No abstract

No abstract

No abstract

In a nuclear power plant (NPP), most of the systems are linked due to processes of fluid flow, heat transfer etc., and their natural tendency to respond to changes during accident conditions. These relationships can be utilized to develop smart applications for plant accident monitoring and management. In this research, the statistical relationships among the process parameters have been analyzed. It has been embarked that the characteristics of a safety system during a particular interval can be estimated by utilizing the other affected parameters, employing statistical correlation and regression models developed from the simulation data offline, when evaluated for the same set of conditions on accident sequence and safety systems. The proposed methodology has been demonstrated for a specific loss of coolant accident scenario using correlation coefficient and neural networks, for the time interval when containment spray system was initiated at the particular stage of accident progression and remained operational for some designed time. Virtual sensor networks were constructed for the estimation of reactor vessel level during that time period, which demonstrates the realization of methodology. The estimations from such virtual sensor networks are expected to improve by utilizing the importance measures and concepts to generalize the neural networks. Also, correlation voting index (CVI) provides a capability to select a set of related outputs, which would be used as a yardstick for comparing results in case, missing or uncertain inputs are present.

During a severe accident, contact of the molten corium with the coolant water may cause an energetic steam explosion which is a rapid increase of explosive vaporization by transfer to the water of a significant part of the energy in the corium melt. This steam explosion has been considered as an adverse effect when the water is used to cool the molten corium and could threaten reactor vessel, reactor cavity, containment integrity. In this study, TROI TS-2 and TS-3 experiments as part of the OECD/SERENA-2 project were analyzed with TEXAS-V. Input parameters were based on actual TROI experiment data. In mixing simulations, calculated results were compared to melt front behavior, void fraction in trigger time and other parameters in experiment results. In explosion simulations, corresponding to TROI experiments an external triggering was employed at the moment that melt front reached heights of 0.4 m. Calculated results of peak pressure and impulse at the bottom were compared with TROI experiment results. Melt front behaviors of the melt was different from the experimental results in both TS-2 and TS-3. Void fraction in triggering time in TS-2 was in good agreement with the experiment results and in TS-3 was slightly overestimated. The peak pressure and impulse at bottom were successfully predicted by TEXAS-V. These calculations will allow establishing whether the limitations and differences observed in the simulations of the experiments are important for the reactor case.