Prediction of Chaotic Time Series of Bridge Monitoring System Based on Multi-Step Recursive BP Neural Network

Jian Xi Yang; Jian Ting Zhou

doi:10.4028/www.scientific.net/AMR.159.138

Paper Titles

An Improvement Test Approach of Look-up Table in SRAM-Based FPGAs
p.116

Segmentation and Semantic Annotation of 3D Objects
p.124

Semantic Enabled 3D Object Retrieval
p.128

Study on Detection System for Bend Elastic Modulus of Veneer Based on CPLD
p.132

Prediction of Chaotic Time Series of Bridge Monitoring System Based on Multi-Step Recursive BP Neural Network
p.138

Improvement of Apriori Algorithm Based on Boolean Matrix
p.144

Objective Evaluation on Half-Circle Skirt Shape Style Based on 3D Virtual Garment Simulation
p.149

P-Type Adiabatic Sequential Circuits for Leakage Reduction of Nanometer Circuits
p.155

The Empirical Study on Factors Influencing the Disclosure of Internal Control Information of Listed Companies
p.162

HomeAdvanced Materials ResearchAdvanced Materials Research Vol. 159Prediction of Chaotic Time Series of Bridge...

Prediction of Chaotic Time Series of Bridge Monitoring System Based on Multi-Step Recursive BP Neural Network

Abstract:

BHM is an important means to assess and predict the safety operation of large bridge in service around the world. Given the missing of real-time monitoring information for some time and the lack of effective theory and technique to capture the missing information and even to predict the evolution of structure, this paper made an attempt to predict the evolution of monitoring information using time series and chaotic theory. Firstly, maximum Lyapunov exponent of available monitoring information is calculated to assess the chaos of the bridge structure. The parameters of reconstructed phase space, correlation dimension and time delay, are calculated by C-C algorithm and G-P algorithm respectively. According to empirical formula, one 3-layer BP neural network is established Ten recursions are carried out. The results show that multi-layer recursive BP neural network is able to predict BHM information. Using chaotic time series to reconstruct phase space and applying multi-layer recursive BP neural network to predict BHM information facilitates further estimation and prediction of bridge safety condition by means of chaotic nonlinear characteristics.

You might also be interested in these eBooks

Micro Nano Devices, Structure and Computing Systems

View Preview

Info:

Periodical:

Advanced Materials Research (Volume 159)

Pages:

138-143

DOI:

https://doi.org/10.4028/www.scientific.net/AMR.159.138

Citation:

Cite this paper

Online since:

December 2010

Authors:

Jian Xi Yang, Jian Ting Zhou

Keywords:

Back Propagation, Bridge Monitoring System, Chaotic Time Series , Multi-Step Recursive, Neural Network (NN), Prediction

Export:

RIS, BibTeX

Price:

Permissions CCC:

Request Permissions

Permissions PLS:

Request Permissions

Сopyright:

Citation:

References

[1] Zonghua Liu, Fundamentals and Applications of Chaotic Dynamics, Beijing, Higher Education Press, (2006).

Google Scholar

[2] Han Min, Chaotic time series prediction theory and method, Beijing, China Water Power Press, (2007).

Google Scholar

[3] Yu Zhang, Time series analysis of chaotic and symbols and practices, Changsha: National Defense University Press, (2007).

Google Scholar

[4] Zhishan Liang, Minli Wang，Electric Power System Short-term Load Forecasting Using Lyapunov Exponents Technique, Proceedings of the Chinese Socity for Electrical Engineering , 1998，18（5）: 368~371.

Google Scholar

[5] Xiang Zhao, Yueming Sun, Application of Lyapunov Exponent to Forecast of Remnant Life of Rotor, Proceedings of the Chinese Socity for Electrical Engineering, 1999, 19（10）: 10~13.

Google Scholar

[6] R Norega, Hong wang, A direct adaptive neural network control for unknown nonlinear systems and its application[J], IEEE trans Neural Network., 1998, 19 (1): 27-34.

DOI: 10.1109/72.655026

Google Scholar