应用小波支持向量机的预应力混凝土碳化深度研究  

STUDY ON CARBONATION DEPTH PREDICTION OF PRESTRESSED CONCRETE BASED ON WAVELET SVM

作　　者：郝文峰[1] 顾建祖[1] 汤灿[1] 

机构地区：[1]江苏大学理学院,江苏镇江212013

出　　处：《工业建筑》2009年第S1期931-934,共4页Industrial Construction

基　　金：建设部研究开发项目(2008K322)

摘　　要：构造了一类小波核函数,并利用小波核函数支持向量机对预应力混凝土碳化深度进行了仿真和预测。将小波函数用于SVM得出一类小波核函数,并将其与径向核函数SVM、神经网络在预应力混凝土碳化深度预测中的应用进行了对比。研究发现,SVM对预应力混凝土碳化深度的预测精度要远远高于神经网络,且小波核函数SVM比径向核函数SVM精度高。小波核函数在一定场合下精度要比径向核函数高,SVM用于预应力混凝土碳化深度的预测是可行的,且具有较高精度。A variety of wavelet kernel functions are constructed.The carbonization depth of the prestressed concrete is emulated and predicted through wavelet kernel function support vector machine.The wavelet function is applied to SVM then a variety of wavelet kernel functions are obtained.It is also compared with the application of radial kernel function SVM and neural network in prediction of prestressed concrete carbonization depth.The research shows that the precision of prestressed concrete carbonization depth predicted by SVM is much higher than that predicted by neural network.The precision of wavelet kernel function proposed in this article is in some circumstance higher than that of radial kernel function.The application of SVM in prediction of prestressed concrete carbonization depth is feasible and has relatively high precision.

关 键 词：预应力混凝土 碳化 支持向量机 小波核函数 预测 

分 类 号：TU528[建筑科学—建筑技术科学]