Hazard identification on fractionation column using rule based expert system

Abstract

A rule-based expert system has been developed in which the knowledge used to assist Hazard and Operability Study is divided into process-specific and process-general components. The process-specific knowledge, which consists of process operation data and the process flow sheet, has been constructed using a HYSIS process simulator. The process-general knowledge which consists of rule-based expert system is developed from simulation result obtained from HYSIS process simulator. The link between Hazard and Operability Study result documentation with data base of process operation is important, in order to analyse the causes and consequences of the whole process deviation on fractionation column. In this Hazard and Operability Study, process deviation studied are flow rate, temperature and pressure. An inference engine mechanism is developed using Visual Basic programming language, for appropriate interaction between these components in order to identify causes and consequences of deviation for each process stream. The base case study used is a fractionation column of an oleochemical plant. The study has contributed to an improvement of hazard identification technique and proposed an algorithm by considering the whole consequences of process deviation in each stream. The algorithm is found applicable for any of process system, however the expert system developed is only limited to the application for fractionation column of oleo chemical plant.