Free to read: CATL is Using AI to Discover Next-Generation Battery Materials

AI could automate battery design and simulation, potentially cutting costs by 80%, says professor

CATL holds 37.6% of the global power battery market in the first seven months of 2024, according to SNE Research.

Contemporary Amperex Technology Co. Ltd. (CATL), the world’s largest battery producer, is focusing its research and development team on the use of artificial intelligence to discover the next generation of battery materials.

Zeng Yuqun, the CATL’s chairman, said his company has 20,000 people working in R&D and discovering the materials that will push forward battery innovation was their main aim. He was talking on Wednesday to Nicolai Tangen, chairman of Norway’s sovereign wealth fund Norges Bank Investment Management.

CATL holds a dominant position in the global power battery market, accounting for 37.6% of the market in the first seven months of 2024, according to South Korean research firm SNE Research.

The quest for new chemical materials is critical for improving battery performance, longevity, and cost-effectiveness. Current battery technology, such as lithium iron phosphate and ternary batteries, relies heavily on chemical materials to balance energy density, safety and cost.

The next frontier in battery innovation lies in discovering new materials. Solid-state batteries are a key area of research, necessitating the discovery of suitable solid-state electrolytes and higher energy density electrode materials.

This complex search process benefits significantly from AI-driven methodologies, which can enhance the traditionally laborious and trial-and-error approach to material discovery.

At a forum on Aug. 31, Ouyang Minggao, of the Chinese Academy of Sciences and a Tsinghua University professor, emphasized the shift from trial and error to simulation in battery R&D. This shift has already increased research efficiency by two to five times. AI technology could automate battery design and simulation, potentially reducing costs by up to 80%, he said.

AI’s role in battery development was also discussed at the World Power Battery Conference earlier this month. E. Weinan, a professor at Peking University, said the complex nature of battery materials presents significant challenges for traditional computing methods. However, AI, with its capability to handle large datasets and complex variables, offers a promising solution.

The battery design automation tool under development aims to streamline battery design, enabling developers to receive automated design recommendations based on specific performance requirements, the professor went on.

CATL is actively applying AI in its research processes. Vice President Meng Xiangfeng disclosed at the World Power Battery Conference that the company is building a supercomputing center to exploit AI in battery material selection, design and development.

Liu Jincheng, chairman of Yiwei Lithium Energy, emphasized the efficiency gains from AI in verifying complex chemical systems for battery electrolytes. By eliminating unnecessary experimental verifications, AI can significantly reduce research costs, which can otherwise range between 10 million yuan ($1.41 million) to 20 million yuan.

Sun Huajun, Chief Technology Officer of BYD’s FinDream Battery Co., noted that AI can facilitate battery cloud management to ensure optimal working conditions. Through real-time monitoring and data analysis, AI algorithms can detect abnormalities and issue early warnings, enhancing the safety and performance of battery systems, he said.