Profile Design and Multifidelity Optimization of Solid Rocket Motor Nozzle

Abstract

This paper presents a new profile modeling method and multifidelity optimization procedure for the solid rocket motor contoured nozzle design. Two quartic splines are proposed to construct the nozzle divergent section profile, and the coefficients of the splines' functions are calculated by a fortran program. Twodimensional axisymmetric and threedimensional compressible Navier–Stokes equations with ReNormalisation Group (RNG) kخµ turbulent models solve the flow field as lowand highfidelity models, respectively. An optimal Latin hypercube sampling method produces the sampling points, and Kriging functions establish the surrogate model combining with the lowand highfidelity models. Finally, the adaptive simulated annealing algorithm is selected to complete the profile optimization, with the objectives of maximizing the thrust and the total pressure recovery coefficient. The optimization improves the thrust by 4.27%, and enhances the recovery coefficient by 4.63%. The result shows the proposed profile modeling method is feasible and effective to enhance the nozzle performance. The multifidelity optimization strategy is valid for improving the computational efficiency.