2022
06/30
相关创新主体

 

创新背景

锂离子电池可充电却不会永远满格多次反复充电和再充电循环后会造成电池衰减进而报废。这是因为锂电池中锂离子往返穿过交换膜进行充电放电,多次进行后一部分锂离子活性下降丧失,数量减少并沉积形成电绝缘外壳,导致电池性能变差。提高电池性能保证使用续航,需要进一步对电池内部进行研究。

 

创新过程

关于电池衰减的研究,美国能源部SLAC国家加速器实验室、普渡大学、弗吉尼亚理工大学和欧洲同步加速器辐射设施的研究人员合作,利用计算机视觉技术,用机器视觉代替人眼对目标进行观察识别发现电池衰减和电极粒子作用息息相关,会随时间而变化。

研究使用X射线分析电池阴极,在10或50个充电周期后,用X射线断层扫描重建阴极的3D图像。然后将3D图片切割成一系列2D切片,使用计算机视觉方法来识别粒子。发现电池的早期衰变似乎与单个电极粒子的特性驱动有关,经过多次充电循环后,粒子之间的组合作用会影响电池衰减,随着时间推移而分解。

最后确定了2000多个单独的粒子,计算了单个粒子的特征和全局特征,研究粒子彼此接触状况和形变程度。发现10次充电循环后,单个粒子的特性如大小、形状和表面粗糙度影响电池,包括粒子的球形程度和粒子体积与表面积的比率。但在50个循环之后,配对和组属性的粒子作用推动粒子分解,包括两个粒子之间的距离、形变程度和相似取向。

粒子之间的相互作用有强相互作用、电磁相互作用、弱相互作用和引力相互作用。通过彼此之间的相互作用,会产生新粒子或发生粒子衰变等粒子转化现象。研究人员表示,电池粒子就像人一样,会有单独行动也会有群体组队,研究需要对两方面都进行观察,了解峰值效率。如果想制造更好的电池,就需要知道如何将粒子组合在一起。
利用电池粒子之间的相互作用制造电池或许是一种延长电池寿命的新方法,开发低成本快速充电的电池。

创新关键点

利用计算机视觉技术观察电池内粒子情况和相互作用,研究电池衰减原因,帮助创新制造电池的新方法。

 

Using computer vision technology to observe particles, innovative battery manufacturing technology

Regarding the study of battery decay, researchers from the U.S. Department of Energy's SLAC National Accelerator Laboratory, Purdue University, Virginia Tech University, and the European Synchrotron Radiation Facility collaborated to use computer vision technology to replace the human eye with machine vision to observe and identify targets. Battery decay is closely related to electrode particle interactions and changes over time.
The study used X-rays to analyze the battery cathode, and after 10 or 50 charging cycles, X-ray tomography was used to reconstruct a 3D image of the cathode. The 3D picture was then sliced into a series of 2D slices, and computer vision methods were used to identify the particles. It was found that the early decay of the battery appears to be driven by the properties of the individual electrode particles, and that the combined action between the particles affects the battery decay after multiple charging cycles, breaking down over time.
Finally, more than 2000 individual particles were determined, the characteristics and global characteristics of individual particles were calculated, and the contact state and deformation degree of particles were studied. After 10 charging cycles, individual particle characteristics such as size, shape and surface roughness were found to affect the battery, including the degree of sphericity of the particles and the ratio of particle volume to surface area. But after 50 cycles, particle interactions with pairing and group properties drive particle decomposition, including the distance between the two particles, the degree of deformation, and similar orientations.
There are strong interactions, electromagnetic interactions, weak interactions and gravitational interactions between particles. Through the interaction with each other, new particles or particle transformation such as particle decay occurs. The researchers said that battery particles, like people, will act alone or in groups, and research needs to observe both aspects to understand peak efficiency. If you want to make better batteries, you need to know how to put the particles together.
Using the interactions between battery particles to create batteries could be a new way to extend battery life and develop low-cost, fast-charging batteries.

智能推荐

  • 人工智能结合教师培训,自适应反馈帮助识别学习困难人群

    2022-07-28

    结合人工智能和职前教师培训,帮助学习者发现自己的学习困难并增加实践机会。

    涉及学科
    涉及领域
    关键词 
  • 开发新型芯片以提高人工智能计算效率

    2022-08-19

    斯坦福大学的工程师们创造了一种更高效、更灵活的人工智能芯片,它可以将人工智能的力量带入微型边缘设备。

    涉及学科
    涉及领域
    关键词 
  • 新技术+基础设施 | 主动维修基础设施,提高维修管理效率

    2022-08-17

    利用现代数字化技术帮助精简基础设施维修措施并进行主动预防性维护,降低维修成本,提高维修效率,进而优化公共事业管理、城市管理和工程管理模式。

    涉及学科
    涉及领域
    关键词 
  • AI+材料化学 | 创新利用机器学习缩短测试电池材料的时间

    2022-08-25

    麻省理工学院的研究人员利用机器学习工具,更快建立精确的计算机模型,帮助科学家缩减电池材料的测试时间。

    涉及学科
    涉及领域
    关键词