Multi-objective optimization of wire-wrapped reactor fuel assemblies using CFD and genetic algorithms

The lead bismuth eutectic cooled fast reactor (LFR), as a novel energy conversion system, offer significant advantages in terms of both economy and safety compared to other sustainable energy sources. However, the optimization of reactor design using traditional control variable methods faces certain limitations when dealing with multiple design variables. Therefore, this study proposes a multi-objective optimization analysis method based on computational fluid dynamics (CFD) and genetic algorithm. This approach aims to find geometric designs of wire that provide the highest heat transfer capability and the lowest flow resistance. This study conducted numerical simulations on a 7-rod fuel assembly of the LFR, analyzing the impact of different wire pitches and diameters on thermal-hydraulic parameters of the coolant. The results indicated that the wire is the primary factor promoting coolant mixing. Variations in wire pitch predominantly affect the heat transfer coefficient. When the pitch is greater than 342.7 mm and the diameter is greater than 4.2 mm, variations in diameter primarily influence the friction factor; below these values, wire pitch is more significant in determining the friction factor. The errors in the multi-objective optimization analysis are within the specified range (10 %), with maximum relative errors of 6.67 % for the heat transfer coefficient and 5.29 % for the friction factor. The maximum heat transfer coefficient of 18,933 W & sdot;m- 2 & sdot;K- 1 is occurred when the wire pitch is 179.2 mm and the diameter is 4.4 mm. The minimum friction factor of 0.161 is observed when the wire pitch is 397.2 mm and the diameter is 3.5 mm.