a Matlab toolbox for magnetization analysis

Table of contents

1. Authors
2. Statement of need
3. Setup and usage instructions
4. Description of the files and directory structure
5. External libraries
6. Explanation document
7. Examples
8. Cite as
9. Contact us
10. License
11. Acknowledgement

Authors

• Marina Zoe Petean
• Matías Ignacio González
• Tomás Francisco González
• Josefina María Silveyra
• Juan Manuel Conde Garrido

University of Buenos Aires – CONICET, Buenos Aires, Argentina.

Statement of need

The accurate description of magnetization is crucial for designing devices with soft magnetic cores. MagAnalyst is an easy-to-use MATLAB toolbox that models the magnetization of soft magnetic materials. It implements a physically based approach, proposed by Silveyra and Conde Garrido, to describe the ideal anhysteretic curve. The equation of state is simple yet accurate for various material scenarios. The toolbox first plots a set of curves to help the user determine the number of component magnetizations to model and the seeds for the fitting process. It then retrieves the model parameters without the need for programming or optimization skills. As a result, even noisy data from challenging scenarios can be accurately described by a single analytic function, with continuous and differentiable magnetization and susceptibility. We believe that MagAnalyst will be a valuable resource for the magnetic materials community.

Setup and usage instructions

You need to create a copy on a local directory on your machine to use MagAnalyst. Obtain a copy by downloading and unzipping the latest release or clone MagAanalyst instead e.g. using: git clone https://github.com/matias-gonz/mag-analyst. You can place the MagAnalyst folder anywhere on your machine.

We recommend using GitHub Desktop, which allows users to easily synchronize with the latest version of MagAnalyst.

After downloading the main folder and placing it in a suitable location, MagAnalyst is ready to use.

MagAnalyst is platform-agnostic and works on Windows, macOS, or Linux. The GUI loads sample projects from paths relative to the toolbox root (for example data/sampleData/), so you can place the MagAnalyst folder anywhere you like; if you move it after opening a sample project you may need to re-import the curve file.

To start using the application with the graphical user interface, simply run the app.mlapp file.

If you prefer to run MagAnalyst using command lines, you can follow the instructions in the mag-analyst/src/demo.m and mag-analyst/src/demo_2_components.m files.

Both demo scripts build their data paths with Matlab's fullfile helper (e.g., fullfile('data','sampleData','Finemet - TA.csv')), so they load the same files on Windows, macOS, and Linux without requiring manual path tweaks.

MagAnalyst was implemented and tested with the Matlab 2022b version on Windows 10. Authors cannot guarantee that the code can run on previous versions.

Description of the files and directory structure

├── app.mlapp                     # App controller
├── data                          # IO data
├── assets                        # Logos
├── articles                      # Articles
└── src
    ├── AnhystereticCurve         # Data and modeled anhyesteretic curve
    ├── ErrorCalculator
    ├── MagneticParameters
    ├── Parser                    # Input parser
    ├── Plotter
    ├── ResidueCalculator
    ├── minimize                  # minimize library files
    ├── interparc                 # interparc library files
    ├── colorDialog.mlapp
    └── fit.m

External libraries

MagAnalyst currently uses the following third-party libraries:

• minimize to find the constrained minimum of the objective function starting at the user’s initial estimates.
• interparc to calculate a set of equally spaced points from an original curve with unevenly spaced points.

Explanation document

In this Explanation document, we provide additional documentation and know-how to understand the background of MagAnalyst.

Examples

These are the examples provided in the data/sampleData folder with the current release:

• Mn-Zn ferrite: A soft ferrite (Mn_0.51Zn_0.44Fe_2.05O₄) with very low anisotropy. Similar to the curve analyzed in JMMM article and in AIP Advances article. The example is illustrated by a video available in YouTube, fitted with one component magnetization and two component magnetizations.
• Finemet - TA: Finemet (r) alloy (Fe_73.5Si_13.5B₉Nb₃Cu₁) nanocrystallized under transverse field annealing and measured along the longitudinal direction of the ribbon. Analyzed in JOM article.
• JNEX-900: A non-oriented electrical steel. Analyzed in AIP Advances article.
• GOSS - 67 MPa: A grain-oriented silicon steel sheet, cut parallel to the rolling direction, longitudinally compressed up to 67 MPa. Analyzed in JOM article.

The files provided include:

• data tables of either anhysteretic curves or hysteresis loops in CSV format, where the first and second column correspond to the horizontal and vertical axes field, respectively (the headers specify the fields and units)
• saved projects in TXT format, where the anhysteretic curves have been fitted with either one or two components

Cite as

If you use MagAnalyst in your work, please cite:

• J. M. Silveyra, M. I. González, T. F. González, and J. M. Conde Garrido, "MagAnalyst: A MATLAB Toolbox for Anhysteretic Magnetization Analysis," IEEE Transactions on Magnetics, 2024. https://doi.org/10.1109/TMAG.2024.3408681

You may also cite the following papers on the approach used by MagAnalyst to describe anhysteretic curves:

• J. M. Silveyra and J. M. Conde Garrido, "A physically based model for soft magnets’ anhysteretic curve," JOM, pp. 1-14, 2023. https://doi.org/10.1007/s11837-023-05704-x
• J. M. Silveyra and J. M. Conde Garrido, "On the anhysteretic magnetization of soft magnetic materials," AIP Advances, vol. 12, p. 035019, 2022. https://doi.org/10.1063/9.0000328
• J. M. Silveyra and J. M. Conde Garrido, "On the modelling of the anhysteretic magnetization of homogeneous soft magnetic materials," Journal of Magnetism and Magnetic Materials, vol. 540, p. 168430, 2021. https://doi.org/10.1016/j.jmmm.2021.168430

If you don't have access to any of these articles, request them through ResearchGate! We will be happy to share them with you.

Contact us

If you encounter bugs or problems using MagAnalyst, have any suggestions for improving the software, or are interested in receiving email updates about MagAnalyst, please write us to jsilveyra@fi.uba.ar.

License

MagAnalyst is provided under the MIT License.

Acknowledgment

The University of Buenos Aires (UBA) and CONICET provided financial support for this project. Any opinions, findings, conclusions, or recommendations expressed in this website are those of the author(s) and do not necessarily reflect the views of UBA or CONICET.