Scenario generation and aggregation to enhance resilience assessment in chemical processes using dynamic simulations

Abstract

An integrated resilience assessment framework for chemical process systems (CPSs) is proposed to address limitations in disturbance-scenario selection and multi-scenario analysis within existing dynamic simulation-based methods. The framework first employs a signed directed graph (SDG) to generate multi-source disturbance scenarios from safety-critical variables (SCVs), thereby ensuring comprehensive and accurate scenario identification. Next, process dynamic simulation models are established, with time-varying process parameters used to depict how system performance evolves under different disturbances. Finally, by weighting individual resilience indices according to disturbance frequencies and scenario occurrence probabilities, the framework yields a clear quantification of overall system resilience. When applied to a methanol–water distillation process, the proposed method precisely captures the system’s inherent resilience under routine disturbances in dynamic response and scenario adaptability. This work provides a systematic theoretical methodology for resilience assessment of complex systems, laying a solid foundation for subsequent decision-making and design optimization considering process resilience. It provides effective support for mitigating process safety accidents induced by operational disturbances and reducing system risks.