Princeton MagNet is a large-scale dataset designed to enable researchers modeling magnetic core loss using machine learning to accelerate the design process of power electronics. The dataset contains a large amount of voltage and current data of different magnetic components with different shapes of waveforms and different properties measured in the real world. Researchers may use these data as pairs of excitations and responses to build up analytical magnetic models or calculate the core loss to derive static models.
Website
Princeton MagNet is currently deployed at https://mag-net.princeton.edu/
MagNet Challenge
Download the Latest Version of the MagNet Handbook (03-25-2023)
Documentation
The web application for Princeton MagNet uses the magnet package, a python package under development where most of
the functionality is exposed. Before magnet is released on PyPI, it can be installed using
pip install git+https://github.com/PrincetonUniversity/magnet.
Please pip install mag-net and pip install . in the magnet folder before running streamlit.
How to Cite
If you used MagNet, please cite us with the following.
-
D. Serrano et al., "Neural Network as Datasheet: Modeling B-H Loops of Power Magnetics with Sequence-to-Sequence LSTM Encoder-Decoder Architecture," IEEE 23rd Workshop on Control and Modeling for Power Electronics (COMPEL), 2022.
-
H. Li, D. Serrano, T. Guillod, E. Dogariu, A. Nadler, S. Wang, M. Luo, V. Bansal, Y. Chen, C. R. Sullivan, and M. Chen, "MagNet: an Open-Source Database for Data-Driven Magnetic Core Loss Modeling," IEEE Applied Power Electronics Conference (APEC), Houston, 2022.
-
E. Dogariu, H. Li, D. Serrano, S. Wang, M. Luo and M. Chen, "Transfer Learning Methods for Magnetic Core Loss Modeling,” IEEE Workshop on Control and Modeling of Power Electronics (COMPEL), Cartagena de Indias, Colombia, 2021.
-
H. Li, S. R. Lee, M. Luo, C. R. Sullivan, Y. Chen and M. Chen, "MagNet: A Machine Learning Framework for Magnetic Core Loss Modeling,” IEEE Workshop on Control and Modeling of Power Electronics (COMPEL), Aalborg, Denmark, 2020.
Team Members
Princeton MagNet is currently maintained by the Power Electronics Research Lab as Princeton University. We also collaborate with Dartmouth College, and Plexim.
Sponsors
This work is sponsored by the ARPA-E DIFFERENTIATE Program, Princeton CSML DataX program, Princeton Andlinger Center for Energy and the Environment, and National Science Foundation under the NSF CAREER Award.
