Aerodynamic surrogate-based optimization of the nose shape of a high-speed train for crosswind and passing-by scenarios. Issue 184 (January 2019)

• Record Type: Journal Article
• Authors: Muñoz-Paniagua, J.; García, J.
• Is Part Of: Journal of wind engineering and industrial aerodynamics. Issue 184(2019)
• Page Start: 139
• Page End: 152
• Publication Date: 2019-01
• Subjects: Shape optimization -- High-speed train -- Genetic algorithm -- Metamodel -- Trains passing by -- Crosswind; Wind-pressure -- Periodicals; Buildings -- Aerodynamics -- Periodicals; Pression du vent -- Périodiques; Constructions -- Aérodynamique -- Périodiques; Buildings -- Aerodynamics; Wind-pressure; Periodicals
• DOI: 10.1016/j.jweia.2018.11.014
• Languages: English
• ISSNs: 0167-6105

Abstract

Abstract: In this paper, the aerodynamic optimization of the nose of a high-speed train in two different scenarios involving lateral stability, namely two trains passing by and a train subjected to crosswind, is performed using genetic algorithms (GA). The requirements for this optimization method involve the parameterization of each optimal candidate as a design vector. Here, we consider three design variables to include the most characteristic geometrical factors affecting the pressure pulse generated when the two trains meet each other and the side force coefficient for crosswind. These design variables represent the nose length, the bluntness of the nose and the A-pillar roundness. A large set of three-dimensional, turbulent, unsteady simulations of realistic train models has been completed, and this information is used to fit a metamodel. Besides, these simulations yield insight into the design space nature, concluding that the nose length is the most relevant design variable for the pressure pulse, but the bluntness of the nose has also an impact on the magnitude of the pressure peaks. Concerning the crosswind scenario, a neural network is trained to hasten the convergence of the optimization method used to minimize the side force coefficient. We use an ANOVA test to determine the influence of each design variable in the side force coefficient. To complete this single-objective optimization, a multi-objective optimization is developed, and a Pareto front is obtained. Highlights: Two issues concerning lateral stability of high-speed trains are considered here: two trains passing by and crosswind. Genetic algorithm (GA) is used as the optimization method to minimize the pressure pulse and the side force coefficient. A metamodel is used to improve the performance of GA. Simulations required are used to yield insight into the design space. Nose length is the most relevant variable in passing by scenario. Bluntness has a notable role in the pressure pulse. Single (minimize side force) and multiobjective optimization problem are proposed. A Pareto front results from the latter.