From customized multiscale modeling to general mesoscience – The principle of compromise
Jinghai Li · AIP conference proceedings · 2013
This presentation reviews a 3-decade research on meso-scale modeling of various complex systems at the Institute of Process Engineering, Chinese Academy of Sciences (IPE, CAS).Starting with the study of particle clustering phenomenon, the EMMS (energy-minimization multiscale) model was established for gas-solid fluidization systems by analyzing the compromise between dominant mechanisms, which was then verified, and further extended to other multiphase systems.In recent years, the EMMS principle was used to define a multi-scale computation paradigm -the EMMS paradigm, featuring the structural similarity between problem, modeling, software and hardware, which was implemented by constructing a supercomputer with a capacity of 1.0 Petaflops.This paradigm was believed to be promising for realizing the virtual process engineering (VPE).The EMMS principle was also believed to be a universal one for all meso-scale phenomena, that is, could be the principle for the emerging science -Meso-science.This presentation reviews the principles, methodology, computation and application of the EMMS paradigm, and gives perspective for VPE and meso-science.