Researchers at the Tokyo University of Science (TUS) have developed a new way to perform the coercivity analysis of magnetic materials by using a combination of data science, machine learning (ML) and an extension of the Ginzburg-Landau model. This is an important step forward because coercivity, a physical property of magnetic materials, is key to optimizing energy efficiency in a variety of applications, including electric motors, said researchers.
“Soft magnetic materials, i.e., materials that can be easily magnetized and demagnetized, play an essential role in transformers, generators and motors,” researchers reported. “The ability of a magnetic material to resist an external magnetic field without changing its magnetization is known as coercivity, a property closely linked to the energy loss. In applications such as electric cars, low-coercivity materials are highly desirable to achieve higher energy efficiency.”
The key challenge cited by the researchers is that coercivity and other magnetic phenomena related to energy losses in soft magnetic materials originate from very complex interactions. However, the currently available tools and frameworks to analyze coercivity “mostly do not consider directly the defects and boundaries in the material, which is fundamental to develop new applications,” said researchers.
The new approach, developed by a research team including Professor Masato Kotsugi at TUS, Japan, makes it easier to link microstructural characteristics to macroscopic material properties (coercivity). This has been difficult to analyze using currently available theories that don’t account for the material’s defects and other types of inhomogeneities, they reported.
A new approach developed by TUS researchers enables the analysis of the effect of microstructures on a material’s coercivity, a magnetic property that is relevant to the development of soft magnetic materials for efficient electric motors and other devices. The method makes it easier to link microstructural characteristics to macroscopic material properties. (Source: Masato Kotsugi from Tokyo University of Science, Japan. License type: CC BY 4.0)
The team focused on finding a way to automate the coercivity analysis of magnetic materials while accounting for their microstructural characteristics. The process entailed collecting data for both simulated and real magnetic materials in the form of microscopic images of their magnetic domains.
The images, after preprocessing, were then used as input for an ML technique called principal component analysis, used to analyze large datasets, and the most relevant information or features in the images were condensed into a two-dimensional “feature space,” explained researchers.
The new approach, along with other ML techniques, such as artificial neural networks, yielded a realistic energy landscape of magnetization reversal in the material in the feature space, according to researchers. The upshot: Using ML to analyze the energy landscape showed good results for both experimental and simulated data.
The research, led by Alexandre Lira Foggiatto from TUS, was published in Communications Physics in November 2022. Researchers indicate that the new approach, beyond leveraging materials informatics for automation and clarification of coercivity in soft magnetic materials, could be used for analyzing other properties, such as temperature and strain/stress.
“Our method can be extended to other systems for analyzing properties such as temperature and strain/stress, as well as the dynamics of high-speed magnetization-reversal processes,” Foggiatto said in a statement.
The research team hopes their functional analysis models will contribute to higher efficiency in electric-vehicle motors and other sustainable transportation applications.
Advertisement
