Presentation Information
[1E11]Prediction and optimization technology for producing petrochemical fraction by co-hydroprocessing of petroleum and low-carbon feedstock oils
○Takahiro Himukai1, Anzai iwao1, Mayumi Yokoi1, Kiyoshi Sase2 (1. ENEOS Corp., Central Technical Research Laboratory, 2. Japan Petroleum and Carbon Neutral Fuels Energy Center)
Keywords:
co-hydroprocessing,neural-network,hydrocracking
Technology is being developed to co-process low-carbon feedstock oil, which consists of environmentally friendly resources such as waste plastics and biomass, in refineries by blending them with petroleum. In this study, co-hydrocessing with reduced-pressure diesel oil was investigated with the aim of producing petrochemicals, for which demand is expected to remain strong in the future. As a result of the reaction analysis, it became necessary to predict and optimize a complex reaction system, so a reaction simulator combining an ordinary physical model and a neural network was created and is presented here.