Surrogate based internal and external ballistic optimization design for solid launch vehicles

This research aims to improve the efficiency of the internal and external ballistics integrated optimization design for solid launch vehicles (SLVs). The solid rocket motor model and trajectory simulation are established and respectively optimized in series and in parallel. The complete set of conventional serial design process and integrated parallel design process are respectively proposed and compared. In order to effectively improve the efficiency of the optimization design for internal and external ballistics of solid launch vehicles, the sequence approximate optimization method based on the radial basis function surrogate model is proposed to reduce the calling times of the original models with high computational cost. An internal and external ballistic integrated optimization design framework is established by 3 stages: Design of Experiment stage, surrogate modeling stage and updating stage. Besides, a multi-stage adaptive sampling method is applied for surrogate model update. Finally, in comparison with the conventional serial design process, the result shows that the proposed integrated parallel design process effectively reduced the initial mass of the objective 3-stage SLV and greatly reduced the time requirement and computational cost. The research results can provide theoretical and technical support for the integrated design of SLVs and provide reference for the optimization of overall parameters.