In the long-term process of use, due to the influence of various factors as material aging, load effect and environmental change, a certain degree of deformation will occur to the foundation and superstructure of the brige, the long-term monitoring for the deformation of the bridge as well as its influencing factors and the establishment of mathematical models for predicting deformation will contribute to the maintenance and management of large bridge. According to the complex characteristics of factors influencing, the improved algorithm of radial basis function neural network and network parameters was studied, and a software based on MATLAB was developed for bridge deformation predication. Combined with the monitoring data of Nanjing second Yangtze river bridge, a RBF network deformation forecasting model that based on 9 # measure point affected by many factors was established, and the deformation forecast and analysis were carried out. Research methods and the calculation and analysis results show that improved radial basis function neural network has higher prediction accuracy and can provide an important reference for the management of bridge.
[1]
Paramasivan Saratchandran,et al.
Performance evaluation of a sequential minimal radial basis function (RBF) neural network learning algorithm
,
1998,
IEEE Trans. Neural Networks.
[2]
Sheng Chen,et al.
Recursive hybrid algorithm for non-linear system identification using radial basis function networks
,
1992
.
[3]
Huang Guo,et al.
Flood level forecast model for tidal channel based on the radial basis function-artificial neural network
,
2003
.
[4]
Ioannis Pitas,et al.
Median radial basis function neural network
,
1996,
IEEE Trans. Neural Networks.
[5]
Simon R. Jones,et al.
Learning in linear systolic neural network engines: analysis and implementation
,
1994,
IEEE Trans. Neural Networks.
[6]
Yung C. Shin,et al.
Radial basis function neural network for approximation and estimation of nonlinear stochastic dynamic systems
,
1994,
IEEE Trans. Neural Networks.