A novel CNN architecture for robust structural damage identification via strain measurements and its validation via full-scale experiments

• Published in: Measurement : journal of the International Measurement Confederation Vol. 239; p. 115393
• Main Authors: Duran, Burak, Emory, Dominic, Eftekhar Azam, Yashar, Linzell, Daniel G.
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 15-01-2025
• Subjects: Bridge health monitoring; Convolutional neural network; damage detection; Strain time-history; Supervised learning; damage detection; Strain time-history; Supervised learning; Bridge health monitoring; Convolutional neural network
• ISSN: 0263-2241
• DOI: 10.1016/j.measurement.2024.115393

•A novel and robust CNN architecture for structural damage identification framework.•This method works without measuring applied loads to the structure.•Measured strain time histories were converted to image data type.•Robustness was validated using full scale experiments on a laboratory structure.•Invisible damage due to a full-scale crash test was accurately detected by as few as four sensors and low sampling rates.•Over 300 live load tests were conducted before and after inducing three damage levels on the structure.
In this study, an innovative two-dimensional Convolutional Neural Network (2D CNN) architecture is proposed and investigated for the classification of bridge damage. Employing unique strain time-history data transformed into grayscale images, the approach seamlessly combines feature extraction and classification, allowing for the precise identification and categorization of structural damage. The method’s effectiveness was validated through field experiments on a full-scale bridge mock-up sample subjected to several controlled damage states under nonstationary, commercial vehicle loads. A wide range of realistic damage conditions, from minor to severe structural damage states, was included in the experimental scenarios together with inherent operational uncertainties. The robustness of the 2D CNN model was rigorously tested against fluctuating loads and introduced noise. Demonstrating remarkable accuracy, the 2D CNN successfully classified different damage states with over 95% accuracy, effectively identifying a damage state that was visually undetectable. Furthermore, the architecture proved to be highly versatile, effectively handling variations in the number of sensors. uncertainties included in the experimental data, and elevated levels of measurement noise.