Abstract: The building of a heavy oil database, which containing massive evaluation data of heavy oil catalytic cracking obtained from the State Key Laboratory of Heavy Oil Processing, China University of Petroleum (Beijing), is introduced. On this basis, by using part of the data of heavy oil properties and product distributions as training data and adopting chemometrics correction theory, mathematical correlation models of heavy oil physicochemical properties and product distributions in catalytic cracking process are set up. Validation of the models is carried out using another part of the data from the database. Results show that the product distributions of heavy oil catalytic cracking can be predicted with mathematical correlation models built by correction theory fast and accurately.

Key words: heavy oil, chemometrics, correction theory, catalytic cracking, model