基于BP、RBF神经网络混凝土抗压强度预测  

Prediction of Concrete Compression Strength Based on BP and PBF Neural Network Theories

作　　者：李红[1] 彭涛 

机构地区：[1]长安大学建筑工程学院,西安710061 [2]河南徐辉建筑工程设计事务所

出　　处：《武汉理工大学学报》2009年第8期33-36,共4页Journal of Wuhan University of Technology

基　　金：国家"863"项目(2007AA03Z109)

摘　　要：为了预测混凝土的抗压强度,在分析BP、RBF神经网络原理的基础上,提出用BP、RBF神经网络模拟混凝土抗压强度与搅拌机各主要影响参数间关系的方法。根据搅拌机的实际工作状况,分别建立了4维输入向量、1维输出向量的BP、RBF神经网络模型,通过19组试验,验证了2种模型的可靠性。结果表明,实测结果与预测结果相接近,该2种神经网络模型能较准确地快速预测混凝土抗压强度。For prediction of concrete compression strength, the method for simulating the relation between concrete compression strength and major influencing factors of stiming mill using BP and RBF neural network is presented. According to the practical working situation of stirring mill, four-dimensional output vector BP and one-dimensional output vector RBF neural network are established. The reliability of the two models is proved by 19 tests. The result indicates that the test data is closed to the prediction, these two neural network models is a preferable method for predicting concrete compression strength precisely and quickly.

关 键 词：BP神经网络 RBF神经网络 预测 混凝土 抗压强度 

分 类 号：TP31[自动化与计算机技术—计算机软件与理论] TU528.1[自动化与计算机技术—计算机科学与技术]