In the nuclear power industry, much effort is spent analyzing potential plant accidents. While the likelihood that any of these events would ever happen is extremely small, these analyses are a critical component of ongoing research, since they help engineers develop strategies that could potentially divert a major crisis. One way that engineers around the world participate in accident analysis is through international standard problems (ISPs). An organizing body defines the standard problems, and researchers from various countries independently work to solve them.

In one recent study (ISP #43), sponsored by the Committee on the Safety of Nuclear Installations, within the Organization of Economic Cooperation and Development, the problem posed was to predict boron mixing in the downcomer of a pressurized water reactor using CFD. Borated water, or water containing soluble boric acid, is used for reactivity control because boric acid is an excellent absorber of neutrons. ISP #43 considered the case where a temporary loss of coolant in the reactor vessel would cause boiling in the vessel. The steam resulting from the boiling would leave the boric acid in the reactor vessel, and condense in the form of pure, or unborated water in the heat exchanger used for steam generation. After some time, a significant slug of unborated water would form in the reactor loop of the heat exchanger. Reinitiation of the pumps would cause this slug of unborated water to travel back toward the reactor core causing a reactivity excursion if it does not mix sufficiently with the existing borated water in the reactor vessel.

The slug enters the vessel through a cold leg, or inlet pipe, and then travels through the downcomer, an annular space outside of the reactor core. After the downcomer, the slug moves into a lower plenum where it turns upward to enter the reactor core. The task set forth in ISP #43 was to predict the mixing in the downcomer and lower plenum as the slug moves toward the reactor core. The experiments (and CFD simulations) used cold water for the slug, and hot water for the borated water in the vessel, with thermal mixing representative of boron mixing in the actual device.

Contours of temperature on a vertical slice in the downcomer show thermal mixing prior to entry into the reactor

Christopher Boyd from the U.S. Nuclear Regulatory Commission simulated this accident scenario using FLUENT, and submitted his results to ISP #43. The final report of this ISP is not yet available. He subsequently presented his results at the 8th International Conference on Nuclear Engineering in April. His transient 3D model focused on thermal mixing in the downcomer region. Being a blind study, the researchers were not allowed to see the data before completing their analyses. Once the data were revealed, the FLUENT predictions were found to be in excellent agreement. The average temperature near the exit of the downcomer, for example, was well within the margin of experimental error. Temperatures were recorded as a function of time as increasing amounts of the pure cold water entered the unit, lowering the average temperature prior to entry into the core. The spreading of the slug near a step change in the downcomer width also closely matched reported observations. Because the water was introduced through one of four available ports, the azimuthal distribution of temperature at different locations was also of interest. In this portion of the analysis, FLUENT results were also found to closely match the trends in the data.

Average temperature at the downcomer exit from the ISP and CFD predictions

(click image for enlarged view)

Overall, the simulation demonstrated that CFD can be used as an integral part of accident analysis in the nuclear industry. Further analysis with more realistic scenarios is now underway.

FluentNEWS