Memory, refining process flow segmented simulation and optimization method, device and equipment

Abstract

The invention discloses a memory, a refining process flow segmented simulation method, device and equipment, and the method comprises the steps: obtaining a raw material nuclear magnetic spectrum of a raw material oil product; generating raw material physical property data corresponding to the raw material oil product; the process flow of each device is simulated according to the upstream and downstream sequence of each device in the refining process, and the process flow simulation method comprises the following steps: generating a mass transfer-reaction equation corresponding to the device according to the device type and the process parameter of the device in the refining process; the device simulation result in the nuclear magnetic spectrum form of the raw material or the device simulation result in the nuclear magnetic spectrum form of the previous device serves as a feature vector of oil product properties, the feature vector and a mass transfer-reaction equation serve as input, and the device simulation result in the nuclear magnetic spectrum form of the device is generated through a mass transfer-reaction solver; and performing physical property calculation according to the device simulation result of the device, and analyzing the device simulation result in the nuclear magnetic spectrum form into product physical property data of the device.

Description

technical field [0001] The invention relates to the field of chemical production, in particular to a method, device and equipment for segmental simulation and optimization of storage and refinery process flow. Background technique [0002] In the field of oil refining and chemical technology, the process simulation method is the theoretical basis to realize the application of process design, device optimization selection, production optimization and so on. [0003] Existing process simulation software adopts a unified architecture model, that is, physical property package-PFD modeling-iterative calculation. The physical property package is the basis of process simulation software. Existing software uses physical property equations based on pure components. PR equations are usually used in the calculation of refining processes, and the molar composition is used as the calculation basis. In each calculation process of process simulation, the components of the physical propert...

AI Technical Summary

Problems solved by technology

However, for the refining process, the complex oil hydrocarbon mixture cannot be divided by pure components, and it is also difficult to determine the molecular weight to calculate the molar composition, and the error is large when switching the molar composition

[0004] After research, the inventor found that in the prior art refining and chemical process flow simulation scheme, in order to keep the data of the actual production calibration, detection and online analysis of the oil stream in the industrial device consistent with the physical property package in the process simulation software, it is necessary to The actual detection data is converted into virtual narrow fraction physical property package components, which will introduce the first error; then, the narrow fraction physical property package needs to be fitted to other properties according to the empirical formula of the simulation system, and then the second error is introduced. Secondary error; then, because the empirical reaction kinetic model ignoring the influence of mass transfer factors is used for calculation in the process simulation environment, a third error will be introduced; finally, if the calculated results are to be applied to industrial devices, It is necessary to reconvert the process simulation software and logistics parameters in the system into macroscopic and continuous production analysis data, so the fourth error will also be introduced

[0005] It can be seen from the above that in the prior art refining and chemical process flow simulation scheme, multiple errors are introduced, so the accuracy of the results is poor, which in turn will lead to poor results when used for process flow optimization

Images