摘要
作为石化工业的关键中间体,环氧乙烷生产过程中催化剂形貌与操作参数的协同优化是提升反应器效能的核心挑战。本研究针对传统实验和模拟方法在催化剂构效关系解析中的高成本瓶颈,融合颗粒解析计算流体力学(PRCFD)与图卷积神经网络(GCN),构建了反应器多物理场的快速预测策略。基于COMSOL平台构建高保真计算流体力学(CFD)模型,研究了圆柱体、单孔及五孔结构催化剂在随机堆积体系中的流动-反应耦合过程,构建了涵盖三种典型颗粒形貌随机堆积构型及四种进气速率的综合研究场景。通过与真实乙烯转化率数据对比,验证了COMSOL模拟参数设置的有效性。模拟表明催化剂颗粒形状和进气速率对乙烯转化率和床层压降的影响呈现强非线性关系。基于有效的模拟数据,采用图卷积神经网络学习催化剂颗粒几何形状与压力、浓度之间的映射关系。训练后的模型能够快速预测不同催化剂和进气速率下的压力和浓度分布,相关系数R2大于0.9。本研究为化工反应器的智能设计提供了兼具物理可解释性与计算效率的创新技术手段。
As a crucial intermediate in the petrochemical industry,the collaborative optimization of catalyst morphology and operating parameters during the ethylene oxide production process represents the core challenge in enhancing reactor performance.In response to the high cost bottleneck of traditional experimental and simulation methods in analyzing the structure-activity relationship of catalysts,this study integrate d particle-resolved computational fluid dynamics(PRCFD)and graph convolutional neural networks(GCN)to construct an intelligent prediction framework for reactor multi-physics fields.A high-fidelity CFD model was established based on the COMSOL platform to investigate the flow-reaction coupling process of cylindrical,single-hole,and five-hole structure catalysts in a randomly packed system.A comprehensive research scenario covering three typical particle morphologies,random packing configurations,and four inlet gas rates was constructed.C ompared with real ethylene conversion data,the effectiveness of the simulation parameter settings in COMSOL was verified.The simulations reveal ed that the effects of catalyst particle shape and inlet gas rate on the ethylene conversion rate and the bed pressure drop exhibit ed a strong non-linear relationship.Based on the valid simulation data,a graph convolutional neural network was employed to learn the mapping relationships between the geometric shapes of catalyst particles and pressure and concentration.The trained model c ould rapidly predict the pressure and concentration distributions under different catalysts and inlet gas rates,with a correlation coefficient R 2 greater than 0.9.This study provide d a new paradigm which combines physical interpretability and computational efficiency for the intelligent design of chemical reactors.
作者
刘廷廷
孟子程
穆丽静
陈锡忠
刘岑凡
LIU Tingting;MENG Zicheng;MU Lijing;CHEN Xizhong;LIU Cenfan(State Key Laboratory of Synergistic Chem-Bio Synthesis,School of Chemistry and Chemical Engineering,Shanghai Jiao Tong University,Shanghai 200240,China;Key Laboratory of Special Equipment Safety and Energy-S aving,State Administration for Market Regulation,China Special Equipment Inspection and Research Institute,Beijing 100029,China)
出处
《化工进展》
北大核心
2025年第8期4571-4581,共11页
Chemical Industry and Engineering Progress
基金
国家重点研发计划(2023YFC3008701)。
关键词
计算流体力学
填充床
乙烯氧化
神经网络
图卷积架构
computational fluid dynamics
packed bed
ethylene oxidation
neural networks
graph convolution architecture
