Catalytic cracking device optimization method based on data model

Abstract

The invention relates to a catalytic cracking device optimization method based on a data model, and provides a reaction device model established according to big production history data. The method does not depend on a complicated process mechanism, and the yield and key properties of a product can be accurately predicted. The variable correlation algorithm provided by the invention can intelligently screen out variables which are strongly correlated with target variables from massive DCS bit numbers and Lims variables, so that the model complexity is reduced to the minimum, and meanwhile thereliability is guaranteed. In addition, the multi-neural network integrated learning prediction model constructed by the invention is high in operation speed, high in convergence and wide in adaptability, and the proposed optimization method can quickly calculate a result on the premise of consuming few computing resources and can meet the real-time online optimization requirement of the device. An intelligent algorithm is adopted to determine the time delay effect of different process parameters, so that the calculation is more accurate.

Description

technical field [0001] The invention relates to the field of petrochemical production process control and optimization, in particular to a method for optimizing a catalytic cracking unit based on a data model. Background technique [0002] Catalytic cracking is an important means of lightening heavy oil, the core device for gasoline and diesel production in refineries, and the main source of benefits for refineries. In recent years, crude oil has become heavier and inferior, the market demand for clean fuels and low-carbon olefins is increasing, the safety and environmental protection indicators are becoming more and more stringent, and the international competition for petrochemical products is becoming increasingly fierce. presented new challenges. Catalytic cracking is the most complex catalytic production device in the petroleum refining industry. It is very complicated to establish a process mechanism model for the entire device. Big data technology can directly mine l...

AI Technical Summary

Problems solved by technology

In other studies, the eight-lumped kinetic model combined with the BP neural network to predict the product yield of catalytic cracking has improved the prediction accuracy compared with the simple lumped model. However, there are only 120 groups of production sample data for training, and the time covered Smaller range, lack of applicability testing under wider working conditions

There are also some studies that apply artificial neural network technology to device control, but there are also problems of limited training samples and narrow application range.

[0004] Therefore, the shortcomings of existing methods are summarized, including the following aspects: (1) The mechanism model is complex, difficult to establish, and takes a long time

(2) The complex mechanism model has a high degree of nonlinearity, slow calculation speed, and poor convergence

(4) The current method does not consider the time delay effect, and the internal causal relationship between the data does not correspond, resulting in poor extrapolation performance of the model

(5) The selection of variables depends on the experience of artificial experts, and some influencing variables may be ignored, or invalid variables may be added, increasing the complexity of the model

Images