NIMS and SoftBank Corp. have collaborated to develop a model using machine learning techniques to predict the cycle lives of high-energy-density lithium-metal batteries. This model has been successful in accurately estimating the longevity of batteries by analyzing their charge, discharge, and voltage relaxation process data without making assumptions about specific battery degradation mechanisms. The goal of this research is to improve the safety and reliability of devices powered by lithium-metal batteries, which have the potential to achieve higher energy densities per unit mass compared to lithium-ion batteries currently in use.
The potential applications of lithium-metal batteries are vast, including use in drones, electric vehicles, and household electricity storage systems. NIMS and SoftBank established the NIMS-SoftBank Advanced Technologies Development Center in 2018 to focus on research related to high-energy-density rechargeable batteries for various systems such as mobile phone base stations, the Internet of Things, and high altitude platform stations. The collaboration has led to advancements in battery technology and the development of techniques for accurately estimating the cycle lives of lithium-metal batteries, which are crucial for ensuring the safety and efficiency of these high-performance batteries.
A lithium-metal battery with an energy density exceeding 300 Wh/kg and a life of over 200 charge/discharge cycles has already been reported, demonstrating the potential for practical use in various applications. However, the complexity of degradation mechanisms in lithium-metal batteries poses a significant challenge in developing models capable of predicting their cycle lives accurately. The research team fabricated a large number of high-energy-density lithium-metal battery cells with a lithium-metal anode and a nickel-rich cathode, using advanced battery fabrication techniques. By evaluating the charge/discharge performance of these cells and applying machine learning methods to the data, the team was able to construct a model that accurately predicts the cycle lives of lithium-metal batteries without relying on specific degradation mechanisms.
Moving forward, the team is focused on improving the accuracy of cycle life predictions using the model and accelerating efforts to commercialize high-energy-density lithium-metal batteries. This will involve leveraging the model in the development of new lithium-metal anode materials to optimize battery performance and longevity. The goal is to address the challenges associated with the complexity of degradation mechanisms in lithium-metal batteries and ensure their safe and reliable use in a wide range of technologies. By continuing to refine the model and collaborate on research and development efforts, NIMS and SoftBank aim to contribute to the advancement of battery technology and the adoption of high-performance lithium-metal batteries in various applications.
Overall, the collaboration between NIMS and SoftBank in developing a model to predict the cycle lives of lithium-metal batteries using machine learning techniques represents a significant advancement in battery technology. The success of this research has the potential to improve the safety and reliability of devices powered by lithium-metal batteries and accelerate their adoption in various technologies. By focusing on developing new materials and refining the model to enhance prediction accuracy, the team is committed to advancing the commercialization of high-energy-density lithium-metal batteries and overcoming the challenges associated with their complex degradation mechanisms.
