Design And Scheduling Optimization Of Refinery Hydrogen System

The continuous increase of the proportion of hydrotreating and hydrocracking processes in the current refineries,leads to the increasing consumption of hydrogen.Thus,it is important for the refinery to optimize the hydrogen system to save fresh hydrogen.Many research efforts have been focused on single impurity hydrogen system,and the flowrates of the hydrogen sources and hydrogen sinks are assumed to be constant.In this thesis,a fuzzy optimization mathematical model of multi-component hydrogen system in refinery based on superstructure is proposed,which considers four components,namely H₂,H₂S,C2+and CH₄,as well as the flowrate variation range of hydrogen sources and hydrogen sinks.The objective functions include annual total cost（TAC）minimization and overall system satisfaction（λ）,respectively.The removal of H₂S（i.e.desulfurization process）and recovery of C2+（i.e.light hydrocarbon recovery）are also considered.GAMS software is used for modeling and solve the proposed mathematical model.Simplified industrial case study results show that,considering both the direct utilization of of HP gas（gas stream from high pressure separater）and LP gas（gas stream from low pressure separater）,the optimized design of the recovery of light hydrocarbon can maximize the economic benefit.The fuzzy optimization method can be used to guide the optimal design of hydrogen system with multi-period variable flowrates.For hydrotreating unit（such as diesel hydrotreating）,the calculation model of reaction hydrogen consumption in the literature is used in this thesis.The chemical hydrogen consumption of hydrodesulphurization,hydrodenittrogenation,olefins and aromatics saturation can be calculated,respectively.Then,combining the empirical data of hydrogen emission and dissolved hydrogen consumption,the total hydrogen consumption can be determined.The accuracy of the method is verified by calculating the hydrogen consumption of diesel hydrogenation unit under four conditions.For hydrocracking unit,the hydrogen consumption correlation model of hydrocracking unit in the literature is improved.The data of related parameters are converted to be dimensionless data to fit the parameters of the regression model,and the hydrogen consumption correlation formulation can be obtained.Case studies show that the average absolute relative deviation（AARDf）s of the calculated hydrogen consumption of the hydrotreating and hydrocracking units are less than 2%,compared with the actual hydrogen consumption.It indicates that the model can be used to estimate/predict the hydrogen demand of the hydrogenation unit.In the actual operation of oil refinries,the hydrogen demand of the hydrogenation units varies with the adjustment of the processing plan and the change in the propertities of processed oil.This thesis proposes a multi-period scheduling optimization model for refinery hydrogen system in refinery.The hydrogen consumption prediction method in Chapter 3 is adopted to determine the hydrogen demand for hydrogenation units.The constraints such as the flowrate,hydrogen concentration,pressure and hydrogen storage capacity of the hydrogen header under multi-periods operation,as well as the flow direction problem in scheduling periodsis are included in the mathematical model.The software GAMS is used for modeling and solve the mathematical model of the case study.Results show that the total hydrogen production of the hydrogen plant during the scheduling periods is 262,975 Nm³,and the flow direction of hydrogen header switches in the third period.