A model-based framework for automating HAZOP analysis of continuous process plants
Abstract
Hazard and operability (HAZOP) analysis is the study of systematically identifying every conceivable deviation from the design intent, and all possible abnormal causes, and adverse hazardous consequences that can occur in a chemical plant. This is difficult, labor- and knowledge-intensive, and time-consuming analysis. HAZOP analysis is typically performed by a group of experts poring over the process P&IDs for weeks. An intelligent system for automating this analysis can reduce the time and effort involved, make the analysis more thorough and detailed, and minimize or eliminate possible human errors. Towards that goal, a model-based framework has been developed in which the knowledge required to perform HAZOP analysis is divided into process-specific and process-general components. The process-specific knowledge consists of process materials properties and the process P&ID. The process-general knowledge consists of the HAZOP-Digraph models of the process units which are qualitative causal models developed specifically for hazard identification. These models are developed in a context independent manner so that they can be used in a wide variety of process P&IDs. An inference mechanism is developed for appropriate interaction between these components in order to identify process-specific abnormal causes and adverse consequences. Based on this framework, an expert system called HAZOPExpert has been developed in an object-oriented architecture using the expert system shell G2. The performance of HAZOPExpert was successfully evaluated on various industrial scale petrochemical plant case studies.
Degree
Ph.D.
Advisors
Venkatasubramanian, Purdue University.
Subject Area
Chemical engineering|Industrial engineering
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.