Maribel Anaya Vejar

Data-Driven Methodologies for Structural Damage Detection Based on Machine Learning Applications

Structural health monitoring (SHM) is an important research area, which interest is the damage identification process. Different information about the state of the structure can be obtained in the process, among them, detection, localization and classification of damages are mainly studied in order to avoid unnecessary maintenance procedures in civilian and military structures in several applications. To carry out SHM in practice, two different approaches are used, the first is based on modelling which requires to build a very detailed model of the structure, while the second is by means of data-driven approaches which use information collected from the structure under different structural states and perform an analysis by means of data analysis . For the latter, statistical analysis and pattern recognition have demonstrated its effectiveness in the damage identification process because real information is obtained from the structure through sensors installed permanently to the observed object allowing a real-time monitoring. This chapter describes a damage detection and classification methodology, which makes use of a piezoelectric active system which works in several actuation phases and that is attached to the structure under evaluation, principal component analysis, and machine learning algorithms working as a pattern recognition methodology. In the chapter, the description of the developed approach and the results when it is tested in one aluminum plate are also included.

Principal Component Analysis and Self-organizing Maps for Damage Detection and Classification under Temperature Variations

The use of statistical techniques for data driven has proven very useful in multivariable analysis as a pattern recognition approach. Among their multiple advantages such as data reduction, multivariable analysis and the definition of statistical models built with data from experimental trials, they provide robustness and allow avoiding the need of the development of physical models which sometimes are difficult for modelling especially when the system is complex. In this paper, a methodology based on Principal Component Analysis (PCA) is developed and used for building statistical baseline models comprising the dynamics from the monitored healthystructureunderdifferenttemperatureconditions.Inasecondstep, fortesting the proposed methodology, data from the structure at different structural states and under different temperature conditions are projected into the baseline models in order to obtain statistical measures (Scores and Q-index) which are included as feature vectors in a Self-Organizing Map for the damage detection and classification tasks. The methodology is evaluated using ultrasonic signals collected from an aluminium plate and a stiffened composite panel. Results show that all the simulated states are successfully classified no matter what the kind of damage or the temperature is present in both structures.