Partenaires

Ampère

Supervisory authorities

CNRS Ecole Centrale de Lyon Université de Lyon Université Lyon 1 INSA de Lyon

Our partners

Ingénierie@Lyon



Search


Home > Thèses et HDR > Thèses en 2022

14/11/2022 - Sorelle H. NGUEDJANG KOUAKEUO

by Laurent Krähenbühl - published on , updated on

Agenda

Ajouter un événement iCal

Sorelle Nguedjang defends her PhD on Nov. 14, 2022 at 10:00AM.
Place : INSA de Lyon, Bâtiment Gustave Ferrié, amphi théâtre AE1 (Villeurbanne)

Internal characterization of magnetic cores: A route for dimensioning and real time condition monitoring

Jury :

  • Rappor-teur/trice : M. FOTSIN Hilaire (Université de Dschang, Cameroun); Mme LEBOUC Afef, G2ELab (Grenoble)
  • Examinateur/trice : Mme CAVASILA Sophie (Université Lyon 1); M. CHEUKEM André (Université de Dschang, Cameroun); M. DANIEL Laurent (Université Paris-Saclay)
  • Directeurs de thèse : M. MOREL Laurent (Ampère); M. TANYI Emmanuel (Université de Buea, Cameroun)
  • Encadrement : M. TSAFACK Pierre (Université de Buea, Cameroun)

Abstract :
A sophisticated control of equipment with magnetic components requires precise knowledge of local information on their magnetic behaviour. To this end, several approaches ranging from modelling, simulation and the setting up of various sensors have been developed to have a better physical understanding of the magnetic interactions that take place within a magnetic core. But then, the data obtained from the numerical resolutions do not have in return an experimental/practical validation. In addition, all the magnetic sensors that have been developed so far do not allow real-time in-situ monitoring of this equipment in operation. This inability is due to the geometric size of the sensors and limitations related to their instrumentation. Is it therefore possible to develop a system of on-board magnetic sensors which will not only allow the characterization of magnetic components, but which will go further by giving the possibility of controlling the behaviour of a magnetic core in real time? As a palliative to the geometric constraint, throughout our thesis, the sensors have been miniaturized using the printed circuit technique. We have therefore made good use of the enormous progress that has been made in the electronic field. Initially, the 1.5cm high magnetic tips were printed directly on the target sheet which provides information on its magnetic state when inserted into the magnetic core of electrical machines. This technique of printed magnetic tips with a thickness of 30µm (PMI) will now allow us to have magnetic signals from 1Hz to 200Hz; hence the access to information on the induction of the internal magnetic field of a magnetic core from the experimental results. With regards to the measurement of the magnetic field intensity, the technique of Giant Magnetoresistors (GMR) was used. Moreover, we have developed a 2in1 mini-wafer of 100µm thickness including GMR and PMI for simultaneous 2D measurements. The local results thus obtained and compared to the averaged induction of the encircling coil reveal the state of homogeneity in a sheet pile with a relative error of 8%. Finally, a numerical evaluation of the magnetic behavior of a laminated core by finite element simulation agrees with the experimental results. An error of 22% on the hysteretic losses parameter. Nevertheless, the variation of the hysteresis cycles throughout the various plates of the stack is the same in simulation as in measurement; hence a validation of the results obtained previously by simulation. This study undeniably contributes to the design of an intelligent magnetic core. A guarantee of real-time monitoring of the magnetic state of a material, of the magnetic core of an electric machine point by point.

Keywords :
Magnetic core condition forecast, magnetic induction sensor, non-invasive sensor, local measurement, magnetic tip technique, conductive ink, printing technique, stacks of magnetic sheets, giant magnetoresistance, finite element simulation, continuous control of magnetic cores, magnetic field distribution.



View online : Texte complet