NIMS and SoftBank Corp. have joined forces to create a cutting-edge model that harnesses the power of machine learning to predict the cycle lives of high-energy-density lithium-metal batteries.
By analyzing charge, discharge, and voltage relaxation process data without making any assumptions about specific battery degradation mechanisms, this groundbreaking model accurately estimates battery longevity. This transformative technique is poised to significantly enhance the safety and reliability of devices powered by lithium-metal batteries, making it a game-changer in battery technology.
Lithium-metal batteries represent a groundbreaking advancement in battery technology, promising significantly higher energy densities compared to current lithium-ion batteries. The potential applications of these batteries span from drones to electric vehicles and household electricity storage systems.
In collaboration with SoftBank, NIMS established the NIMS-SoftBank Advanced Technologies Development Center to drive research on high-energy-density rechargeable batteries for various cutting-edge technologies, including mobile phone base stations, the Internet of Things (IoT), and high altitude platform stations (HAPS).
While a lithium-metal battery surpassing 300 Wh/kg in energy density and lasting over 200 charge/discharge cycles has been reported, the challenge lies in translating this potential into practical, safe applications.
Understanding the complex degradation mechanisms unique to lithium-metal batteries and developing accurate predictive models for their cycle lives poses a substantial hurdle. However, with continued research and development, the successful integration of high-performance lithium-metal batteries holds immense promise for the future of energy storage technology.
This pioneering research team has successfully engineered a substantial number of high-energy-density lithium-metal battery cells, featuring lithium-metal anodes and nickel-rich cathodes. Through the utilization of cutting-edge battery fabrication techniques, the team meticulously assessed the charge/discharge performance of these cells.
Moreover, they developed a groundbreaking model that leverages machine learning methods to accurately predict the cycle lives of lithium-metal batteries based on charge, discharge, and voltage relaxation process data. Remarkably, this model operates without any reliance on specific assumptions about battery degradation mechanisms.
Furthermore, the team is dedicated to enhancing the precision of cycle life predictions and expediting the practical deployment of high-energy-density lithium-metal batteries. They are committed to achieving this by harnessing the model to aid in the development of new lithium-metal anode materials.
Journal reference:
- Qianli Si, Shoichi Matsuda, Youhei Yamaji, Toshiyuki Momma, Yoshitaka Tateyama. Data-Driven Cycle Life Prediction of Lithium Metal-Based Rechargeable Battery Based on Discharge/Charge Capacity and Relaxation Features. Advanced Science, 2024; DOI: 10.1002/advs.202402608