Load monitoring is, nowadays, a design matter. In particular, shape sensing by means of the inverse Finite Element Method (iFEM) is becoming more and more popular for its ability to reconstruct the correct deformed shape of a structure under random loads, even in presence of very sparse measured strain data. In the present work, the iFEM methodology is reviewed in detail with a plate subjected to different type of loads (axial, bending, shear and torque). In particular, it is analyzed the iFEM strengths and weaknesses by means of different checks. Among them, it is performed a sensors grid sensitivity and it is investigated how the results are affected by noisy measurements. Furthermore, it is proposed a methodology to perform a structural health monitoring by taking advantage of the iFEM ability to reconstruct the strain field of a structure by means of just few sensors. In particular, it is defined an anomaly detection index based upon the comparison between the strain read at a target sensor location and the one reconstructed (in the same position) trough the iFEM algorithm. This index enables to detect both the presence and the position of a crack. Two different crack positions are employed to numerically test the proposed method. Moreover, the anomaly detection method is experimentally validated trough a fatigue crack propagation test.
Il monitoraggio dei carichi è diventato un importante aspetto nella fase di progettazione di un componente. In particolare, lo shape sensing attraverso il metodo degli elementi finiti inverso (iFEM) stà diventando sempre più popolare per la sua abilità nel ricostruire la deformata di una struttura sottoposta a carichi random, anche nel caso di misure di deformazioni molto rade. In questo lavoro, si analizza nel dettaglio il metodo degli elementi finiti inverso attraverso una lastra soggetta a diversi tipi di carico (carico assiale, taglio, momento flettente e torcente). In particolare, si analizzano i punti di forza e i punti deboli dell’iFEM attraverso diverse prove. Tra esse, si segnala una prova per testare la sensibilità dei risultati alla griglia dei sensori e alle misure di deformazione soggette a rumore. Inoltre, si propone un metodo per eseguire il monitoraggio strutturale di un componente, sfruttando l’abilità dell’iFEM nel ricostruire il campo di deformazioni attraverso solo alcuni sensori. In particolare, viene proposto un indice di anomalia basato sul confronto tra la deformazione misurata da un sensore e ricostruita attraverso l’iFEM nella stessa posizione. Questo indice permette di identificare sia la presenza che la posizione di una cricca in maniera automatica. Il metodo è testato numericamente attraverso due diverse posizioni della cricca. Infine, il metodo proposto è validato sperimentalmente attraverso un test di propagazione a fatica di una cricca.
On the combination of load monitoring and damage identification with iFEM
COLOMBO, LUCA
2016/2017
Abstract
Load monitoring is, nowadays, a design matter. In particular, shape sensing by means of the inverse Finite Element Method (iFEM) is becoming more and more popular for its ability to reconstruct the correct deformed shape of a structure under random loads, even in presence of very sparse measured strain data. In the present work, the iFEM methodology is reviewed in detail with a plate subjected to different type of loads (axial, bending, shear and torque). In particular, it is analyzed the iFEM strengths and weaknesses by means of different checks. Among them, it is performed a sensors grid sensitivity and it is investigated how the results are affected by noisy measurements. Furthermore, it is proposed a methodology to perform a structural health monitoring by taking advantage of the iFEM ability to reconstruct the strain field of a structure by means of just few sensors. In particular, it is defined an anomaly detection index based upon the comparison between the strain read at a target sensor location and the one reconstructed (in the same position) trough the iFEM algorithm. This index enables to detect both the presence and the position of a crack. Two different crack positions are employed to numerically test the proposed method. Moreover, the anomaly detection method is experimentally validated trough a fatigue crack propagation test.File | Dimensione | Formato | |
---|---|---|---|
2017_10_Colombo.pdf
non accessibile
Descrizione: Testo della tesi
Dimensione
60.1 MB
Formato
Adobe PDF
|
60.1 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/10589/136193