Adjoint Optimization Method for Head Shape of High- Speed Maglev Train

Abstract

The head shape of a high-speed maglev train was optimized in this study, based on the adjoint method, and the aerodynamic drag of four optimized train models were simulated and compared using different control point generation methods. The effectiveness of using the adjoint method to develop a compressible model for a maglev train was verified. The results show that the adjoint matrix optimization method can quickly and effectively capture the shape characteristics of the train head that are sensitive to aerodynamic resistance. When the design variables of the head are not defined separately, the grid control point set and surface control point set can be used to carry out the adjoint closed-loop optimization of the train head shape, and the exchange control point generation method can be used to perform closed-loop optimization. The results of a numerical simulation show that the optimized train model reduces aerodynamic resistance by approximately 4.8%.