The present thesis is focused on developing innovative methodologies to be applied in multi-axis vibration control testing. This test practice can closely replicate via shakers excitation the vibration environment that a structure needs to withstand during its operational life. In the field of multi-axis dynamic testing, this thesis investigates the critical aspects related to the definition process of the Multiple Input Multiple Output (MIMO) control target. In MIMO random control tests in fact, the control target is a full reference Spectral Density Matrix (SDM) in the frequency band of interest. The diagonal terms are the Power Spectral Densities (PSDs), representative for the acceleration operational levels at critical control locations. The off-diagonal terms are the Cross Spectral Densities (CSDs), representative for the cross-correlation that needs to be replicated between pairs of control channels. De facto, test specifications are often given in terms of PSDs only, relying on the legacy of single-axis testing. Information about the CSDs is often missing. However, an accurate definition of the CSD profiles can enhance the MIMO random testing, as these terms influence both the specimen responses (strain/stress) and the shaker's voltages (drives). The challenges are linked to the algebraic constraint that the resulting reference SDM must be positive (semi)definite, in order to be realizable and controllable by the control algorithm. In this context, the thesis proposes three newly developed methods which cope with the MIMO control target issues. The present methods are based on an innovative phase selection algorithm, which ensures the compliance with all the exiting physical dependencies, and it provides realizable control targets. In addition to meet the existing mathematical constraints, the proposed techniques offer further features that improve the testing practice. Two novel techniques, namely Minimum Drives Method (MDM) and Minimum Single Drive Method (MSDM), are focused on the drives power minimisation. The objective is to replicate the fixed test specifications without stressing the delicate excitation system. The third novel technique instead, namely Extreme Dynamic Response Method (EDRM), aims to combine the test specifications to get out the maximum dynamic response from the specimen. In this case, the objective is to excite the specimen with the most demanding multi-axial vibration environment, thus reducing the time-to-failure. The detailed analytic derivation and implementation steps of the proposed methods are followed by real-life testing, carried out with the three-axial electro-dynamic excitation system at the University of Ferrara. This avant-garde test facility is composed of three independent shakers of 10 kN to excite test specimens in three orthogonal directions simultaneously. The experimental results validate the novelties of this thesis. Both the MDM and the MSDM exhibit high performances on preserving the excitation system from DAC overloads. In particular, the MDM equally reduces all the multiple drives, and it guarantees the replication of the test specifications by exploiting the minimum power of the excitation system. The MSDM instead, capitalises its full potential when most of the power is required by a single drive of the multiple shakers system. It improves the safety performances by reducing the power of the overworked drive. Finally, the EDRM points out significant benefits when performing accelerated multi-axial fatigue tests. It ensures the shortest fatigue-life duration of the specimen thanks to the proper combination of the user-defined test specifications. These original features open the possibility to candidate the proposed MIMO target generation methods as innovative solutions to be included in the standard practice for multi-axial vibration control testing.

La presente tesi è focalizzata sullo sviluppo di metodologie innovative da applicare nei test dinamici controllati in cui le vibrazioni sono di tipo multi-assiale. L’obbiettivo dei test dinamici controllati è quello di replicare fedelmente in laboratorio l'ambiente vibrazionale che l’oggetto dovrà successivamente sopportare durante la sua vita operativa. Mediante tale tesi, si vuole dunque indagare gli aspetti critici relativi al processo di definizione del target di tipo MIMO (Multiple Input Multiple Output). Nei test di controllo MIMO, il target viene definito come SDM (Spectral Density Matrix), i cui termini diagonali sono delle PSDs (Power Spectral Densities) rappresentative dei livelli di accelerazione nei punti di controllo. I termini fuori dalla diagonale invece, sono delle CSDs (Cross Spectral Densities) rappresentative della correlazione tra le varie coppie di canali controllati. Di fatto, le specifiche dei test sono spesso fornite solo in termini di PSDs, e le informazioni relative alle CSDs sono mancanti. Una definizione accurata dei profili CSD può tuttavia migliorare il test controllato, poiché tali termini influenzano sia le risposte dell’oggetto (deformazione/sollecitazione), sia le potenze utilizzate dagli shaker per eseguire il test. In tale contesto, la tesi propone tre metodi di generazione del target MIMO basati su un innovativo algoritmo di selezione delle fasi che garantisce il rispetto di tutti i vincoli esistenti, sia matematici che fisici. Le tecniche proposte offrono ulteriori funzionalità che migliorano lo svolgimento dei test. In particolare, il Minimum Drives Method (MDM) e il Minimum Single Drive Method (MSDM), si concentrano sulla minimizzazione della potenza richiesta dagli shaker. L'obiettivo è replicare le specifiche del test senza stressare il delicato e costoso sistema d'eccitazione. Il terzo metodo proposto invece, l’Extreme Dynamic Response Method (EDRM), mira a combinare le specifiche del test per ottenere la massima risposta dinamica dell’oggetto in prova. In questo caso, l'obiettivo è quello di sottoporre l’oggetto alle condizioni di prova più danneggianti possibili, riducendo così il tempo totale di guasto. Per la validazione delle metodologie proposte, è stata condotta una compagna sperimentale eseguita con lo shaker triassiale installato presso i laboratori del Dipartimento di Ingegneria dell'Università di Ferrara. Si tratta di un sistema all'avanguardia composto da tre shaker elettrodinamici da 10 kN ciascuno, in grado di eccitare simultaneamente gli oggetti nelle tre direzioni ortogonali. I risultati sperimentali confermano che, sia l’MDM che l’MSDM, garantiscono elevate prestazioni nel preservare il sistema di prova da sovraccarichi di potenza. In particolare, l’MDM riduce in egual modo tutti i drive del sistema e replica le specifiche sfruttando la minima potenza complessiva. L'MSDM, invece, sfrutta al meglio il suo potenziale quando gran parte della potenza viene richiesta dal singolo drive del sistema di prova. In questi casi, l’MSDM è in grado di ridurre la potenza dell'azionamento sovraccaricato salvaguardando così l’intera apparecchiatura. Infine, l'EDRM evidenzia vantaggi significativi durante l'esecuzione di prove a fatica multi-assiale. Il metodo proposto, grazie alla corretta combinazione delle specifiche, garantisce sempre la minor durata del componente in esame, sintomo che l'oggetto è stato testato nelle condizioni più danneggianti possibili. Grazie alle caratteristiche di originalità ed affidabilità mostrate durante la campagna sperimentale, è possibile candidare i tre metodi proposti in questa tesi come soluzioni innovative da includere nelle procedure standard per l’esecuzione dei test dinamici controllati di tipo multi-assiale.

