Hydrofining catalyst performance prediction method

Abstract

The invention discloses a hydrofining catalyst performance prediction method which comprises the following steps: obtaining basic data during crude oil hydrogenation reaction, and normalizing the datato form a database; Establishing a generalized regression neural network model, taking the data influencing the properties of the hydrofining catalyst in the obtained database as input data, and taking the data related to the properties of the product obtained after the hydrogenation reaction as output data; Training the generalized regression neural network model; And predicting the performanceof the hydrofining catalyst. According to the method, the performance of the catalyst can be quickly and effectively predicted, unnecessary repeated labor and energy consumption are reduced, the modelis high in learning speed, high in generalization ability and high in prediction precision, and the performance of the hydrofining catalyst during crude oil hydrogenation catalytic reaction can be quickly and accurately predicted.

Description

technical field [0001] The invention relates to the field of catalyst performance prediction. More specifically, it relates to a method for predicting the performance of a hydrorefining catalyst. Background technique [0002] At present, with the gradual decline in the quality of crude oil, people's demand for pure oil products continues to increase. In the process of petroleum refining, more and more attention is paid to hydrogenation technology, and the most important part of hydrogenation technology is the catalyst. use. [0003] At present, the research method of catalyst research and development through traditional experimental means not only has a certain degree of blindness, but also requires a large number of experiments to try and screen, which wastes manpower and material resources, and because the process is mostly continuous equipment, it usually requires a long operation period for research. It is difficult to respond to the rapidly changing market and provide...

AI Technical Summary

Problems solved by technology

[0003] At present, the research method of catalyst research and development through traditional experimental means not only has a certain degree of blindness, but also requires a large number of experiments to try and screen, which wastes manpower and material resources, and because the process is mostly a continuous device, it usually requires a long operation period for research. It is difficult to respond to the rapidly changing market and provide timely feedback to provide solutions, which affects the economic benefits of the enterprise

Images