Skip to Main content Skip to Navigation

Estimation en ligne de paramètres de machines électriques pour véhicule en vue d'un suivi de la température de ses composants

Abstract : To make hybrid and full electric vehicles competitive, the performance enhancement of the embedded electric motors is essential. For this purpose, the presented work focuses in particular on the online estimation of temperature variations inside a permanent magnet synchronous motor. Therefore, we propose to estimate the temperature dependant parameters, namely the winding resistance and the magnet flux. The knowledge of these parameters or their temperature variations allows indeed to avoid performance degradation by adapting the torque control and makes possible a thermal monitoring, especially for machine availability improvement without risk of damage. Aiming at this double objective, we propose two parameter observers based on a realistic model of the machine. Thus, we first consider an electrical model of the motor that takes account of possible differences between measured and exact signals and ensures robustness to the mechanical parameters. An offline least squares algorithm for parameters identification is proposed to validate the model. Then, based on this realistic model, we design two observers, called Luenberger and Kreisselmeier. According to some observability conditions, the first one estimates the flux and position while the latter estimates one or several parameters among resistance, flux and inductance. These observers use the currents, voltages, and position for the Kreisselmeier observer, as only measurements. Theoretical and simulation studies are conducted on the observers to validate their efficiency and for a better understanding of their setting parameters. In particular, we present a study assessing the measurement noise impact on the Luenberger observer to improve its robustness to noise over a wide speed range. Finally, the implementation of the observers on different test benches provides promising results, both on parameters estimation and on the feasibility of temperature variations estimation.
Document type :
Complete list of metadata

Cited literature [54 references]  Display  Hide  Download
Contributor : ABES STAR :  Contact
Submitted on : Tuesday, March 11, 2014 - 3:47:09 PM
Last modification on : Friday, April 29, 2022 - 10:56:09 AM
Long-term archiving on: : Wednesday, June 11, 2014 - 12:50:19 PM


Version validated by the jury (STAR)


  • HAL Id : pastel-00958055, version 1


Nicolas Henwood. Estimation en ligne de paramètres de machines électriques pour véhicule en vue d'un suivi de la température de ses composants. Autre. Ecole Nationale Supérieure des Mines de Paris, 2014. Français. ⟨NNT : 2014ENMP0001⟩. ⟨pastel-00958055⟩



Record views


Files downloads