Open AccessPrediction of Maximum Deformation of Single Nail Riveting Based on RBF Neural Network
Author(s) -
Jinyu Gu,
Jian Yin,
Yongdang Chen,
Chengcheng Zhao,
Yunfei Cheng
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1576/1/012024
Subject(s) - rivet , artificial neural network , deformation (meteorology) , finite element method , parametric statistics , approximation error , computer science , python (programming language) , structural engineering , engineering , artificial intelligence , algorithm , materials science , mathematics , statistics , composite material , operating system
The influencing factors of riveting deformation are more complicated, and the specific relationship between the amount of deformation and each factor is difficult to express with general expressions, which is a non-linear problem. Aiming at this problem, this study uses RBF neural network to establish a model of the relationship between the maximum deformation of single nail riveting and various factors. Then, 1000 sample sizes were designed using the LHS method, with 90% of the sample size as training and 10% as testing. Secondly, the secondary development of the finite element software is carried out by using Python language to realize parametric modeling and batch processing. Finally, using the RBF neural network model to predict the maximum deformation of a single nail riveting, the maximum relative error and the average relative error were 8.43% and 2.948%, respectively. The results show that the RBF neural network can be applied in the field of prediction of maximum deformation of riveting and has high prediction accuracy.