观点
电池(电)
计算机科学
班级(哲学)
领域(数学)
系统工程
人工智能
工程类
物理
声学
数学
量子力学
功率(物理)
纯数学
作者
Aysegul Kilic,Burcu Oral,Damla Eroğlu,Ramazan Yıldırım
标识
DOI:10.1016/j.est.2023.109057
摘要
New alternatives for various elements of Li-ion battery (LIB) have been investigated to overcome its limitations, creating a new class of batteries called beyond LIBs. There has been a great deal of effort to find more functional materials and better design parameters to improve the performance of these systems, and machine learning (ML) is also used extensively for this purpose. We assessed the state of the art in beyond LIBs, briefly reviewed ML applications, and provided our viewpoints on the potential contribution of ML to the field. It seems that, while the diversity of the materials, parameters, and problems in different beyond LIB systems should be taken into account, the new approaches, including the construction of battery-specific databases or other data-sharing mechanisms, should be developed to transfer the knowledge from more frequently studied systems to others as well as from the material-level performance to system-level performance.
科研通智能强力驱动
Strongly Powered by AbleSci AI