基于PSO-BP神经网络的隧道内气动压力幅值预测  

Prediction of aerodynamic pressure amplitude in tunnel based on PSO-BP neural network

作　　者：崔峰 王汉封[1,2] 舒卓乐 CUI Feng;WANG Hanfeng;SHU Zhuole(School of Civil Engineering,Central South University,Changsha 410075,China;National Engineering Research Center for High-speed Railway Construction Technology,Central South University,Changsha 410075,China)

机构地区：[1]中南大学土木工程学院,湖南长沙410075 [2]中南大学高速铁路建造技术国家工程研究中心,湖南长沙410075

出　　处：《中南大学学报（自然科学版）》2023年第9期3752-3761,共10页Journal of Central South University:Science and Technology

基　　金：国家自然科学基金资助项目(52078505)。

摘　　要：将BP神经网络技术用于隧道内气动压力变化幅值的预测,使用粒子群算法(particle swarm optimization,PSO)对其进行优化,构建PSO-BP神经网络模型。为了验证模型的准确性和可靠性,利用收集到的数据样本对模型进行训练测试,并引入交叉验证法评估2种模型的性能。研究结果表明:PSO-BP神经网络能够准确预测不同情况下的气动压力幅值,而且在平均相对误差、平均绝对误差、均方根误差以及样本的决定系数等方面均比未优化的BP神经网络的好,具有更高的预测精度。通过建立的PSO-BP压力幅值预测模型,得到了压力幅值在不同条件下的变化规律。The BP neural network technology was used to predict the amplitude of aerodynamic pressure change in tunnel,and the particle swarm optimization(PSO)was introduced to optimize it,and the PSO-BP neural network model was constructed.In order to verify the accuracy and reliability of the models,the collected data samples were used to train the models,and cross-validation was introduced to evaluate the performance of the two models.The results show that the PSO-BP neural network can accurately predict the aerodynamic pressure amplitude under different conditions,and the average relative error,the average absolute error,the root-meansquare error and the determination coefficient of samples are better than those of the unoptimized BP neural network.therefore,it has higher prediction accuracy.Based on the PSO-BP pressure amplitude prediction model.The variation of pressure amplitude under different conditions is obtained.

关 键 词：高速列车 隧道 气动压力幅值 PSO-BP神经网络模型 交叉验证 

分 类 号：U25[交通运输工程—道路与铁道工程]