Artificial Intelligence-Driven Prediction and Classification of Cube and Goss Texture Evolution in Pure Iron, Electrical Steels, and Ultra-Low Carbon Steels: An Integrated Framework Using Multimodal Microstructure Characterization and Intelligent Materials Informatics

Unofficial AI-generated summary based on the public title and abstract. Not an official translation.

🔬研究者:Materials informatics and AI method applied to crystallographic texture prediction, offering a data-driven framework for computational materials science.

🏢実務担当者:Steel manufacturers can use the AI framework for real-time quality control and process optimization of electrical steel production.

🏛政策担当者:Supports energy efficiency and decarbonization through advanced materials manufacturing, relevant for industrial policy and green innovation.

Alternative Title 1 Artificial Intelligence for Crystallographic Texture Engineering in Pure Iron, Electrical Steels, and Ultra-Low Carbon Steels Subtitle A Comprehensive Framework for Cube and Goss Texture Prediction Using EBSD, XRD, SEM, TEM, Optical Microscopy, and Intelligent Microstructure Analytics Alternative Title 2 AI-Based Intelligent Texture Recognition and Microstructure Evolution in Pure Iron, Electrical Steels, and Ultra-Low Carbon Steels Subtitle Machine Learning Approaches for Automated Cube and Goss Texture Classification, Recrystallization Analysis, and Property Prediction Alternative Title 3 Artificial Intelligence-Assisted Microstructure Evolution and Texture Prediction in Ferrous Materials Subtitle Applications of Deep Learning and Materials Informatics for Cube and Goss Texture Optimization in Pure Iron and Electrical Steels Alternative Title 4 Machine Learning Framework for Cube and Goss Texture Evolution in Pure Iron, Electrical Steels, and Ultra-Low Carbon Steels Subtitle Combining Multimodal Microstructure Characterization with Physics-Informed Artificial Intelligence for Intelligent Texture Engineering Alternative Title 5 Intelligent Microstructure Characterization Using Artificial Intelligence Subtitle Automated Prediction of Cube and Goss Texture Evolution through Optical Microscopy, SEM, TEM, EBSD, and XRD Alternative Title 6 Physics-Informed Artificial Intelligence for Texture Evolution in Ferrous Materials Subtitle A Unified Framework for Predicting Cube and Goss Texture Development, Grain Growth, Recrystallization, and Magnetic Properties Alternative Title 7 AI-Driven Crystallographic Texture Prediction for Advanced Electrical Steels Subtitle Deep Learning-Based Analysis of Cube and Goss Texture Evolution Using EBSD and Multiscale Microstructure Characterization Alternative Title 8 Artificial Intelligence and Materials Informatics for Intelligent Texture Engineering Subtitle Automated Recognition of Cube and Goss Texture in Pure Iron, Ultra-Low Carbon Steels, and Grain-Oriented Electrical Steels Alternative Title 9 Digital Texture Engineering Using Artificial Intelligence Subtitle An Intelligent Framework Integrating EBSD, SEM, TEM, XRD, and Machine Learning for Crystallographic Texture Prediction Alternative Title 10 Towards Autonomous Texture Characterization in Ferrous Materials Subtitle Artificial Intelligence, Computer Vision, and Digital Twins for Cube and Goss Texture Recognition and Microstructure Evolution Alternative Title 11 Artificial Intelligence for Intelligent Texture Design and Microstructure Evolution Subtitle Applications in Pure Iron, Electrical Steels, Ultra-Low Carbon Steels, and Advanced Magnetic Materials Alternative Title 12 Deep Learning-Based Texture Informatics for Ferrous Materials Subtitle Automated Cube and Goss Texture Prediction through Multimodal Microstructure Characterization Alternative Title 13 Artificial Intelligence Meets Physical Metallurgy Subtitle Predicting Crystallographic Texture Evolution and Material Properties Using EBSD, XRD, SEM, TEM, and Machine Learning Alternative Title 14 Computational Texture Engineering Using Artificial Intelligence Subtitle A Data-Driven Framework for Cube and Goss Texture Optimization in Pure Iron and Electrical Steel Processing Alternative Title 15 Next-Generation Intelligent Texture Analysis Subtitle Integrating Deep Learning, Physics-Informed Neural Networks, and Microstructure Characterization for Automated Materials Engineering Detailed Description Abstract-Level Description The crystallographic texture of engineering materials plays a fundamental role in determining