Contribution à l'étude des machines synchrones à aimants permanents en présence de défauts inter-spires : modélisation, détection de défauts inter-spires

Abstract : Embedded electric systems in aircraft applications have to satisfy to specifications including weight, performances and reliability leading to the use of Permanent Magnet Synchronous Machines (PMSM). Due to these constraints, prototypes are often different from classical PMSM. Online monitoring of these systems is challenging but is still important regarding safety applications (electro mechanical airfoil, braking?). The main aim is to detect major incidents in order to use redundant systems. This work concerns modeling of PMSM under both healthy and inter-turn fault cases in order to defined online fault detection methods without additional sensors. Two approaches are presented and compared to experimental tests concerning PMSM under inter-turn fault conditions. One of them is based on Permeance Network (PN) modelling. It can be considered to be the best compromise. The next part deals with two fault indicators methods based on healthy models of Inverter and PMSM and shows improvements brought by these approaches compared to classical methods. The whole indicators are analyzed using Pattern Recognition (PR)
Document type :
Theses
Complete list of metadatas

Cited literature [122 references]  Display  Hide  Download

https://hal.univ-lorraine.fr/tel-01750066
Contributor : Thèses Ul <>
Submitted on : Thursday, March 29, 2018 - 12:37:00 PM
Last modification on : Friday, August 24, 2018 - 6:36:44 PM
Long-term archiving on : Friday, September 14, 2018 - 5:37:08 PM

File

DDOC_T_2012_0364_LEBOEUF.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : tel-01750066, version 1

Collections

Citation

Nicolas Leboeuf. Contribution à l'étude des machines synchrones à aimants permanents en présence de défauts inter-spires : modélisation, détection de défauts inter-spires. Autre. Université de Lorraine, 2012. Français. ⟨NNT : 2012LORR0364⟩. ⟨tel-01750066⟩

Share

Metrics

Record views

77

Files downloads

79