[1] Tu Yongming, Lu Senlu, Wang Chao,. Damage identification of steel truss bridgesbased on deep belief network [J]. Journal of Southeast University (English Edition), 2022, 38 (4): 392-400. [doi:10.3969/j.issn.1003-7985.2022.04.008]
Copy
Damage identification of steel truss bridgesbased on deep belief network(
)
基于深度信念网络的钢桁架桥梁损伤识别
)Journal of Southeast University (English Edition)[ISSN:1003-7985/CN:32-1325/N]
- Volumn:
- 38
- Issue:
- 2022 4
- Page:
- 392-400
- Research Field:
- Traffic and Transportation Engineering
- Publishing date:
- 2022-12-20
Info
- Title:
- Damage identification of steel truss bridgesbased on deep belief network
- 基于深度信念网络的钢桁架桥梁损伤识别
- Author(s):
- Tu Yongming; Lu Senlu; Wang Chao
- Key Laboratory of Concrete and Prestressed Concrete Structures of Ministry of Education, Southeast University, Nanjing 211189, China
National Prestressing Engineering Research Center, Southeast University, Nanjing 211189, China
School of Civil Engineering, Southeast University, Nanjing 211189, China
- 涂永明; 卢森露; 王潮
- 东南大学混凝土及预应力混凝土结构教育部重点实验室, 南京 211189; 东南大学国家预应力工程技术研究中心, 南京 211189; 东南大学土木工程学院, 南京 211189
- Keywords:
- deep learning; restricted Boltzmann machine; deep belief network; structural damage identification
- PACS:
- U446.3
- DOI:
- 10.3969/j.issn.1003-7985.2022.04.008
- Abstract:
- To improve the accuracy and anti-noise ability of the structural damage identification method, a bridge damage identification method is proposed based on a deep belief network(DBN). The output vector is used to establish the nonlinear mapping relationship between the mode shape and structural damage. The hidden layer of the DBN is trained through a layer-by-layer pre-training. Finally, the backpropagation algorithm is used to fine-tune the entire network. The method is validated using a numerical model of a steel truss bridge. The results show that under the influence of noise and modeling uncertainty, the damage identification method based on the DBN can identify the accurate damage location and degree identification compared with the traditional damage identification method based on an artificial neural network.
- 为了提高结构损伤识别方法的精度和抗噪性, 基于深度信念网络提出了一种桥梁损伤识别方法.首先, 将桥梁前5阶不完全的模态数据作为输入数据, 将结构的损伤位置以及损伤程度作为输出向量, 建立模态振型与结构损伤的非线性映射关系;然后, 采用逐层预训练的方式对深度信念网络的隐含层进行训练;最后, 利用反向传播算法进行微调, 从而优化网络.基于钢桁架桥梁的数值模拟结果表明, 在噪声以及建模不确定性的影响下, 相较于传统的基于人工神经网络的损伤识别方法, 基于深度信念网络的损伤识别方法可以更为精确地识别损伤位置和程度.
References:
[1] Santos A, Figueiredo E, Silvia M, et al. Genetic-based EM algorithm to improve the robustness of Gaussian mixture models for damage detection in bridges [J]. Structural Control and Health Monitoring, 2017, 24(3): 1886. DOI: 10.1002/stc.1886.
[2] De Lautour O R, Omenzetter P. Nearest neighbor and learning vector quantization classification for damage detection using time series analysis [J]. Structural Control & Health Monitoring, 2010, 17(6):614-631. DOI: 10.1002/stc.335.
[3] Gui G, Pan H, Lin Z, et al. Data-driven support vector machine with optimization techniques for structural health monitoring and damage detection [J]. KSCE Journal of Civil Engineering, 2017, 21(2):523-534. DOI: 10.1007/s12205-017-1518-5.
[4] Ozdagli A I, Koutsoukos X. Machine learning based novelty detection using modal analysis [J]. Computer-Aided Civil and Infrastructure Engineering, 2019, 34(12): 1119-1140. DOI: 10.1111/mice.12511.
[5] Bandara R P, Chan T H T, Thambiratnam D P. Frequency response function based damage identification using principal component analysis and pattern recognition technique[J]. Engineering Structures, 2014, 66(4):116-128. DOI: 10.1016/j.engstruct.2014.01.044.
[6] Lam H F, Yuen K V, Beck J L.Structural health monitoring via measured ritz vectors utilizing artificial neural networks [J]. Computer-Aided Civil & Infrastructure Engineering, 2010, 21(4): 232-241. DOI: 10.1111/j.1467-8667.2006.00431.x.
[7] Nadith P, Li J, Ling L, et al. Structural damage identification based on autoencoder neural networks and deep learning [J]. Engineering Structures, 2018, 172:13-28. DOI: 10.1016/j.engstruct.2018.05.109.
[8] Zhang Y, Miyamori Y, Mikami S, et al. Vibration-based structural state identification by a 1-dimensional convolutional neural network [J]. Computer-Aided Civil and Infrastructure Engineering, 2019, 34(9):822-839. DOI: 10.1111/mice.12447.
[9] Guo Q, Feng L, Zhang R, et al.Study of damage identification for bridges based on deep belief network [J]. Advances in Structural Engineering, 2020, 23(3):1562-1572. DOI: 10.1177/1369433219898058.
[10] H�E4;ggstr�F6;m J, Blanksv�E4;rd T. Assessment and full scale failure test of a steel truss bridge [C]// IABSE Workshop Helsinki 2015. Helsinki, Finland, 2015:288-295. DOI: 10.2749/222137815815622951.
Memo
- Memo:
-
Biography: Tu Yongming(1978—), male, doctor, professor, tuyongming@seu.edu.cn.
Foundation item: The National Natural Science Foundation of China(No. 51378104).
Citation: Tu Yongming, Lu Senlu, Wang Chao. Damage identification of steel truss bridges based on deep belief network[J].Journal of Southeast University(English Edition), 2022, 38(4):392-400.DOI:10.3969/j.issn.1003-7985.2022.04.008.