RBF神经网络在坝肩断(夹)层位移监测中的应用

Application of RBF Neural Network to Displacement Monitoring of Abutment Fault and Interlayer

针对RBF神经网络可解决BP算法收敛速度慢、易陷于局部最小等问题,介绍了RBF下神经网络的训练与设计方法,构建了某坝坝肩断(夹)层位移数据分析的RBF神经网络,并对测试样本进行了预测。结果表明,采用该法的预测误差约为±0.2 mm,预测效果较好。

RBF neural network can overcome the BP algorithm drawbacks including slow convergence and easy to trap in local minimum. This paper introduces the training and design method of RBF neural network and establishes an RBF neural network model for analysis of the displacement data of a certain abutment fault and interlayer, which predicts the test samples. The results show that the prediction error by this method is approximately 0.2 mm and prediction effect is good.