Experiment on distributed dynamic simulation for safety design of chemical plants

Abstract

To meet the market challenges, chemical plants need to provide safer plant operation. Safety design approach is used to ensure the safety during the design stage, which satisfies the safety during the plant operation. In such approach, simulation practices are widely used to provide quantitative measures to assess the fault propagation and abnormal situations. As most of the chemical plants are becoming more complex, simulation tools need to provide more intelligent and distributed environment to meet the performance requirements. This research work proposes a distributed simulation environment to support safety design activities of complex chemical plants using intelligent agents. The proposed approach is based on dividing the complex plant design model into smaller and controlled sub-processes called control group units (CGUs). The design and simulation activities of each CGU will be carried out independently on a set of computation resources. Intelligent agents are developed to integrate the plant partitions where simulation results are exchanged among the different simulation sessions. The proposed distributed simulation environment is used to assess the fault propagation within each plant segment (i.e. CGU) and between the adjacent CGUs. The proposed solution is used during the design of a case study HDS chemical plant.