摘要
在分析普通混凝土强度各影响因素的基础上,选取6个影响因素组成输入层,以混凝土28 d强度作为输出,建立径向基函数网络,经网络训练和仿真结果对比,表明所建网络结构合理、收敛速度快、精度高,可以满足普通混凝土强度预测要求,具有广泛的应用前景。
After analyzed the influencing factors on common concrete strength,the paper is based on Radial Basis Function neural network with six influencing factors as input parameters and 28 d strength as output one.After massive network trains and emulation results comparison,it is indicated that the network to forecast the strength of the normal concrete has advantages like reasonable structure,quick convergence and high accuracy.It proves that the network has a widely application prospect in forecasting the strength of the concrete.
出处
《山西建筑》
2008年第16期66-67,共2页
Shanxi Architecture
关键词
径向基神经网络
混凝土强度
预测
Radial Basis Function neural network,concrete strength,forecast
