Executive Summary : | A wide spectrum of nuclear reactors are in operation across the length and breadth of India. Safe and reliable operation of these reactors and their longevity is highly dependent on safety standards and reliable design practices. Central to all the reactors (LWR, BWR, PWR, PHWR, FBR, CANDU etc), in operation is the extraction of heat energy from the bundle of fuel rods, which is of paramount importance to design and development. How to optimally extract heat from the rod bundles with reasonable critical heat flux and critical power ratio margins is the central dogma of this field. Depending on the nature of design specifics and availability of the nuclear reserves available in the country, the bundle of nuclear fuel rods use various fuel sources. Typical coolants used are heavy water, pressurized heavy water, liquid sodium etc. More over, the rod bundle design is also influenced by a number of what if scenarios, that are envisioned and suitable safe guards are implemented. The design of these rod bundle configurations are based on scale down experiments that build correlations for sub-channel analysis codes. Since the advent of high throughput computing, full scale 3-D CFD simulations use Reynolds averaged Navier-Stokes calculations. However, the trillions of degrees of freedom involved in performing these simulations makes this approach computationally intensive. Hence, simpler approaches such as, sub-channel analysis codes have become popular due to their reasonable CPU time, as they have reasonable accuracy. Present study proposes to develop high fidelity sub-channel analysis codes. The study will perform limited CFD calculations on rod bundles to upscale the sub-channel analysis procedure to match the CFD based simulations. To this end, a novel thermal resistance net-work based up-scaling strategy is proposed. The study will include CFD simulations on 19 or 217 rod bundles using OPENFOAM solvers on the Aqua Blue HPC cluster of IIT Madras. The results obtained from the CFD calculations will be cast in the frame-work of a thermal resistance network, to enable the fine tuning of the modelling constants used in the sub-channel analysis. |