Full-field Measurements for Anomaly Detection of Mechanical Systems using Convolutional Neural Networks and LSTM Networks 111 Key findings include: 1. Effectiveness of Machine Learning Models: LSTM networks proved to be highly effective, achieving excellent performance with a peak accuracy of 94.28% in the model. LSTMs excelled at capturing temporal dependencies in time-series data, making them particularly well-suited for tasks involving sequential patterns. 2. Challenges in Complex Non-Linear States: The models performed well overall but struggled with more complex non-linear states involving multiple bumpers, which introduced intricate vibrational patterns. 3. Frequency-Based Performance Variations: The ability to differentiate between linear and non-linear states depended on the frequency, with the models performing best at natural vibrational frequencies like 2118 Hz and 2792 Hz. However, accuracy declined at higher frequencies, particularly at 4960 Hz, where the distinction between the states became less clear. The LSTM approach applied to videos generated with laser Doppler vibrometers showed great potential for real-time SHM applications, particularly in complex scenarios and high-frequency performance. This research contributes to the development of machine learning methods for real-time anomaly detection in structural systems. Acknowledgments This work was supported by the GHAIA (Geometric and Harmonic Analysis with Interdisciplinary Applications) project, under the H2020-MSCA-RISE program of the European Commission. This project has received funding from the European Union’s Horizon 2020 research and innovation program under the Marie Skłodowska-Curie grant agreement No 777822. References 1. M. Arul and A. Kareem, “Data anomaly detection for structural health monitoring of bridges using shapelet transform,” arXiv preprint arXiv:2009.00470, 2020. 2. N. D. Boffa, M. Arena, E. Monaco, M. Viscardi, F. Ricci, and T. Kundu, “About the combination of high and low frequency methods for impact detection on aerospace components,” Progress in Aerospace Sciences, vol. 129, p. 100789, 2022. 3. D. Caballol, A´ . P. Raposo, and F. Gil-Carrillo, “Non-destructive testing of concrete layer adhesion by means of vibration measurement,” Construction and Building Materials, vol. 411, p. 134548, 2024. 4. N. M. Kalimullah, A. Shelke, and A. Habib, “A deep learning approach for anomaly identification in PZT sensors using point contact method,” Smart Materials and Structures, vol. 32, no. 9, p. 095027, 2023. 5. X. Ye, P. Wu, A. Liu, X. Zhan, Z. Wang, and Y. Zhao, “A deep learning-based method for automatic abnormal data detection: Case study for bridge structural health monitoring,” International Journal of Structural Stability and Dynamics, vol. 23, no. 11, p. 2350131, 2023. 6. C. T. do Cabo and Z. Mao, “An Optical Mode Shape-Based Damage Detection Using Convolutional Neural Networks,” in Rotating Machinery, Optical Methods & Scanning LDV Methods, Volume 6: Springer, 2022, pp. 157–162. 7. Z. Tang, Z. Chen, Y. Bao, and H. Li, “Convolutional neural network-based data anomaly detection method using multiple information for structural health monitoring,” Structural Control and Health Monitoring, vol. 26, no. 1, p. e2296, 2019. 8. S.-Y. Kim and M. Mukhiddinov, “Data Anomaly Detection for Structural Health Monitoring Based on a Convolutional Neural Network,” Sensors, vol. 23, no. 20, p. 8525, 2023. 9. C. T. do Cabo and Z. Mao, “An Optical Temporal and Spatial Vibration-Based Damage Detection Using Convolutional Neural Networks and Long Short-Term Memory,” in Rotating Machinery, Optical Methods & Scanning LDV Methods, Volume 6: Springer, 2022, pp. 159–165. 10. R. Yang et al., “CNN-LSTM deep learning architecture for computer vision-based modal frequency detection,” Mechanical Systems and signal processing, vol. 144, p. 106885, 2020. 11. M. Ma and Z. Mao, “Deep-convolution-based LSTM network for remaining useful life prediction,” IEEE Transactions on Industrial Informatics, vol. 17, no. 3, pp. 1658–1667, 2020. 12. A. A. M. Al-Saffar, H. Tao, and M. A. Talab, “Review of deep convolution neural network in image classification,” in 2017 International conference on radar, antenna, microwave, electronics, and telecommunications (ICRAMET), 2017: IEEE, pp. 26–31.
RkJQdWJsaXNoZXIy MTMzNzEzMQ==