抄録
This paper presents a novel design technique for high-speed trains using a multi-objective optimization method to balance plural aerodynamic properties. The technique involves the use of an evolutionary algorithm, a shape parameterization technique that uses B-spline curves and Coons patches, and a computational simulation that uses a Message Passing Interface. To demonstrate the capability of the method, a train nose is designed that has optimized aerodynamic drag and aerodynamic forces with respect to affecting other trains. A 10th generation evolutionary calculation with 512 individuals was conducted, and physically reasonable Pareto solutions were successfully obtained.
