The implementation of machine learning methods for structural health monitoring applications has proven to be very powerful, especially in detecting damage and compensating for environmental and operational conditions. The use of guided waves in this area has also shown to be a powerful tool due to their sensitivity to structural changes in the propagation medium. In this work, two strategies for detecting damage and distinguishing their positions and for dealing with temperature variations without an additional classical temperature compensation technique are investigated. For this purpose, four unsupervised dimensionality reduction learning methods were used and compared: Principal Component Analysis, Kernel Principal Component Analysis, t-distributed stochastic neighbour embedding and Autoencoder. The first strategy (score plot) consists of using the latent dimensions directly to distinguish the data points of different states of the structure, and the second (DI plot) proposes a method to use Q- and T2-statistics, which have been proposed in previous work for PCA, computed using the compressed representation of the monitoring data. To this end, monitoring data from intact and damaged states of a 500x500x2 mm plate of carbon-fibre–reinforced polymer recorded by 12 piezoelectric transducers at different temperatures are examined. As a reversible damage model, a 10 mm thick aluminium disc is placed at four different locations on the plate. The results primarily show the success of the methods used with DI plot in detecting damage regardless of varying temperature. The autoencoder in the first strategy also demonstrates promising performance in detecting and distinguishing the position of the damage, even in the presence of varying temperature conditions.
Schneller Präzisionsbau erfordert Modulbauweisen mit leichten, gut handhabbaren Bauelementen, die mit hohen Wiederholungsraten hergestellt werden können. Ein modularisiertes Stabwerk stellt eine solche Bauweise dar. Damit die Stabelemente kontinuierlich, schnell und wirtschaftlich hergestellt werden können, braucht es gegenüber dem klassischen Stahlbetonbau neue Fertigungs‐ und Konstruktionskonzepte, bei denen die Konstruktion hocheffizient ausgelegt sein muss (form follows force). Darüber hinaus muss die Konstruktion konsequent die Anforderungen aus einer schnellen Serienfertigung erfüllen. Für solche neuen Fertigungsprozesse bieten sich Qualitätssicherungskontrollen mit Künstlicher Intelligenz (KI) an, die ihre Daten aus einer Vielzahl unterschiedlicher Sensoren und Sensorarten gewinnen. Eine solche KI‐gestützte Produktion wird durch ein Künstliches Neuronales Netz (KNN) umgesetzt, das gegenüber konventionellen Ansätzen die Möglichkeit bietet, nichtlineare Zusammenhänge von heterogenen Daten abbilden zu können. In diesem Beitrag wird die KNN‐basierte Form der Qualitätssicherung (QS) an fließgefertigten Stabelementen exemplarisch für die Beurteilung von Betonoberfläche und Geometrie beschrieben, die einen ersten Schritt in eine Prozessregelung darstellen kann. Erste Untersuchungen haben gezeigt, dass damit sehr effizient auch kleine Fehlstellen eindeutig erkannt und lokalisiert werden können.
Concrete steel towers are increasingly being used for onshore wind turbines. The lower part consists of separated segmented concrete rings connected with dry joints. Due to slight deviations from the axisymmetric cross-section, closely spaced modes occur. Therefore, the influences of small system changes on closely spaced modes, particularly the mode shapes, should be investigated to enable reliable vibration-based monitoring. In this context, the influence of imperfections due to the waviness of the dry joints requires attention. As no acceleration measurements on concrete towers considering small system changes have been performed so far, this has not yet been investigated. Therefore, an experiment is carried out using a large-scale laboratory model of a prestressed concrete segment tower. The system modifications are introduced by changing the preload. This changes the influence of imperfections of the surfaces of the horizontal dry joints, estimated by measuring strain and displacement at the lowest joint. An increasing preload causes the first two pairs of bending modes to move closer together. This enables to study the effect of the closeness of natural frequencies on the related mode shapes based on the same structure. Thus, the known effects of increasing uncertainty of the alignment and a rotation of the mode shape in the mode subspace with closer natural frequencies can be shown experimentally. In this work, the operational modal analysis (OMA) methods Bayesian-OMA (BAYOMA) and Stochastic Subspace Identification (SSI) are used. Local imperfections can significantly affect modal parameters, so these should be considered for vibration-based monitoring.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.