摘要
本文基于多相流测试、数值仿真和深度学习方法开展了气固流化床数字孪生建模与应用研究。首先，基于气固流化床的工作原理及流程过程，搭建了一种多层次的流化床数字孪生框架；进一步重点在功能应用层分析了数字孪生体在设备故障诊断、参数控制优化等场景下的核心技术以及应用实现路径；最后，从数字孪生体构建的标准化、业务化及可视化等层面总结了气固流化床装置数字孪生技术面临的挑战。本研究可为气固流化床设备的信息化、数字化及运维管控的科学化提供理论参考。
关键词： 气固流化床；数字孪生；多相流测试；数值模拟；深度学习
Abstract
Based on multiphase flow testing, numerical simulation and deep learning methods, digital twin modeling and application research of gas-solid fluidized bed were carried out in this paper. Firstly, based on the working principle and flow process of gas-solid fluidized bed, a multi-level digital twin framework of fluidized bed is built. In the functional application layer, the core technology and application path of digital twin in the scenario of equipment fault diagnosis and parameter control optimization are analyzed. Finally, the challenges of digital twin technology in gas-solid fluidized bed units are summarized from the aspects of standardization, operation and visualization of digital twin construction. This study can provide theoretical reference for the information, digitization and scientific operation and maintenance control of gas-solid fluidized bed equipment.
Key words： Gas-solid fluidized bed; Digital twins; Multiphase flow test; Numerical simulation; Deep learning
参考文献 References
[1] 肖荣美. 《工业能效提升行动计划》专家解读系列文章之五：数字化绿色化协同促进工业节能提效. 2022.
[2] 陶飞,刘蔚然,刘检华等.数字孪生及其应用探索[J].计算机集成制造系统,2018,24(01):1-18.
[3] 贺兴,艾芊,朱天怡等.数字孪生在电力系统应用中的机遇和挑战[J].电网技术,2020,44(06):2009-2019
[4] 沈沉,贾孟硕,陈颖等.能源互联网数字孪生及其应用[J].全球能源互联网,2020,3(01):1-13.
[5] 余斌,朱伟佳.石化行业数字孪生技术的应用探索[J].化工进展,2019,38(S1):278-282.
[6] Pitchaikani A, P.K., Strandberg M, et al., Liquid cooling applications in twin builder-industrial paper in Tokyo[J]: The 2nd Japanese Modelica Conference. 2018.
[7] 安世亚太科技股份有限公司数字孪生体实验室, 数字孪生体技术白皮书 (2019) 2019: 北京.
[8] Fourgeau, E., et al., System Engineering Workbench for Multi-views Systems Methodology with 3DEXPERIENCE Platform. The Aircraft RADAR Use Case[J]. Springer International Publishing, 2016.
[9] Ge, W., L. Guo, and J. Li, Toward Greener and Smarter Process Industries[J]. Engineering, 2017. 3(2): p. 152-153.
[10] 金涌,程易,白丁荣等.中国流态化技术研发史略[J].化工进展,2023,42(06):2761-2780.
[11] 杨越林. Beckmann重排流化床反应器的冷模实验与模拟[D].浙江大学,2022.
[12] 张海东. 气-固流化床鼓泡-湍动流型转变特性及流动控制研究[D].天津大学,2020.
[13] 白玲. 流化床稠密气固两相瞬态流动的离散元模拟与高速摄影实验研究[D].江苏大学,2020.
[14] 王维,洪坤,鲁波娜等.流态化模拟:基于介尺度结构的多尺度CFD[J].化工学报,2013,64(01):95-106.
[15] 高希. 鼓泡和湍动流化床两相流体力学新模型研究[D].浙江大学,2013.
[16] 徐骥,葛蔚,任瑛等.Particle-Mesh Ewald(PME)算法的GPU加速[J].计算物理,2010,27(04):548-554.
[17] 陈飞国,葛蔚,李静海.复杂多相流动分子动力学模拟在GPU上的实现[J].中国科学(B辑:化学),2008(12):1120-1128.
[18] 吴锦坤. 鼓泡流化床流动特性的直接数值模拟[D].浙江大学,2006.
[19] Guo, L., J. Wu, and J. Li, Complexity at Mesoscales: A Common Challenge in Developing Artificial Intelligence[J]. Engineering, 2019. 5(5): p. 924-929.