This PhD thesis presents and discusses the development of SHM technique based on strain field in order to detect and monitor crack growth in bonded composite joints. Backface strain (BFS) field was used to develop the monitoring method. In the beginning, the concept of the monitoring technique based of BFS field was developed for Aluminum-Carbon fibre reinforced polymer (CFRP) adhesively bonded single lap (SL) joint. The characteristics of BFS profile along the exposed surface near and within the bonded zone was investigated by finite element (FE) analyses for ‘no-crack’ and ‘with crack’ conditions. It was observed that, the BFS profile has a negative minimum peak which has a linear correlation with the crack position. Experiments were carried out based on this correlation to detect crack and to monitor fatigue crack growth (FCG) in single lap joint. An array of conventional electrical strain gages as well as an array of fibre Bragg grating (FBG) sensors were installed on the backface of Aluminum substrate to capture BFS profile for monitoring crack position. Validation was done by comparing the crack length detected by this technique with the crack length measured using optical microscope. The array of FBG sensors was installed in the Aluminum substrate which is homogeneous material. In order to apply this BFS profile based monitoring technique for bonded joints having substrates made of non-homogeneous materials such as in woven composites, the spectral response of FBG sensors to be used to obtain BFS profile was analysed combining transfer matrix (T-matrix) with digital image correlation (DIC) method. DIC technique was used to capture non-uniform strain field in woven composites and used in T-Matrix to simulate FBG response. It was seen that FBGs having smaller gage length produce sharp peaks of reflected spectra even when large strain gradient exists and therefore are suitable for woven applications, whereas peak splitting occurs for FBGs having larger gage lengths, thus making them not recommended for non-uniform strain fields. However, FBG with small gauge length tended to provide results which are affected by local inhomogeneities of the underlying microstructure, thus making it difficult to exploit the SHM techniques based on FE models. Moreover, uncertainty about the exact position of gratings may affect accuracy of results. After that, experiments were carried out using FBGs having different gauge lengths to confirm the simulated observations. Finally, distributed optical sensing technique using optical backscatter reflectometer (OBR) was assessed to capture the BFS profile for SHM. The great advantage of this system over the more conventional arrays of FBG is that it can transform an ordinary optical fibre into quasi-continuous sensor and the entire length of the fiber can be used for sensing strains. BFS profiles were captured for CFRP-CFRP SL bonded joint for different static loads using OBR optical sensing. Two step validation processes had been implemented. First, the OBR sensed BFS profiles were validated by FE analyses. Then FE analyses were validated by two dimensional (2D) DIC technique.
Questa tesi di dottorato presenta e discute lo sviluppo della tecnica di monitoraggio dell'integrità strutturale basata su campo di deformazione. Lo scopo è quello di rilevare e monitorare la propagazione di cricche negli incollaggi di differenti parti in materiale composito, basandosi sul campo di deformazione di backface (BFS). A questo scopo, il materiale scelto per tale studio è il polimero rinforzato con fibre di alluminio-carbonio (CFRP) incollato con tecnica a giro singolo (SL). Le caratteristiche del profilo BFS lungo la superficie di incollaggio nei pressi e all'interno della stessa, è stata studiata per elementi finiti (FE), implementando dapprima una analisi 'no-crack' e, successivamente inserendo una cricca all’interno dell’incollaggio. Ciò che si è osservato vede il profilo BFS avere un picco minimo negativo che presenta correlazione lineare con la posizione della cricca all’interno dell’incollaggio. Gli esperimenti sono stati effettuati sul giunto a giro singolo, sulla base di questa correlazione per rilevare la presenza o meno della cricca e per monitorare la relativa crescita sotto carico di fatica (FCG). Una serie di tradizionali estensimetri elettrici, nonché una serie di sensori in fibra ottica con reticolo di Bragg (FBG), sono stati installati sulla parte posteriore del supporto in alluminio al fine di misurare il profilo BFS per il controllo di posizione di cricca. Si è proceduto quindi alla validazione, eseguita confrontando la lunghezza della fessura rilevata da questa tecnica, con la lunghezza del difetto misurata utilizzando microscopio ottico. Normalmente, l’utilizzo dei sensori FBG è riservato allo studio e al monitoraggio di difettologie all’interno di materiali omogenei, per applicare questa tecnica basata sul profilo BFS in materiali disomogenei come giunti incollati in tessuto di composito e alluminio, si è valutata la risposta spettrale di tali sensori combinando rispettivamente la matrice di trasferimento (T-matrix) e il metodo della correlazione di immagine digitale (DIC). La correlazione di immagine digitale è quindi stata utilizzata