基于逐步回归-BP神经网络的大坝变形监测模型  

Deformation Monitoring Model of Dam Based on SREG-BP Neural Network

作　　者：戚丹[1] 包腾飞[1] 康业渊[1] 柴丽莎[1] 

机构地区：[1]河海大学水利水电学院,江苏南京210098

出　　处：《水电能源科学》2013年第8期109-111,98,共4页Water Resources and Power

基　　金：新世纪优秀人才支持计划基金资助项目(NCET-11-0628);国家自然科学基金资助重点项目(51139001)

摘　　要：针对拱坝坝肩抗力体是坝体的重要受力部位,应用结合逐步回归分析方法的BP神经网络构建监测模型,以提高BP神经网络的泛化能力和模型的预测准度和精度,利用C语言编程训练,成功完成了预测,且传统模型预测数据的残差平方和大于改进后的残差平方和。实例分析结果表明,该监测模型可行、有效,并具有通用性。For the fact that arch dam resistance body is an important stressed component, BP neural network corn bined with stepwise regression method is applied to establish monitoring model for improving the generalization ability of BP neural network and the prediction accuracy and precision of the model. The prediction is completed successfully by using C language programming. The residual square sum for the prediction data of the traditional model is greater than that of the modified model. The analysis results of an example show that the monitoring model is feasible and effective, and it has generality.

关 键 词：坝肩抗力体变形 大坝变形监测 监测模型 BP神经网络 逐步回归 

分 类 号：TV698.11[水利工程—水利水电工程]