METHOD, SYSTEM AND COMPUTER PROGRAM PRODUCT FOR THE ONLINE MONITORING OF AN OIL REFINERY

摘要

A computer-implemented method for monitoring an oil refinery plant (100), comprising: from a dataset of production information of the oil refinery plant (100) comprising production scheduling information (602a), laboratory analysis information (602b) and production control operational information (602c), wherein the laboratory analysis information (602b) includes the penetration value (305) of the residue (303) of the vacuum distillation unit (300) and at least one crude characterization parameter, and wherein the production control operational information (602c) includes at least one temperature parameter and at least one flow rate parameter of a vacuum distillation unit (300): constructing (1060, 606) a predictive model (607) by applying a computerized, numerical algorithm based on machine learning techniques; and calculating model evaluation metrics; obtaining a prediction value (1070) for the penetration value (305) of the residue (303) of the vacuum distillation unit (300) by applying the constructed predictive model (607) to a dataset (605b) comprising a most recent data sample of information of the oil refinery plant (100); depending on the obtained prediction value (1070), actuating or not on at least one production control operational variable of the oil refinery plant (100). A system for monitoring an oil refinery plant (100). A computer program product comprising computer program instructions/code for performing the steps of the method. A computer-readable memory/medium that stores program instructions/code for performing the steps of the method.