Sensorless speed and position estimation of a PMSM
Author
Hernaez Ojeda, Cesar
Term
4. term
Education
Publication year
2014
Submitted on
2014-06-03
Pages
75
Abstract
Dette projekt undersøger sensorløs estimering af hastighed og position for en overflademonteret permanentmagnet‑synkronmaskine (SPMSM) ved mellem og høje hastigheder. Der foreslås båndbredder for både stationære og dynamiske tilstande, og den resulterende vinkelpositions‑estimeringsfejl analyseres. Rotorpositionen estimeres ud fra modspænding (back‑EMF), den spænding som de roterende magneter inducerer i viklingerne, og hastigheden estimeres med et faselåst kredsløb (PLL), der følger signalets fase og frekvens. En autotuning er udviklet til begge estimationsdele for at opnå god ydeevne ud fra den valgte båndbredde og den målte positionsfejl. Motoren styres med feltorienteret regulering (FOC), og en spændingskildeomformer (VSI) med rumvektormodulation (SVM) regulerer strømmerne. Arbejdet omfatter en litteraturgennemgang for at vælge egnede sensorløse back‑EMF‑metoder, implementering og afprøvning i simulation samt laboratorietests, hvor det samlede sensorløse system valideres og sammenlignes med en sensoreret løsning.
This project investigates sensorless estimation of rotor speed and position for a surface‑mounted permanent magnet synchronous machine (SPMSM) operating at medium and high speeds. It proposes bandwidth settings for steady‑state and dynamic conditions and analyzes the resulting angular position estimation error. Rotor position is estimated from back electromotive force (back‑EMF), the voltage induced by the rotating magnets, and speed is estimated using a phase‑locked loop (PLL) that tracks signal phase and frequency. An auto‑tuning approach is designed for both estimators to achieve good performance given the chosen bandwidth and measured position error. The motor is controlled with field‑oriented control (FOC), and a voltage source inverter (VSI) with space vector modulation (SVM) regulates the currents. The work includes a literature review to select suitable back‑EMF sensorless methods, implementation and testing in simulation, and laboratory experiments that validate the complete sensorless control system and compare it with a sensored solution.
[This abstract was generated with the help of AI]
Documents