Abstract: In this paper, we use deep learning to investigate a loss factor function (LF) for measuring energy dispersal in bridge structures in Ho Chi Minh City, Vietnam. The LF is calculated from the power spectral density (PSD) of random vibration signals to account for the mechanical parameters required for detecting structural changes. The LF is applied to many different types of bridge decks such as a prestressed concrete bridge, precast reinforced concrete bridge, and cable-stayed bridge. In addition, to ensure the new parameters are working effectively for the evaluation, a deep learning-based signal processing platform is used along with a convolutional neural network (CNN) to create the training. The training process helps eliminate interference values and errors. This demonstrates that the LF is sensitive to many different real-life structures while previous parameters are sensitive to only particular structures.

评估桥梁结构能量扩散的基于深度学习的信号处理方法

[1]Anitescu C, Atroshchenko E, Alajlan N, et al., 2019. Artificial neural network methods for the solution of second order boundary value problems. Computers, Materials & Continua, 59(1):345-359.

[2]Gao YQ, Mosalam KM, 2018. Deep transfer learning for image-based structural damage recognition. Computer-aided Civil and Infrastructure Engineering, 33(9):748-768.

[3]Guo HW, Zhuang XY, Rabczuk T, 2019. A deep collocation method for the bending analysis of Kirchhoff plate. Computers, Materials & Continua, 59(2):433-456.

[4]Nguyen TD, Nguyen HQ, Pham TB, et al., 2021. A novel proposal in using viscoelastic model for bridge condition assessment. In: Bui TQ, Cuong LT, Khatir S (Eds.), Structural Health Monitoring and Engineering Structures. Springer, Singapore.

[5]Samaniego E, Anitescu C, Goswami S, et al., 2020. An energy approach to the solution of partial differential equations in computational mechanics via machine learning: concepts, implementation and applications. Computer Methods in Applied Mechanics and Engineering, 362:112790.

[6]Wang ZL, Cha YJ, 2021. Unsupervised deep learning approach using a deep auto-encoder with a one-class support vector machine to detect damage. Structural Health Monitoring, 20(1):406-425.

[7]Yu YJ, Cao H, Yan XY, et al., 2020. Defect identification of wind turbine blades based on defect semantic features with transfer feature extractor. Neurocomputing, 376:1-9.