中文核心期刊
CSCD来源期刊
中国科技核心期刊
RCCSE中国核心学术期刊

重庆交通大学学报(自然科学版) ›› 2007, Vol. 26 ›› Issue (5): 13-16.

• • 上一篇    下一篇

神经网络参数识别法在重庆石板坡大桥中的应用

黎曦,顾安邦,乔为国   

  1. 重庆交通大学土木建筑学院,重庆 400074
  • 收稿日期:2006-06-26 修回日期:2006-07-05 出版日期:2007-10-15 发布日期:2015-01-22
  • 作者简介:黎曦(1981一),男,四川内江人,硕土生,从事桥梁结构研究.

Application of Neural Network Method to Parameters Identification of Shibanpo Bridge in Chongqing

LI Xi,GU An-bang,QIAO Wei-guo   

  1. School of Civil Engineering&Architecture,Chongqing Jiaotong University,Chongqing 400074,China
  • Received:2006-06-26 Revised:2006-07-05 Online:2007-10-15 Published:2015-01-22

摘要: BP神经网络法的自适应学习能力、非线性映射能力、鲁棒性和容错能力以及快速收敛能力可有效解决连续刚 构桥施工控制中参数估计的核心问题,通过实例证明,其参数估计结果与实测数据吻合性较好,识别精度较高,有相 当的实践意义.尤其是对于必须考虑非线性影响、不确定系统的控制等问题,如果经典算法识别精度低,可考虑采用 非经典神经网络算法进行重要参数的识别.

关键词: 连续刚构桥, 施工控制, BP神经网络, 参数识别, 弹性模量

Abstract: With the capacity of adaptive self-study,non-linear representation,good robusticity and error tolerance as well as rapid convergence,BP neural network could efectively solve the core problem of parameters identification in the construction of continuous rigid flame bridge.Through the example,a conclusion is drawn that the result of parameters estimation using BP network is very identical with the measure,which has practical significance with satisfactory accuracy.If the accuracy in the classic method is rather low especially in problems with non-linear and uncertain system control and SO on,the BP neural network can be used well.

Key words: continuous rigid frame bridge, construction control, BP neural network, parameters identitTcation, elastic modulus

中图分类号: