Single-objective aerodynamic optimization of a streamlined bridge deck subjected to shape modification using a polynomial emulator and genetic algorithm
Sévérin Tinmitondé,Xuhui He,Lei Yan
DOI: https://doi.org/10.1007/s00158-022-03459-8
IF: 4.279
2022-12-01
Structural and Multidisciplinary Optimization
Abstract:Traditional approach based on wind tunnel tests used in bridge wind-resistant design is no more efficient since long-span bridges require an iterative process involving multiple design variables and processes. To ease the design process, scholars have proposed aerodynamic shape optimization based on the Kriging surrogate model. Despite the prowess of the Kriging model, it is accurate only when using high-fidelity data (experimental or Large Eddy Simulation data) and is not suitable for noisy datasets, making this model costly and hindering its practical use. To tackle this issue, the present study proposed a polynomial surrogate combined with a genetic algorithm to determine the optimal shape of a streamlined bridge deck cross-section. First, a uniform sampling plan considering 57 different geometries was generated. Then, the polynomial surrogate was used to predict the aerodynamic coefficients and the flutter velocity based on the dataset obtained from the unsteady Reynolds-average Navier–Stokes computational fluid dynamics simulations. The accuracy of the surrogate model is evaluated using error metrics, such as the sum-squared error, R -squared, and the root mean squared error. The results obtained from the statistical metrics demonstrate that the proposed polynomial surrogate provides an accurate prediction of the force coefficients and their derivatives, as well as the critical flutter velocity. A benchmark example is used to demonstrate the efficiency of the proposed model. Finally, a genetic algorithm was introduced to determine an optimal shape, and the results demonstrate that the proposed optimization framework can improve the design process remarkably.
mechanics,computer science, interdisciplinary applications,engineering, multidisciplinary