Prediction method for refining property of pure component based on hierarchical group construction

Abstract

The invention discloses a method for predicting refining properties of pure components based on hierarchical group construction, which is used for predicting octane number and cetane number of pure component compounds in petroleum products, and avoids redundancy of feature sets through hierarchical construction of component feature sets and introduction of hierarchical group construction. When a third-level component descriptor is added into a feature set, a Bayesian rule is introduced, so that posterior probability distribution estimation can be carried out on the feature set, and a feature set with a higher posterior probability is selected without only paying attention to prediction precision. On the basis, the Bayesian rule is introduced again, so that posterior probability distribution estimation can be carried out on the final model, and the over-fitting risk of the final prediction model is avoided. The method can be applied to a crude oil and product blending unit in the petrochemical industry, and the petroleum refining precision is effectively improved.

Description

technical field [0001] The invention relates to the field of refining and chemical industry data analysis, in particular to a method for predicting refining and chemical properties of pure components based on hierarchical group construction. Background technique [0002] Due to the limitations of analytical chemistry and computer hardware, traditional refining unit models mostly use lumped kinetic models. Raw materials and products are often divided into lumps based on macroscopic properties such as boiling point or solubility. Ten lumped and eleven lumped models are widely used in catalytic cracking units. However, the lumped models based on macro-level division naturally have multi-component properties and cannot characterize the component information in detail, making it difficult to extend such lumped models to new feedstocks and catalyst systems. However, the lumped model at the molecular level can calculate the composition and properties of the raw materials from the ...

AI Technical Summary

Problems solved by technology

However, the lumping based on macro-level division naturally has multi-component properties, and cannot characterize the component information in detail, making it difficult to extend this kind of lumped model to new raw materials and catalyst systems

Images