SHM under varying environmental conditions: an approach based on model order reduction and deep learning. (1st July 2022)

• Record Type: Journal Article
• Authors: Torzoni, Matteo; Rosafalco, Luca; Manzoni, Andrea; Mariani, Stefano; Corigliano, Alberto
• Is Part Of: Computers & structures. Volume 266(2022)
• Publication Date: 2022-07-01
• Subjects: Structural Health Monitoring -- Reduced order modeling -- Deep learning -- Damage identification -- Temperature effect -- Environmental and operational variability; Structural engineering -- Data processing -- Periodicals; Electronic data processing -- Structures, Theory of -- Periodicals; 624.171
• DOI: 10.1016/j.compstruc.2022.106790
• Languages: English
• ISSNs: 0045-7949

Abstract

Highlights: Within a structural health monitoring framework, we propose a simulation-based strategy that exploits vibration and thermal data to manage the confounding influence of temperature fluctuations on the damage identification process. The proposed strategy combines model order reduction techniques and deep learning algorithms. The tasks of damage detection and damage localization are handled as a supervised classification exploiting a fully convolutional neural network architecture. In order to speed up the offline assembling of the training dataset of vibration and temperature recordings, a parametric order reduction of the thermo-mechanical model of the monitored structure is exploited. By correlating the effects of temperature fluctuations on the vibration recordings to the thermo-mechanical coupling, environment-insensitive and damage-sensitive features are automatically selected and extracted by the neural network. Two case studies has been adopted to validate the proposed procedure, respectively involving a cantilever beam and a portal frame. Abstract: Data-driven approaches to structural health monitoring (SHM) have been recently shown to be a powerful paradigm, helping to lead to an evolution of traditional scheduled-based maintenance methodologies towards condition-based ones. Nevertheless, only few of them provide monitoring scenarios accounting for the varying loading and environmental conditions, which can potentially lead to misinterpretations of the structural state. In this paper, we propose a damage localization strategy that efficiently exploits vibration and temperature data to account for the effects of temperature fluctuations on the structural response. By allowing for a finite number of a priori defined damage scenarios, deep learning techniques are used to handle the damage localization task as a supervised classification, conditioned on temperature data. The training dataset is generated through a parametrized thermo-mechanical model of the structure, under a prescribed variability of loading and thermal conditions. To relieve the computational burden associated to the data generation process, a parametric order reduction strategy is also exploited. Results relevant to two case studies, a cantilever beam and a portal frame, are adopted to testify the capability of the proposed procedure to locate the damage, also when characterized by a rather small reduction of the local stiffness properties.