Catalytic reforming process modeling method based on data and mechanism mixing

Abstract

The invention discloses a catalytic reforming process modeling method based on data and mechanism mixing. The method includes: establishing a mechanism model; discretizing the mechanism model to obtain a discrete model; performing dynamic simulation on the discrete model by adopting a solver capable of solving a large-scale NLP problem to obtain reactor internal parameters which cannot be obtained by actual measurement; constructing a data model, and adopting a neural network learning mode during training; and finally, enabling the trained data model to take the input measured value of the reactor in the catalytic reforming process and the output of the mechanism model after dynamic simulation as the input, comparing the outputwith the output measured value of the reactor in the catalytic reforming process, if the difference value is greater than a threshold value, optimizing the mechanism model again, otherwise, ending the process to obtain the required catalytic reforming process model combining the mechanism model and the data model is obtained. According to the method, the respective defects of mechanism modeling and data modeling are made up, and the universality and adaptability of the catalytic reforming process model are improved.

Description

technical field [0001] The invention belongs to the technical field of automatic control, and relates to a catalytic reforming process modeling method based on data and mechanism mixing. Background technique [0002] Catalytic reforming is one of the main processing technologies in the oil refining industry, and it is also one of the effective methods to solve the shortage of high-octane gasoline and aromatic hydrocarbon raw materials. Only by establishing a more accurate model can the catalytic reformer operate better, increase the yield and obtain better economic benefits. [0003] The process of reforming reaction is that the reforming feed is mixed with circulating hydrogen, and heat exchange is performed with the reforming reaction product. After the heat exchange, it enters the heating furnace to be heated to a certain temperature, and then enters the first reforming reactor for reforming reaction. , to generate a reforming reaction product that comes out of the first...

AI Technical Summary

Problems solved by technology

These foreign modeling software are built for the actual situation of their own country, and the internal mechanism of the software is strictly confidential and cannot be obtained. It is not necessarily applicable, so it is very necessary to establish a mechanism model that is in line with the actual situation in our country

The advantage of the mechanism model is that the internal parameters can be obtained well, and the parameters are easy to adjust, and the model has strong adaptability, but the difficulty is that the catalytic reforming process is too complex and some parameters cannot be obtained, so the mechanism model wants to establish It is very difficult to be accurate. In addition, a large-scale solver is required in the optimization solution, and the solution speed is slow.

[0005] In recent years, with the development of machine learning, artificial intelligence and other technologies, many experts and scholars at home and abroad have gradually used such technologies to establish data models for the catalytic reforming process in order to overcome the shortcomings of traditional mechanism modeling methods. The data model of can be seen as a black box model. This model does not need to know too much about the internal structure of the mechanism process and does not require a solver. However, if only the data model is used, if the future changes of the data exceed a certain range, such a model The reliability of the

[0006] With the world's energy shortage and increasingly fierce competition in the international market, compared with similar enterprises in developed countries, domestic enterprises generally have problems such as high production costs and poor economic benefits.

Images