AI技术助力变革能源等领域材料研发范式

Group 1 - The core focus of TianGong Intelligent Materials Technology Co., Ltd. is the integration of "artificial intelligence + new materials" to drive the research paradigm from traditional trial-and-error to data-driven approaches, aiding the upgrade of the new materials industry [1] - TianGong has developed a pioneering HENaMat model for sodium-ion battery anode materials, which won second place at the 2025 "AI Navigation Cup" competition organized by the China Internet Association [1] Group 2 - Traditional lithium-ion batteries are nearing their physical energy density limits, while sodium-ion batteries, despite their potential in resources and costs, are limited by lower energy density and cycle life, primarily used in two-wheeled vehicles, low-speed vehicles, and backup power for base stations [3] - TianGong's team has constructed an artificial intelligence prediction model based on crystal graph convolutional neural networks (CGCNN) for sodium-ion battery anode materials, achieving over 90% accuracy in predicting band gaps, voltages, and specific capacities, while reducing computation time from thousands of hours to minutes, enhancing efficiency by approximately 3000 times [3] - The company has established a complete performance evaluation system that extends from band gap prediction to voltage and specific capacity assessment, providing scalable solutions for materials gene engineering, applicable not only to sodium-ion battery anode development but also extendable to lithium-ion batteries and other energy materials [3]