Xiaotong Du
Fields of expertise
Modeling, design and optimization of Inductive power transfer system.
Keywords: Electromagnetic field; Finite element simulation; Machine Learning
Keywords: Electromagnetic field; Finite element simulation; Machine Learning
Short Biography
Xiaotong Du is a Ph.D. candidate in Power Electronics Laboratory at EPFL. She received the B.Sc. and M.Sc degrees from Xi'an Jiaotong University, Xi'an, China, respectively in 2016 and 2019. During the B.Sc. degree, she spent two years at École Centrale de Nantes, Nantes, France, as an double degree program student and received Dipl.-Ing degree.In October 2019, she joined Power Electronics Laboratory, EPFL, as a Doctoral Research Assistant. Her research interests include modeling, design, and optimization of inductive power transfer and resonant converters.
Professional course
Advanced Methods for Model Identification
4 credits
EPFL
01.2021
Machine Learning for Engineers
4 credits
EPFL
02.2022
Field-Based Insights into the Implementation of Renewable Energies
2 credits
Summer School Mont-Soleil
08.2021
Publications
Selected publications
Xiaotong Du; Chengmin Li; Dražen Dujić |
Design and Characterization of PCB Spiral Coils for Inductive Power Transfer in Medium-Voltage Applications |
Xiaotong Du; Dražen Dujić |
Inductive Power Transfer System with Series Connected Primary and Independent Secondary Coils |
Research
Thesis outline
The research activities during my Ph.D is around the topic " Coil link design in Inductive power transfer system as auxiliary power supply in medium voltage converters" . Specificaly, my work covers the state of art of auxiliary power supple (APS) for medium voltage (MV) converters, various IPT solutions, and the design of coil link, which includes compensation network design and the coil geometry design in both without or with ferrite scenarios. At the end of my Ph.D, the systematic design optimization flow and prototype of the IPT system is expected. The planed outline and the progress of my on-going thesis is as follows (with the related publications):1. Compensation network design as part of coil link design. Several common compensation networks are presented and analyzed, with different performance and characteristics compared. Considering the requirement of constant and independent output
among coil pairs in our application scenario, the compensation network structure is determined, and the component dimensioning rule is introduced. (Publication: https://ieeexplore.ieee.org/document/9254476)
2. Design approach of coil geometry in the condition without ferrite on coil backside based on analytic modeling of coil inductance and resistance. The brute-force search is executed in the coil geometry design space to find optimal coil geometry in perspective of
high efficiency and small coil size. (Publication: https://ieeexplore.ieee.org/document/9627815)
3. Optimization design of coil geometry in the condition with the ferrite on the coil backside. The neural network method is utilized to determine coils’ characteristics with the ferrite on their backside. Based on coil characteristics of different coil geometries, the
Pareto front of designs is searched with the genetic algorithm in coil geometry design space. (on-going work)
Teaching & PhD
Teaching and Service Activites
I have been serving as TA in various courses of EPFL in the field of power electronics:2019-2020 Spring semester: Power Electronics (EE-365), together with other TAs: Nikolina Djekanovic, Miodrag Basic, Milan Utvic, Nicolai Hildebrandt
2020-2021 Autumn semester: Industrial electronics I (EE-465), together with another TA: TA Ignacio Alejandro Polanco Lobos
2021-2022 Autumn semester: Industrial electronics I (EE-465), together with TAs: Nikolina Djekanovic, Tianyu Wei
2021-2022 Spring semester: Industrial electronics II (EE-565), together with another TA: Renan Pillon Barcelos