Fluidized Catalytic Cracking (FCC) and Chemical Looping Combustion (CLC): Kinetics , First Principle Models (FPM), Data Driven Models (DDM) and Artificial Intelligence (AI)

这项研究将与加拿大帝国石油公司合作开展，目的是改进FCC（流化催化裂化）和CLC（化学落屑燃烧）过程。建议的方法包括：a)在CREC立管模拟器mark II中，使用帝国石油公司提供的催化剂和VGO（真空气油）建立催化裂化动力学，b)使用Barracuda cpfd（计算粒子流体动力学）确定FCC大型装置的FPMs（第一原理模型），c)使用FMP项目数据库和帝国石油公司的操作数据建立涉及可调参数和模型训练器的DDMs（数据驱动模型）。d)利用Matlab/Python软件开发一种新的crecc - hybrid AI（人工智能）算法。该项目解决了当今帝国炼油厂以及北美其他炼油厂面临的挑战。这些问题是VGO原料成本、汽油价格以及二氧化碳和其他温室气体排放增加的快速变化的结果。因此，CREC立管模拟器与FCC和CLC的人工智能培训模型一起，将提供一个独特的平台，以确定提高FCC过程和全球VGO炼油厂操作性能的最佳方法。

This research will be developed in partnership with Imperial Oil, Canada with the aim of improving FCC (Fluidized Catalytic Cracking) and CLC (Chemical Lopping Combustion) processes. The proposed approach involves: a) The establishment of catalytic cracking kinetics using catalysts and VGO (vacuum gas oil) provided by Imperial Oil, in the CREC Riser Simulator-Mark II, b) The determination of FPMs (First Principle Models) for FCC large scale units using Barracuda CPFDs (Computational Particle Fluid Dynamics), c) The establishment of DDMs (Data Driven Models) involving adjustable parameters and model trainers, using the FMP project data library and Imperial Oil operational data, d) The development of a new CREC-Hybrid AI (Artificial Intelligence) algorithm using the Matlab/Python software. This project addresses challenges faced nowadays by Imperial Oil refineries as well as other refineries in North America. These issues are the result of rapid changes in the cost of VGO feedstocks, gasoline value and increased CO2 and other greenhouse gas emissions. Thus, the CREC Riser Simulator together with AI training models for FCC and CLC, will provide a unique platform to determine the best approaches to enhance the performance of FCC processes and VGO refinery operations worldwide.