A NOVEL METHOD BASED ON THE CORRELATION ANALYSIS AND RBFNN FOR RESTORING NONLINEAR BRIDGE DEFLECTION
The data acquiring system of bridge structural health monitoring has got massive data for long time. Whether the data is right or wrong will affect the end judgment result directly. Despite the acquisition system of the bridge structural health monitoring system(BSHMS) is designed by the advanced sensors, the abnormity deflection is occurred by the acquisition system is the main reason to bring the illusive alarm(above 80％). A novel method based on the correlation analysis of bridges checking points and the RBF neural networks is proposed for restoring nonlinear deflection abnormity data. This paper puts forward the definition of the correlation, mining relations among checking points, then according to none-linear approximation of Radial Basis Function(RBF) neural networks, establishes the RBF neural network data prototype to restore abnormity data.. Compared with conventional methods(its MSE is 2e-9), the proposed approaches MSE is 0.6974) assures more accurate and accords with practice. Simulation results verify the effectiveness of the designed method and theoretical discussions.
Shunren Hu Weimin Chen Yumei Fu
Key Laboratory for Optoelectronic Technology & System Under Ministry of Education, Chongqinq Univers Key Laboratory for Optoelectronic Technology & System Under Ministry of Education, Chongqinq Univers
国际会议
5th International Conference on e-Engineering & Digital Enterprise Technology(第5届e工程及数字企业国际学术会议)
贵阳
英文
280-284
2006-08-16（万方平台首次上网日期，不代表论文的发表时间）