Article: Optimisation of drag coefficient of a car based on back propagation neural network and genetic algorithm Journal: International Journal of Vehicle Performance (IJVP) 2024 Vol.10 No.4 pp.466 - 484 Abstract: In order to reduce the time cost of automotive aerodynamic performance development, this paper proposes a new method that combines error back propagation neural networks and optimisation algorithms for global optimisation based on the traditional method, with the aim of achieving a reduction in the time cost of optimal design. In this paper, a simplified car model is chosen as the research object, and the aerodynamic characteristics of the vehicle are enhanced by modifying key geometric parameters of the simplified car body. The target is to minimise the aerodynamic drag coefficient. Five pivotal geometric parameters are selected as design variables, and a Latin hypercube sampling method is employed to generate 50 sets of sample data. Subsequently, the samples are modelled using ANSYS Fluent 2021 R1 software in order to calculate the drag coefficients. The resulting data, including the calculated drag coefficients and the geometric parameters, are employed to train a BP neural network. Subsequently, genetic algorithms are applied to identify the optimal design. The findings demonstrate that the optimised vehicle model achieves a 24.52% reduction in drag coefficient. Inderscience Publishers - linking academia, business and industry through research

Title: Optimisation of drag coefficient of a car based on back propagation neural network and genetic algorithm

Authors: Zihou Yuan; Hongwei Zhang; Wangyang Xiang; Yanming Du; Xingren Zheng

Addresses: Hubei Key Laboratory of Digital Textile Equipment, Wuhan Textile University, Wuhan, Hubei, 430200, China ' Hubei Key Laboratory of Digital Textile Equipment, Wuhan Textile University, Wuhan, Hubei, 430200, China ' Hubei Key Laboratory of Digital Textile Equipment, Wuhan Textile University, Wuhan, Hubei, 430200, China ' Hubei Key Laboratory of Digital Textile Equipment, Wuhan Textile University, Wuhan, Hubei, 430200, China ' Hubei Key Laboratory of Digital Textile Equipment, Wuhan Textile University, Wuhan, Hubei, 430200, China

Abstract: In order to reduce the time cost of automotive aerodynamic performance development, this paper proposes a new method that combines error back propagation neural networks and optimisation algorithms for global optimisation based on the traditional method, with the aim of achieving a reduction in the time cost of optimal design. In this paper, a simplified car model is chosen as the research object, and the aerodynamic characteristics of the vehicle are enhanced by modifying key geometric parameters of the simplified car body. The target is to minimise the aerodynamic drag coefficient. Five pivotal geometric parameters are selected as design variables, and a Latin hypercube sampling method is employed to generate 50 sets of sample data. Subsequently, the samples are modelled using ANSYS Fluent 2021 R1 software in order to calculate the drag coefficients. The resulting data, including the calculated drag coefficients and the geometric parameters, are employed to train a BP neural network. Subsequently, genetic algorithms are applied to identify the optimal design. The findings demonstrate that the optimised vehicle model achieves a 24.52% reduction in drag coefficient.

Keywords: automotive aerodynamics; aerodynamic characteristics; computational fluid dynamics; CFD; BP-ANN; GA.

DOI: 10.1504/IJVP.2024.142091

International Journal of Vehicle Performance, 2024 Vol.10 No.4, pp.466 - 484

Received: 18 Apr 2024
Accepted: 10 Jul 2024
Published online: 07 Oct 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article

Title: Optimisation of drag coefficient of a car based on back propagation neural network and genetic algorithm

Keep up-to-date