Индийские ученые создали крупнейшую базу данных сплавов на базе искусственного интеллекта

Researchers at the Indian Institute of Technology Madras (IIT Madras) have developed an Artificial Intelligence platform to accelerate the discovery of sustainable, high-performance metallic alloys for applications including electric vehicles, aerospace, renewable energy and marine infrastructure. The researchers have built one of the world's largest publicly available databases of advanced metallic alloys by using Large Language Models (LLMs) to extract and organise information from more than 10,000 scientific research papers. The databases and supporting software are available through the Alloy Tattvasar platform and GitHub for researchers, startups and industries. AI system extracts data from scientific literature The IIT Madras team developed an automated AI pipeline to extract alloy compositions, manufacturing processes and testing conditions from scientific literature. The system can process information covering more than 350 material properties, while retaining the conditions under which individual measurements were obtained. The work addresses a challenge in materials research, where experimental information is distributed across journal articles, tables and figures. Manual extraction and organisation of this information can require substantial time and may make comparisons between different materials difficult. The platform uses Retrieval-Augmented Generation (RAG) to retrieve relevant examples during information extraction. According to the researchers, this approach improves the extraction of information from scientific text and tables. Database contains more than 185,000 records The platform has generated two databases containing more than 185,000 structured records. IIT Madras said these are the world's largest publicly available multicomponent alloy databases. The research also incorporates environmental, economic and social indicators alongside material performance data. This allows candidate alloys to be assessed using both technical characteristics and sustainability-related parameters. The researchers demonstrated the database using high-entropy alloys for three application areas. These included lightweight structural materials for automotive and aerospace applications, soft magnetic materials for electric motors and transformers, and corrosion-resistant alloys for marine infrastructure, offshore engineering, chemical processing and energy systems. Research team and funding The research was conducted by Aravindan Kamatchi Sundaram, Mohit Chakraborty, Sai Mani Kumar Devathi and B. Pabitramohan Prusty under the guidance of Dr Rohit Batra, Assistant Professor in the Department of Metallurgical and Materials Engineering at IIT Madras. The project received funding from the Anusandhan National Research Foundation (ANRF), the Defence Research and Development Organisation's Directorate of Industry and Academia (DRDO-DIA), and the Wadhwani School of Data Science and AI at IIT Madras. Computational resources were provided by the Robert Bosch Centre for Data Science and AI (RBCDSAI). The findings have been published in Advanced Science. The research team plans to expand the system to extract information from figures and microstructural images and incorporate life-cycle assessment methods. Future work will also cover polymers, ceramics and composite materials.