Researchers Map Hidden Energy Loss in Electric Motor Magnets
Energy waste from magnetic domain reversal in electric motors costs the industry billions annually. Researchers at Tokyo University of Science have developed a method to identify and measure this loss by analyzing the internal magnetic structures that conventional approaches miss.
When magnetic fields inside a motor reverse direction repeatedly, they generate heat within the soft magnetic materials that form the motor core. This process, called iron loss or magnetic hysteresis loss, becomes worse at high operating temperatures, where thermal effects partially demagnetize the materials and compound energy waste.
The problem centers on magnetic domains-tiny magnetic regions whose arrangement directly affects how materials respond to heat and energy loss. Some soft magnetic materials contain intricate structures called maze domains, named for their zig-zag, labyrinth-like appearance. These structures change abruptly with temperature shifts, but scientists have struggled to understand them because multiple interacting factors are involved.
Physics-Based AI Framework Reveals Energy Barriers
A team led by Professor Masato Kotsugi and Dr. Ken Masuzawa from Tokyo University of Science, working with collaborators from three other Japanese universities, developed the entropy-feature-eXtended Ginzburg-Landau (eX-GL) model to map these structures. The approach combines physics-based analysis with machine learning to explain how temperature affects magnetization reversal.
"Conventional simulations oversimplify real materials, while experiments reveal complexity without a clear way to quantify cause and effect," Kotsugi said. "Our framework addresses these limitations and mechanistically explains the temperature-dependent magnetization reversal process."
The researchers captured microscopic images of magnetic domains in a rare-earth iron garnet sample at different temperatures, then analyzed them using the eX-GL model. The first stage used persistent homology, a mathematical method that identifies topological features in the image data to detect uneven structural characteristics. Machine learning pattern recognition then determined which features mattered most, producing a digital free-energy map showing how magnetic microstructures evolve as energy changes.
The analysis identified four major energy barriers that strongly influence magnetization reversal. These barriers involve exchange interactions, demagnetizing effects, and entropy-the measure of disorder in a system.
Complexity Driven by Competing Forces
The team discovered that maze domains grow more complex as domain wall length increases. This complexity results from interactions between entropy and exchange forces-the physical forces that align magnetic spins within the material.
The findings appear in Scientific Reports. Because free energy is a universal thermodynamic metric, the eX-GL approach can be applied to other systems with similar characteristics beyond magnetic materials.
For professionals working in materials science or motor design, the method offers a way to identify efficiency losses that remain invisible to standard analysis techniques. The approach automates interpretation of complex magnetization processes, revealing mechanisms difficult to detect using conventional methods.
Learn more about AI for Science & Research and AI Data Analysis applications in your field.
Your membership also unlocks:
