Optimization of DC Starters Based on Reluctance Network Accounting for Armature Reaction Magnetic Field

Abstract

In the new generation of “Stop & Start” starters, a DC series motor with high power density is used. In this application, non-symmetrical machines are used due to the single wave winding in brushed motors. The saturation of the iron core in this application reaches a really high level. Moreover, armature reaction in DC machine with a large stator pole spans makes this phenomenon even more enhanced. All these criteria lead to model the whole machine in highly saturated conditions which is time consuming. In this paper, a fast reluctance network (RN) model for an equivalent machine is proposed. It takes into account the armature reaction in high saturation conditions. It also allows a good accuracy and permits us to show the relevance of taking into account this phenomenon in torque calculation. This model is validated by a finite element (FE) method in many operating points, and by a statistical approach. It shows high robustness and low computation time. Finally, an optimization procedure using this reluctance network model and FE is presented in order to show the accuracy of the RN compared to FE model.

Keywords::

• micro-hybrid starter
• brushed DC motor
• reluctance network
• global optimization

Additional information

Notes on contributors

Sara Bazhar

Sara Bazhar received her Dipl. Ing. degree in Electrical Engineering from Ecole Nationale Supérieure d’Electricité et de Mécanique of Casablanca, Morocco on 2012 and her PhD degree in Electrical Engineering from Université de Lorraine on July 2017, after working on electrical machine design and modeling at Groupe de Recherche en Electrotechnique et Electronique de Nancy (GREEN) laboratory. Now, her main interests are on electrical machines and transformers design and modeling.

Julien Fontchastagner

Julien Fontchastagner received the Dipl. Ing. and the MSc degree from INP Toulouse, France, both in 2003, and the PhD degree from Toulouse University in 2007, all in Electrical Engineering. From 2003 to 2008, he was with the Laboratory of Plasma and Energy Conversion (LAPLACE), at Université de Toulouse. Currently, he is with the Groupe de Recherche en Electrotechnique et Electronique de Nancy (GREEN), at Université de Lorraine, as an Associate Professor, where he works on solving inverse problems for electrical devices design. His research interestsinclude various topics of electric power engineering.

Noureddine Takorabet

Noureddine Takorabet received the Master’s degree in Electrical Engineering from University of Nancy, France, in 1994 and PhD from INPL in 1996. He is currently Professor at Université de Lorraine, member of the laboratory Group of Research in Electrical Engineering of Nancy (GREEN), France. In particular, he is interested in electromechanical conversion. His main research activities deal with electromagnetic devices modeling and optimization.

Nicolas Labbe

Nicolas Labbe is graduated from the INPG, Institut National Polytechnique de Grenoble, where he earned his PhD in Electrical Engineering in 1996, after earning there in 1993 both his Dipl. Ing. and MSc degrees. He joined VALEO Electrical Systems in 1999, a French company specialized in automotive powertrain systems. Currently, he is Electrical Engineering Manager for starter motors, and Valeo Master Expert. His research interests include various topics of electric power engineering.

Raphaël Andreux

Raphaël Andreux was with the Valeo company, St Quentin Fallavier, France, from 2010 to 2015. From 2010 to 2012, he was a PhD student with University of Lorraine, France, in collaboration with the Valeo company. In 2013, he received his PhD degree in electrical engineering. His research fields were on electromagnetism, brushed DC motor, and optimization methods applied on electromagnetic devices.