Target Generation Techniques for Multi-Axis Random Vibration Control: Development, Implementation and Experimental Validation

D'ELIA, GIACOMO
2021

Abstract

The present thesis is focused on developing innovative methodologies to be applied in multi-axis vibration control testing. This test practice can closely replicate via shakers excitation the vibration environment that a structure needs to withstand during its operational life. In the field of multi-axis dynamic testing, this thesis investigates the critical aspects related to the definition process of the Multiple Input Multiple Output (MIMO) control target. In MIMO random control tests in fact, the control target is a full reference Spectral Density Matrix (SDM) in the frequency band of interest. The diagonal terms are the Power Spectral Densities (PSDs), representative for the acceleration operational levels at critical control locations. The off-diagonal terms are the Cross Spectral Densities (CSDs), representative for the cross-correlation that needs to be replicated between pairs of control channels. De facto, test specifications are often given in terms of PSDs only, relying on the legacy of single-axis testing. Information about the CSDs is often missing. However, an accurate definition of the CSD profiles can enhance the MIMO random testing, as these terms influence both the specimen responses (strain/stress) and the shaker's voltages (drives). The challenges are linked to the algebraic constraint that the resulting reference SDM must be positive (semi)definite, in order to be realizable and controllable by the control algorithm. In this context, the thesis proposes three newly developed methods which cope with the MIMO control target issues. The present methods are based on an innovative phase selection algorithm, which ensures the compliance with all the exiting physical dependencies, and it provides realizable control targets. In addition to meet the existing mathematical constraints, the proposed techniques offer further features that improve the testing practice. Two novel techniques, namely Minimum Drives Method (MDM) and Minimum Single Drive Method (MSDM), are focused on the drives power minimisation. The objective is to replicate the fixed test specifications without stressing the delicate excitation system. The third novel technique instead, namely Extreme Dynamic Response Method (EDRM), aims to combine the test specifications to get out the maximum dynamic response from the specimen. In this case, the objective is to excite the specimen with the most demanding multi-axial vibration environment, thus reducing the time-to-failure. The detailed analytic derivation and implementation steps of the proposed methods are followed by real-life testing, carried out with the three-axial electro-dynamic excitation system at the University of Ferrara. This avant-garde test facility is composed of three independent shakers of 10 kN to excite test specimens in three orthogonal directions simultaneously. The experimental results validate the novelties of this thesis. Both the MDM and the MSDM exhibit high performances on preserving the excitation system from DAC overloads. In particular, the MDM equally reduces all the multiple drives, and it guarantees the replication of the test specifications by exploiting the minimum power of the excitation system. The MSDM instead, capitalises its full potential when most of the power is required by a single drive of the multiple shakers system. It improves the safety performances by reducing the power of the overworked drive. Finally, the EDRM points out significant benefits when performing accelerated multi-axial fatigue tests. It ensures the shortest fatigue-life duration of the specimen thanks to the proper combination of the user-defined test specifications. These original features open the possibility to candidate the proposed MIMO target generation methods as innovative solutions to be included in the standard practice for multi-axial vibration control testing.
DALPIAZ, Giorgio
MUCCHI, Emiliano
TRILLO, Stefano
File in questo prodotto:
File Dimensione Formato  
PhD_Thesis_DELIA.pdf

accesso aperto

Descrizione: PhD Thesis DELIA
Tipologia: Tesi di dottorato
Dimensione 13.34 MB
Formato Adobe PDF
13.34 MB Adobe PDF Visualizza/Apri

I documenti in SFERA sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11392/2478822
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact