| 基于改进的BP神经网络胶铆接头的力学性能预测研究 |
| |
| 引用本文： | 刘文杰,邢彦锋,陆瑶.基于改进的BP神经网络胶铆接头的力学性能预测研究[J].农业装备与车辆工程,2021(3). |
| |
| 作者姓名： | 刘文杰 邢彦锋 陆瑶 |
| |
| 作者单位： | 上海工程技术大学机械与汽车工程学院 |
| |
| 基金项目： | 国家自然科学基金资助项目“铝钢零件混联薄板结构焊装偏差分析建模与工艺控制研究”(51575335);上海科学技术基金项目“电致塑性自冲铆接车身装配质量控制及应用”(16030501300)。 |
| |
| 摘 要： | 利用遗传算法优化BP神经网络的连接权值和阈值,并将改进的BP神经网络应用于胶铆接头力学性能预测中,建立了胶铆接头最大拉剪力预测模型。结果表明:GA-BP神经网络比BP神经网络的收敛时间长,但GA-BP网络预测相关系数更好,回归性能更好,具有更好的泛化能力。对训练好的神经网络预测模型进行验证,发现GA-BP神经网络预测的均值绝对误差为BP神经网络均值绝对误差的40%,GA-BP神经网络具备更好的预测性能。
|
| 关 键 词： | AA6111铝合金 胶铆接头 遗传算法 BP神经网络 |
Prediction Model of Mechanical Properties of Riv-bonding Joints Based on Improved BP Neural Network |
| |
| Liu Wenjie,Xing Yanfeng,Lu Yao.Prediction Model of Mechanical Properties of Riv-bonding Joints Based on Improved BP Neural Network[J].Agricultural Equipment & Vehicle Engineering,2021(3). |
| |
| Authors: | Liu Wenjie Xing Yanfeng Lu Yao |
| |
| Institution: | (School of Mechanical and Automotive Engineering,Shanghai University of Engineering Science,Shanghai 201620,China) |
| |
| Abstract: | The genetic algorithm(GA)is used to optimize the connection weight and threshold of BP neural network(BPNN),and the improved BPNN is used to predict the mechanical properties of the rubber riveting head.The results show that GA-BPNN has a longer convergence time than BPNN,while GA-BPNN predicts better correlation coefficients,better regression performance,and better generalization ability.Verification data were substituted into the trained neural network prediction model for verification.The mean absolute error predicted by GA-BPNN is 40%of the BPNN,and GA-BPNN has better prediction performance. |
| |
| Keywords: | AA6111 aluminum alloy Riv-bonding joint genetic algorithm BP neural network |
| 本文献已被 CNKI 维普 等数据库收录！ |
