Multiobjective Optimization Design of Double-Row Blades Hydraulic Retarder with Surrogate Model

Abstract

For the design of double-row blades hydraulic retarder involves too many parameters, the solution process of the optimal parameter combination is characterized by the large calculation load, the long calculation time, and the high cost. In this paper, we proposed a multiobjective optimization method to obtain the optimal balanced solution between the braking torque and volume of double-row blades hydraulic retarder. Moreover, we established the surrogate model for objective function with radial basis function (RBF), thus avoiding the time-consuming three-dimensional modeling and fluid simulation. Then, nondominated sorting genetic algorithm-II (NSGA-II) was adopted to obtain the optimal combination solution of design variables. Moreover, the comparison results of computational fluid dynamics (CFD) values of the optimal combination parameters and original design parameters indicated that the multiobjective optimization method based on surrogate model was applicable for the design of double-row blades hydraulic retarder.