基于BP和RBF神经网络的折叠网壳帐篷风压系数预测对比研究

Comparative study on Wind Pressure Coefficient Prediction of Folding Reticulated Shell Tent Based on BP and RBF Neural Networks

为了全面掌握折叠网壳帐篷风荷载分布特性,分别利用BP神经网络和RBF神经网络对折叠帐篷在风速为20m/s时风向角0°、45°、90°三种工况下风压系数进行预测对比分析。与风洞试验结果相比,BP神经网络和RBF神经网络在折叠帐篷风压系数的分布趋势均能够达到较好的吻合。在风压系数的定量分析上,RBF神经网络在时间效率和准确性上均比BP神经网络表现更好。两种神经网络模型在风压系数预测上的优缺点,可为类似工程神经网络模型构建提供参考。

To comprehensively understand wind load distribution characteristics in folding reticulated shell tents,both the BP neural network and the RBF neural network are used to predict and compare the wind pressure coefficient under three working conditions:wind direction angles of 0°,45°,and 90°,with a wind speed of 20m/s.Comparing the results with those from wind tunnel tests,it indicates that the wind pressure coefficient distribution trend predicted by both the BP neural network and the RBF neural network aligns well.In quantitative analysis,the RBF neural network outperforms the BP neural network in terms of time efficiency and accuracy.Understanding the strengths and weaknesses of these neural network models in predicting wind pressure coefficients provide valuable references for constructing similar neural network models.