Study on a leakage rate predictive model with Application in Multi-Conditions conversion for Double-Offset butterfly valves

Abstract

A significant number of valves constitute a vital component of the containment pressure boundary of a nuclear power plant, and their leakage rates under accident conditions must be tightly controlled. In this study, a methodology for the construction of a predictive model of valve leakage rate is put forth the study of valve leakage characteristics under accident conditions. Taking the double-offset butterfly valves (DOBV) as the research object, its multi-parameter leakage rate predictive model is developed through finite element analysis (FEA) of the microscopic and macroscopic contact mechanics of the sealing, that are then combined with an existing interfacial leakage model. This model is able to predict the effects of different parameters, including the temperature, pressure, and humidity. It is then verified by comparing it with the experimental values of the multi-operating conditions leakage test. Based on this model, the impact of operating condition parameters on the leakage rate of the valve is numerically studied and a dimensionless leakage rate relationship is proposed: $Q_{\text{r}}=E_{\text{P}}{E}_{\text{T}}{E}_{\text{m}}$, where $E_{\text{P}}$,$E_{\text{T}}$,and $E_{\text{m}}$ are pressure, temperature, and humidity conversion factors, respectively. These factors are in a power function relationship with their respective dimensionless state parameters. Further analysis demonstrates that the powers of these relationships are independent of roughness and weakly correlated with seal material and seal structure. The leakage pattern of this valve under two hypothetical accident scenarios is comparatively analyzed by applying the conversion equation, and some useful conclusions related to dynamic leakage rate control are drawn. The methodology presented in this paper is central to the contact seal leakage mechanism and can be extended to other containment penetrations.