An experimental study of torsional vibrations in a rotating mechanical system is carried out by using a dedicated test bench, supported by advanced data acquisition and signal processing techniques. The primary objective is to identify and characterize the torsional phenomena occurring along a transmission shaft operating under varying rotational speed conditions. Based on the theoretical background of torsional vibrations and measurement techniques, a laboratory test bench was designed and constructed, consisting of an electric motor, a transmission shaft, and two inertial masses mounted at the shaft extremities. After evaluating several experimental configurations, a measurement system based on automotive magnetic sensors coupled with toothed wheels was selected as the most suitable solution. Data processing algorithms were developed in MATLAB to perform peak detection, arrival-time estimation, and rotational speed calculation. Furthermore, two different interpolation techniques were investigated and compared in order to improve the accuracy of signal peak localization. The acquired data were subsequently resampled onto a uniformly spaced time grid, enabling spectral analysis and the estimation of the Frequency Response Function (FRF). The experimental results allowed for the identification of the torsional resonance frequency of the system and revealed differences in the dynamic response of the inertial masses located at the opposite ends of the shaft. Furthermore, the interpolation method, based on a more sophisticated approach, provided improved spectral resolution and enhanced robustness to measurement noise. Overall, the results demonstrate the effectiveness of the proposed methodology for the experimental investigation of torsional vibrations, offering a solid basis for future research and condition-monitoring applications in rotating machinery.
Il presente lavoro di tesi affronta lo studio sperimentale delle vibrazioni torsionali in un sistema rotante mediante la realizzazione di un banco prova dedicato e lo sviluppo di procedure di acquisizione ed elaborazione dei dati. L’obiettivo è l’individuazione e la caratterizzazione dei fenomeni torsionali che si manifestano lungo un albero di trasmissione soggetto a variazioni di regime rotazionale. A partire da un’analisi teorica delle vibrazioni torsionali e delle tecniche di misura, è stato progettato un banco prova costituito da un motore elettrico, un albero di trasmissione e due masse poste alle estremità del sistema. Dopo la valutazione di diverse configurazioni sperimentali, è stata adottata una soluzione basata su sensori magnetici automotive accoppiati a corone dentate. L’elaborazione dei dati è stata sviluppata in ambiente MatLab mediante algoritmi per l’identificazione dei picchi, la determinazione degli Arrival time e la stima della velocità di rotazione. Sono stati inoltre confrontati due differenti metodi di interpolazione per migliorare l’accuratezza nella localizzazione dei massimi del segnale. I dati sono stati successivamente ricampionati su una griglia temporale uniforme per consentire l’analisi spettrale e la stima della Funzione di Risposta in Frequenza. I risultati hanno permesso di identificare la frequenza di risonanza torsionale del sistema e di evidenziare differenze nel comportamento dinamico delle masse alle estremità dell’albero. Inoltre, il metodo di interpolazione più avanzato ha mostrato migliori prestazioni in termini di definizione spettrale e robustezza al rumore. Lo studio conferma l’efficacia della metodologia proposta per l’analisi delle vibrazioni torsionali e fornisce una base sperimentale utile per future attività di monitoraggio e per lo sviluppo di algoritmi dedicati ai sistemi rotanti.
Progettazione e sviluppo di un banco prova per l’analisi sperimentale delle vibrazioni torsionali in sistemi rotanti
DE LUCA, MARIA CATERINA
2025/2026
Abstract
An experimental study of torsional vibrations in a rotating mechanical system is carried out by using a dedicated test bench, supported by advanced data acquisition and signal processing techniques. The primary objective is to identify and characterize the torsional phenomena occurring along a transmission shaft operating under varying rotational speed conditions. Based on the theoretical background of torsional vibrations and measurement techniques, a laboratory test bench was designed and constructed, consisting of an electric motor, a transmission shaft, and two inertial masses mounted at the shaft extremities. After evaluating several experimental configurations, a measurement system based on automotive magnetic sensors coupled with toothed wheels was selected as the most suitable solution. Data processing algorithms were developed in MATLAB to perform peak detection, arrival-time estimation, and rotational speed calculation. Furthermore, two different interpolation techniques were investigated and compared in order to improve the accuracy of signal peak localization. The acquired data were subsequently resampled onto a uniformly spaced time grid, enabling spectral analysis and the estimation of the Frequency Response Function (FRF). The experimental results allowed for the identification of the torsional resonance frequency of the system and revealed differences in the dynamic response of the inertial masses located at the opposite ends of the shaft. Furthermore, the interpolation method, based on a more sophisticated approach, provided improved spectral resolution and enhanced robustness to measurement noise. Overall, the results demonstrate the effectiveness of the proposed methodology for the experimental investigation of torsional vibrations, offering a solid basis for future research and condition-monitoring applications in rotating machinery.| File | Dimensione | Formato | |
|---|---|---|---|
|
De Luca Maria Caterina Tesi.pdf
embargo fino al 25/04/2028
Dimensione
14 MB
Formato
Adobe PDF
|
14 MB | Adobe PDF |
I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/20.500.12075/29707