An adjoint-based optimization method for reducing the axial force of a reactor coolant pump

Abstract

To alleviate the wear of a thrust bearing in a reactor coolant pump (RCP) while ensuring the hydraulic performance of the pump, an adjoint-based optimization method is proposed in this study. This method reduces the axial force of the RCP impeller and synchronously improves the impeller’s hydraulic efficiency. By combining the adjoint solution with the radial basis function (RBF)-based mesh deformation, the optimization proceeds along the gradient direction, which greatly reduces the time and cost of the calculation. In the adjoint method, the adjoint equations in the rotating coordinate system are established, a joint objective function of the head constraint, hydraulic efficiency, and axial force is expressed, and then the blade surface sensitivity to the joint objective function is determined. In the RBF mesh deformation, the control points on the blade strand are evenly spaced, which ensures the smoothness of the deformed 3D twisted blade. Using the proposed optimization method, the hydraulic axial force of the impeller is reduced by approximately 3.8%, while the hydraulic efficiency of a scaled RCP impeller is increased by approximately 3.2%, and the head remains at an almost constant value. The obtained results validate the feasibility of the adjoint method for optimizing the design of centrifugal pumps.