Most fuel modeling computer codes of a decade ago did not explicitly account for fuel thermal conductivity degradation with burnup. The overestimation of thermal conductivity is compensated by other code adjustments. Following the approval of the legacy fuel thermal codes, the quantity, quality and comprehensiveness of fuel measurement data have increased dramatically. Recently, the US NRC has informally requested the nuclear industry review the thermal performance codes with the consideration of thermal conductivity degradation. This paper compares the legacy thermal conductivity models with the most recent available models that are published publicly and may have been incorporated into present generation fuel performance codes. Also included in this paper is one method which AREVA NP Inc. developed to evaluate the thermal conductivity degradation effect. The effect of thermal conductivity degradation on the PWR licensing limits, such as centerline fuel melt and strain, is demonstrated.

No abstract

It has been established by other authors [1] that the accelerations of the water confined by the reactor core baffle plates has a significant effect on the responses of all the fuel assemblies during LOCA or seismic transients. This particular effect is a consequence of the water being essentially incompressible, and thus experiencing the same horizontal accelerations as the imposed baffle plate motions. These horizontal accelerations of the fluid induce lateral pressure gradients that cause horizontal buoyancy forces on any submerged structures. These forces are in the same direction as the baffle accelerations and, for certain frequencies at least, tend to reduce the relative displacements between the fuel and baffle plates. But there is another confinement effect — the imposed baffle plate velocities must also be transmitted to the water. If the fuel assembly grid strips are treated as simple hydro-foils, these horizontal velocity components change the fluid angle of attack on each strip, and thus may induce large horizontal lift forces on each grid in the same direction as the baffle plate velocity. There is a similar horizontal lift due to inclined flow over the rods when axial flow is present. These combined forces appear to always reduce the relative displacements between the fuel and baffle plates for any significant axial flow velocity. Modeling this effect is very simple. It was shown in previous papers [2,3] that the mechanism for the large fuel assembly damping due to axial flow may be the hydrodynamic forces on the grid strips, and that this is very well represented by discrete viscous dampers at each grid elevation. To include the imposed horizontal water velocity effects, on both the grids and rods, these dampers are simply attached to the baffle plate rather than “ground”. The large flow-induced damping really acts in a relative reference frame rather than an absolute or inertial reference frame, and thus it becomes a flow-induced coupling between the fuel and baffles. This has a significant effect on the fuel assembly motions and tends to reduce the relative displacements and impact forces between fuel assemblies and baffle walls.

The objective of this paper is to develop a purely mechanistic fuel assembly structural model that will predict the fuel assembly’s static and dynamic characteristics from the knowledge of the fuel assembly’s geometry and component properties. This model provides a method for analyzing the static and dynamic lateral and axial properties of the fuel assembly. A comparison of various in-air fuel assembly test data such as lateral and axial stiffnesses and lateral natural frequencies is provided to demonstrate the analytical model. The fuel assembly model developed by Shah, Brenneman, etc. (1), achieved very good agreement with assembly lateral impact test data by utilizing a “3-beam” model. In that model, the fuel rod-to-spacer grid interfaces were represented by spring and friction elements. The fuel assembly was restrained at each grid position by means of rotational springs, which were benchmarked to the test frequencies. This newly developed model eliminates the need for using rotational springs at the grid locations. Hence, it fully simulates the fuel assembly lateral and axial behavior based on the fuel assembly geometric properties. The fuel assembly model is a 2-D planar model of beams in both lateral and axial directions. The grids are modeled with plate elements. At each grid location there are springs, preload, and frictional sliders representing the lateral and axial connectivity characteristics to the fuel assembly beam model. As the Zircaloy grid preloads relax from irradiation, they can be easily simulated by removing the preload. Hence, this model can represent the fuel assembly structural properties for all aspects of fuel assembly cycles. This model can be used to analyze the fuel assembly lateral static stiffness, first mode and higher order lateral natural frequencies, mode shapes, axial stiffness, in-grid stiffness, through-grid stiffness, and fuel assembly lateral and axial seismic and LOCA response. The model will also estimate the fuel rod frequencies and mode shapes. This model may eliminate the need for some expensive prototype fuel assembly testing.

No abstract