MagNet: A Machine Learning Framework for Magnetic Core Loss Modeling

Haoran Li, Seungjae Ryan Lee, Min Luo, Charles R. Sullivan, Yuxin Chen, Minjie Chen

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

This paper presents a two-stage machine learning framework - MagNet - for magnetic core loss modeling. The first stage of MagNet is a waveform transformation network, which generates 2-D images(tensors) and extracts both the frequency and time domain features from the magnetic excitation waveforms; the second stage of MagNet is a convolutional neural network (CNN), which is trained to recognize the patterns in the 2-D images and predict the core loss based on regression. MagNet is supported by a hardware-in-the-loop (HIL) data acquisition system. The system can automatically generate a large amount of data to train the neural network models. MagNet achieved an average relative error of around 5% for single-frequency core loss prediction. In addition to experimental measurements, MagNet can also be trained with data provided on the datasheets of magnetic materials to improvethe accuracy.

Original languageEnglish (US)
Title of host publication2020 IEEE 21st Workshop on Control and Modeling for Power Electronics, COMPEL 2020
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781728171609
DOIs
StatePublished - Nov 9 2020
Event21st IEEE Workshop on Control and Modeling for Power Electronics, COMPEL 2020 - Aalborg, Denmark
Duration: Nov 9 2020Nov 12 2020

Publication series

Name2020 IEEE 21st Workshop on Control and Modeling for Power Electronics, COMPEL 2020

Conference

Conference21st IEEE Workshop on Control and Modeling for Power Electronics, COMPEL 2020
CountryDenmark
CityAalborg
Period11/9/2011/12/20

All Science Journal Classification (ASJC) codes

  • Control and Optimization
  • Modeling and Simulation
  • Energy Engineering and Power Technology
  • Electrical and Electronic Engineering

Keywords

  • convolutional neural network
  • core loss modeling
  • machine learning
  • wavelet transform

Fingerprint Dive into the research topics of 'MagNet: A Machine Learning Framework for Magnetic Core Loss Modeling'. Together they form a unique fingerprint.

Cite this