

10.1016/j.engstruct.2018.05.109

[46]

10.1002/tal.1400

[47]

Deep learning and its applications to machine health monitoring

Rui Zhao, Ruqiang Yan, Zhenghua Chen et al.

Mechanical Systems and Signal Processing 10.1016/j.ymssp.2018.05.050

[48]

10.1111/mice.12409

[49]

Long Short-Term Memory

Sepp Hochreiter, Jürgen Schmidhuber

Neural Computation 10.1162/neco.1997.9.8.1735

[50]

10.1109/cac48633.2019.8996864

Showing 50 of 68 references

Metrics

62

Citations

68

References

Details

Published: Mar 26, 2024
Vol/Issue: 24(1)
Pages: 447-465
License: View

Authors

M

Mohammadreza Ahmadzadeh

School of Civil Engineering, College of Engineering, University of Tehran, Tehran, Iran

S

Seyed Mehdi Zahrai

School of Civil Engineering, College of Engineering, University of Tehran, Tehran, Iran

M

Maryam Bitaraf

School of Civil Engineering, College of Engineering, University of Tehran, Tehran, Iran

Cite This Article

Mohammadreza Ahmadzadeh, Seyed Mehdi Zahrai, Maryam Bitaraf (2024). An integrated deep neural network model combining 1D CNN and LSTM for structural health monitoring utilizing multisensor time-series data. Structural Health Monitoring, 24(1), 447-465. https://doi.org/10.1177/14759217241239041

An integrated deep neural network model combining 1D CNN and LSTM for structural health monitoring utilizing multisensor time-series data

You May Also Like