Fully convolutional networks for structural health monitoring through multivariate time series classification
Open Access
- 24 September 2020
- journal article
- Published by Springer Science and Business Media LLC in Advanced Modeling and Simulation in Engineering Sciences
- Vol. 7 (1), 1-31
- https://doi.org/10.1186/s40323-020-00174-1
Abstract
We propose a novel approach to structural health monitoring (SHM), aiming at the automatic identification of damage-sensitive features from data acquired through pervasive sensor systems. Damage detection and localization are formulated as classification problems, and tackled through fully convolutional networks (FCNs). A supervised training of the proposed network architecture is performed on data extracted from numerical simulations of a physics-based model (playing the role of digital twin of the structure to be monitored) accounting for different damage scenarios. By relying on this simplified model of the structure, several load conditions are considered during the training phase of the FCN, whose architecture has been designed to deal with time series of different length. The training of the neural network is done before the monitoring system starts operating, thus enabling a real time damage classification. The numerical performances of the proposed strategy are assessed on a numerical benchmark case consisting of an eight-story shear building subjected to two load types, one of which modeling random vibrations due to low-energy seismicity. Measurement noise has been added to the responses of the structure to mimic the outputs of a real monitoring system. Extremely good classification capacities are shown: among the nine possible alternatives (represented by the healthy state and by a damage at any floor), damage is correctly classified in up to $$95 \%$$ 95 % of cases, thus showing the strong potential of the proposed approach in view of the application to real-life cases.
Keywords
Funding Information
- MIUR (PRIN 15-2015 LYYXA8)
This publication has 33 references indexed in Scilit:
- General Realization Algorithm for Modal Identification of Linear Dynamic SystemsJournal of Engineering Mechanics, 2008
- The nature of noise wavefield and its applications for site effects studies: A literature reviewEarth-Science Reviews, 2006
- Reducing the Dimensionality of Data with Neural NetworksScience, 2006
- Parameter identification in explicit structural dynamics: performance of the extended Kalman filterComputer Methods in Applied Mechanics and Engineering, 2004
- Review Paper: Health Monitoring of Civil InfrastructureStructural Health Monitoring, 2003
- Damage assessment by FE model updating using damage functionsComputers & Structures, 2002
- Statistical Damage Classification Under Changing Environmental and Operational ConditionsJournal of Intelligent Material Systems and Structures, 2002
- Vibration–based structural damage identificationPhilosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 2001
- A Summary Review of Vibration-Based Damage Identification MethodsThe Shock and Vibration Digest, 1998
- Small sample size effects in statistical pattern recognition: recommendations for practitionersIEEE Transactions on Pattern Analysis and Machine Intelligence, 1991