their magnetic, mechanical, electrical, and formability characteristics. In body-centered cubic (BCC) ferrous materials such as Pure Iron, Grain-Oriented Electrical Steels (GOES), Non-Grain-Oriented Electrical Steels (NGOES), and Ultra-Low Carbon (ULC) Steels, the evolution of Cube ({001}<100>) and Goss ({110}<001>) textures governs critical performance metrics including magnetic permeability, core loss, deep drawability, yield strength, recrystallization behavior, and anisotropy. Traditional characterization methods—Optical Microscopy, Scanning Electron Microscopy (SEM), Transmission Electron Microscopy (TEM), Electron Backscatter Diffraction (EBSD), and X-ray Diffraction (XRD)—provide high-resolution structural information but often require significant manual analysis, specialized expertise, and substantial processing time. This research proposes a unified Artificial Intelligence (AI)-based framework that integrates multimodal microstructural characterization with advanced data-driven algorithms to enable automated recognition, classification, and prediction of Cube and Goss texture evolution. The framework combines image processing, machine learning, deep learning, computer vision, graph neural networks, vision transformers, and Physics-Informed Neural Networks (PINNs) to analyze EBSD orientation maps, SEM and TEM micrographs, optical microscopy images, and XRD diffraction patterns. By learning complex relationships between thermomechanical processing parameters, microstructure evolution, crystallographic texture, and resulting material properties, the system provides rapid and reliable predictions of recrystallization, grain growth, texture intensity, orientation distribution, and magnetic and mechanical performance. The proposed methodology supports intelligent texture engineering by enabling real-time microstructure assessment, automated defect detection, predictive process optimization, and digital twin development for industrial steel processing. It is applicable to rolling, annealing, secondary recrystallization, and heat-treatment processes used in electrical steels and advanced ferrous alloys. Beyond improving characterization efficiency, the framework establishes a pathway toward autonomous materials laboratories and AI-assisted physical metallurgy, where experimental observations, computational models, and data-driven intelligence operate within a unified materials informatics ecosystem. The research also explores the integration of explainable AI, uncertainty quantification, and physics-based constraints to ensure that predictions remain scientifically interpretable and physically consistent. The resulting framework provides a foundation for next-generation texture engineering, intelligent manufacturing, and accelerated materials discovery, making it highly relevant to advanced research in materials science, metallurgical engineering, computational mechanics, and artificial intelligence. Research Theme Artificial Intelligence → Multimodal Microstructure Characterization → Texture Evolution → Materials Informatics → Intelligent Texture Engineering → Property Prediction → Digital Twin → Autonomous Materials Processing Novel Scientific Contributions A multimodal AI framework integrating Optical Microscopy, SEM, TEM, EBSD, and XRD for comprehensive texture analysis. Automated identification and quantification of Cube ({001}<100>) and Goss ({110}<001>) textures in Pure Iron, Electrical Steels, and Ultra-Low Carbon Steels. AI-based prediction of microstructure evolution, including recrystallization, grain growth, and abnormal grain growth during thermomechanical processing. Fusion of computer vision, graph neural networks, vision transformers, and Physics-Informed Neural Networks to improve texture prediction accuracy while maintaining physical consistency. Correlation of crystallographic texture with magnetic, mechanical, and forming properties, enabling predictive optimization of electrical steel performance. Integration with digital twin technology for intelligent process control, online quality monitoring, and autonomous texture engineering in industrial manufacturing. A scalable materials informatics platform that accelerates alloy development, reduces experimental effort, and supports data-driven decision-making in advanced ferrous materials research.

• openalex https://doi.org/10.5281/zenodo.21369618first seen 2026-07-17 05:04:18 · last seen 2026-07-17 05:04:19