An intelligent simulation architecture for hazard and operability analysis of large-scale process plant

Develops a knowledge-based simulation architecture for the hazard and operability (HAZOP) analysis phase of large-scale process plant modelling. This work is motivated by the desire for a tool that allows a HAZOP expert to build and modify simulation models at a simulation-language independent level and without the constant presence of a simulation software expert. HAZOP is a safety analysis technique for identifying possible deviations from expected process-plant behaviour and highlighting the consequential operating difficulties. The analysis process generally takes the form of a qualitative and discrete simulation process that answers various `what-if´ questions. The input to the qualitative simulation is a logically complete set of guidewords, such as `more flow´, `less flow´, `more temperature´ etc. The results of the simulation are used to determine all possible states of plant units thereby helping to find the potential hazards