Probabilistic Assessment of Rupture Discs Using Classical Regression and Monte Carlo sampling method

Rupture discs serve as pressure control devices in the oil, gas, and petrochemical industries, functioning similarly to safety valves. These discs rupture when the applied pressure reaches their design limit, thereby releasing the tank’s fluid, reducing system pressure, and preventing damage to the system. This study focuses on flat rupture discs modeled as circular plates with specific diameters and thicknesses. External pressure is applied to the disc surface, and using the Johnson-Cook structural and failure model for ۳۰۴ steel, the rupture pressure is determined. The problem was simulated using finite element analysis in the Abaqus software on several rupture disc samples. The rupture pressure for a set of samples was calculated, and based on the results, a regression relationship was developed between the problem parameters and the rupture pressure. This regression model was derived using classical regression techniques and statistical analysis software, offering a predictive tool for determining rupture pressure in flat discs. The proposed model demonstrates high accuracy compared to the simulation results and can predict rupture pressure under various geometric conditions. This predictive capability enhances the design and functionality of rupture discs, reducing potential costs associated with the failure of pressurized systems. The findings provide a robust framework for improving rupture disc performance and ensuring the safety of critical industrial applications.

Keywords:

Rupture disc , finite element analysis , classical regression , Johnson-Cook model