The purpose of this thesis work is to study a method for identifying the delay introduced by an angular position sensor of an IPM motor. This work aims in particular to evaluate its performance in different cases, particularly the case in which there are no low-pass filters after the sensors, which is the ideal case, and the case in which these are present, assessing the different results. The study begins with modeling an IPM drive and analyze of the electromechanical dynamics, followed by an analysis of the algorithm used to identify and estimate the delay introduced by the sensor, with a study of the foundations on which the algorithm is based. The results of these simulations are shown, demonstrating how in the ideal case, the algorithm functions perfectly. In the case where low-pass filters are present, the results of the delay estimation worsen as the value of the delay coefficient decreases. This is because the ratio between the delay coefficient and the distortion introduced by the filters decreases as the delay decreases. Nevertheless, within the range of delay values studied, there is still an improvement in performance in either case, although the improvement decreases as the delay decreases in the case with low-pass filters. With further studies able to separate the delay introduced by the filter from that of the sensor, this algorithm will certainly find its place in the industrial field.
Il presente lavoro di tesi mira a studiare un metodo per identificare il ritardo introdotto da un sensore di posizione angolare in un motore IPM (Interior Permanent Magnet), valutandone le prestazioni in differenti scenari. Particolare attenzione è dedicata a due casi distinti: uno in cui non sono presenti filtri passa-basso dopo i sensori, configurazione considerata ideale, e uno in cui tali filtri sono attivi. Inizialmente, viene condotta un’analisi della dinamica elettrica e meccanica del motore IPM, seguita dall’esame dell’algoritmo impiegato per identificare e stimare il ritardo sensoriale, con una discussione delle basi teoriche su cui si fonda tale algoritmo. Successivamente, vengono presentati i modelli Simulink utilizzati per condurre le simulazioni nei due scenari sopra citati. I risultati delle simulazioni evidenziano il corretto funzionamento dell’algoritmo nel caso ideale, mentre nel caso con filtri passa-basso si osserva un deterioramento delle stime del ritardo all’aumentare della riduzione del coefficiente di ritardo. Questo è dovuto al diminuire del rapporto tra il coefficiente di ritardo e la distorsione introdotta dai filtri, che diminuisce all’aumentare del ritardo. Tuttavia, nell’intervallo di valori di ritardo considerati nello studio, si osserva comunque un miglioramento delle prestazioni in entrambi i casi, sebbene il miglioramento sia meno significativo nel caso con filtri passa-basso. Con ulteriori approfondimenti volti a separare il ritardo introdotto dal sensore da quello introdotto dal filtro, si prevede che questo algoritmo possa trovare ampio impiego nell’ambito industriale.
Self-identification of sensor's delay angle for IPMSM motor control
Fantozzi, Giacomo
2022/2023
Abstract
The purpose of this thesis work is to study a method for identifying the delay introduced by an angular position sensor of an IPM motor. This work aims in particular to evaluate its performance in different cases, particularly the case in which there are no low-pass filters after the sensors, which is the ideal case, and the case in which these are present, assessing the different results. The study begins with modeling an IPM drive and analyze of the electromechanical dynamics, followed by an analysis of the algorithm used to identify and estimate the delay introduced by the sensor, with a study of the foundations on which the algorithm is based. The results of these simulations are shown, demonstrating how in the ideal case, the algorithm functions perfectly. In the case where low-pass filters are present, the results of the delay estimation worsen as the value of the delay coefficient decreases. This is because the ratio between the delay coefficient and the distortion introduced by the filters decreases as the delay decreases. Nevertheless, within the range of delay values studied, there is still an improvement in performance in either case, although the improvement decreases as the delay decreases in the case with low-pass filters. With further studies able to separate the delay introduced by the filter from that of the sensor, this algorithm will certainly find its place in the industrial field.| File | Dimensione | Formato | |
|---|---|---|---|
|
2024_04_Fantozzi.pdf
accessibile in internet solo dagli utenti autorizzati
Dimensione
3.1 MB
Formato
Adobe PDF
|
3.1 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/10589/217252