Research on location and layout of auto-body sheet metal based on NSGA-Ⅱ and RBF neural network

• Main Authors: Peng WANG, Jiachuan XU, Fan CAO, Di LI
• Format: Article
• Language: zho
• Published: Hebei University of Science and Technology 2019-06-01
• Series: Journal of Hebei University of Science and Technology
• Subjects: vehicle engineering; auto-body sheet metal; locating scheme; NSGA-Ⅱ algorithm; RBF neural network
• Online Access: http://xuebao.hebust.edu.cn/hbkjdx/ch/reader/create_pdf.aspx?file_no=b201903001&flag=1&journal_

In order to solve the problem of low efficiency and easy clamping deformation in the location layout design of auto-body sheet metal,alocation layout design method of auto-body sheet metal based on NSGA-Ⅱ and RBF neural network is proposed.With the minimum deviation transfer path and the highest stability as constraints, the first three locating points are optimized by using NSGA-Ⅱ algorithm.With the support of finite element samples, BP and RBF neural network prediction models are constructed and compared, and the results of RBF neural network with higher prediction accuracy are selected as individual fitness values.The GA and PSO are used to optimize and compare the RBF neural network. The solution value of the PSO with faster convergence speed and higher accuracy is chosen as the optimal solution of the fourth location point.Using the seat-mounted beam as a model to verify the research content.The results show that the maximum clamping deformation under the optimized positioning layout is only 27% of the maximum clamping deformation before optimization.Therefore, RBF neural network can effectively predict clamping deformation of sheet metal.The research results have reference value for further research on auto-body welding fixture design and location layout of fuselagethin-walled parts.