Shape optimization of fuel rods in nuclear reactor pressure vessel via Smoothed particle hydrodynamics (SPH) optimization framework

In this paper, a fuel rod shape optimization framework is proposed to enhance coolant flow and heat transfer in the nuclear reactor based on Non-uniform rational B-splines (NURBS) and Smoothed particle hydrodynamics(SPH). By using NURBS, a typical iso-geometric method, various shape features can be described parametrically, including classic cylindrical and helical cruciform fuel rods(HCF). This is the first attempt of NURBS in the field of fuel rod shape optimization. NURBS involves fewer design parameters, reducing the burden of optimization. Furthermore, a SPH solver is introduced to compute objective functions including pressure drop and temperature. The meshless of SPH avoids computational instability caused by frequent meshing during fuel rod deformation. The SPH solver demonstrates its effectiveness in the numerical simulation of the thermal-fluid coupling problem in 2 × 2 fuel rods. In addition, the surrogate-assisted evolutionary algorithm (SAEA) is employed to optimize the design variables including twist angles and control points. The shape optimization framework, centered on NURBS (modeling), SPH (solution), and SAEA (optimization), successfully fills the gap in the evolutionary process from cylindrical fuel rods to helical cruciform fuel rods. Compared to the classical cylindrical fuel rod with mixing vanes, the average and maximum temperatures of the 3 × 3 optimized fuel rods are reduced by 25.6 % and 20.6 %, respectively. And the pressure drop is reduced by 1.37 %.