per catturare il campo di deformazione non uniforme nel tessuto da utilizzare nella T-matrix per simulare la risposta FBG Si è visto che FBG aventi lunghezza minore, producono una risposta nella misura che si adatta particolarmente al monitoraggio di materiali in tessuto composito, mentre reticoli FBG con lunghezza maggiore, producono una misura in cui si avverte la divisione di picco, rendendoli così non raccomandati per materiali non uniformi. Tuttavia, FBG con piccolo tratto utile tendono a fornire risultati affetti da disomogeneità locali della microstruttura sottostante, rendendo così difficile sfruttare le tecniche SHM basate su modelli FE. Inoltre, l’incertezza sull'esatta posizione delle griglie, può influenzare la precisione dei risultati. Al fine di conseguire la validazione finale, si è condotta una campagna di prove utilizzando sensori FBG aventi differenti lunghezze di gauge, per confermare le osservazioni simulate. Al fine di ottenere delle misurazioni idonee al processo SHM, rispetto ai normali sensori FBG, si è adottata la tecnica di rilevamento ottico con ottica backscatter reflectometer (OBR). Il grande vantaggio di questo sistema rispetto alle matrici convenzionali di FBG, risiede nella possibilità di trasformare una normale fibra ottica nel sensore come quasi continua, consentendo la rilevazione di ceppi. I Profili BFS sono stati misurati per CFRP-CFRP SL a giunto incollato per carichi statici diversi, utilizzando rilevamento ottico OBR. La validazione delle misure da OBR si sono svolte seguendo due passi: in primo luogo, i profili BFS misurati con il metodo OBR sono stati convalidati tramite analisi ad elementi finiti, le quali sono state a loro volta validate dalla correlazione delle immagini digitali.
Experimental methods for the assessment of structural integrity of composite bonded joints
KHARSHIDUZZAMAN, MD
Abstract
This PhD thesis presents and discusses the development of SHM technique based on strain field in order to detect and monitor crack growth in bonded composite joints. Backface strain (BFS) field was used to develop the monitoring method. In the beginning, the concept of the monitoring technique based of BFS field was developed for Aluminum-Carbon fibre reinforced polymer (CFRP) adhesively bonded single lap (SL) joint. The characteristics of BFS profile along the exposed surface near and within the bonded zone was investigated by finite element (FE) analyses for ‘no-crack’ and ‘with crack’ conditions. It was observed that, the BFS profile has a negative minimum peak which has a linear correlation with the crack position. Experiments were carried out based on this correlation to detect crack and to monitor fatigue crack growth (FCG) in single lap joint. An array of conventional electrical strain gages as well as an array of fibre Bragg grating (FBG) sensors were installed on the backface of Aluminum substrate to capture BFS profile for monitoring crack position. Validation was done by comparing the crack length detected by this technique with the crack length measured using optical microscope. The array of FBG sensors was installed in the Aluminum substrate which is homogeneous material. In order to apply this BFS profile based monitoring technique for bonded joints having substrates made of non-homogeneous materials such as in woven composites, the spectral response of FBG sensors to be used to obtain BFS profile was analysed combining transfer matrix (T-matrix) with digital image correlation (DIC) method. DIC technique was used to capture non-uniform strain field in woven composites and used in T-Matrix to simulate FBG response. It was seen that FBGs having smaller gage length produce sharp peaks of reflected spectra even when large strain gradient exists and therefore are suitable for woven applications, whereas peak splitting occurs for FBGs having larger gage lengths, thus making them not recommended for non-uniform strain fields. However, FBG with small gauge length tended to provide results which are affected by local inhomogeneities of the underlying microstructure, thus making it difficult to exploit the SHM techniques based on FE models. Moreover, uncertainty about the exact position of gratings may affect accuracy of results. After that, experiments were carried out using FBGs having different gauge lengths to confirm the simulated observations. Finally, distributed optical sensing technique using optical backscatter reflectometer (OBR) was assessed to capture the BFS profile for SHM. The great advantage of this system over the more conventional arrays of FBG is that it can transform an ordinary optical fibre into quasi-continuous sensor and the entire length of the fiber can be used for sensing strains. BFS profiles were captured for CFRP-CFRP SL bonded joint for different static loads using OBR optical sensing. Two step validation processes had been implemented. First, the OBR sensed BFS profiles were validated by FE analyses. Then FE analyses were validated by two dimensional (2D) DIC technique.File | Dimensione | Formato | |
---|---|---|---|
PhD Thesis_Md Kharshiduzzaman.pdf
non accessibile
Descrizione: PhD Thesis
Dimensione
14.66 MB
Formato
Adobe PDF
|
14.66 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/109744