Identification Of Bridge Structures Modification Based On Automatic Parameter Extraction And Environmental Impact Separation

As an important basic transport facility,the damage of bridge would cause great harm on society.In order to identification the overall early performance modificatons of bridge structure,a reasonable health monitoring system must be established.In this paper,a continuous dynamic health monitoring system is established on the bridge to obtain the dynamic response signal of bridge structure.Based on the automatic modal parameter extraction technique,a large number of dynamic parameters are obtained,and then the s tatistical identification model of dynamic parameters and environmental factors is established to eliminate the influence of environmental factors.The index that is sensitive to the whole structure modifications is extracted and the overall performance of the structure is found to be changed.In order to obtain a large number of accurate modal parameters,the Covariance-dependent random subspace method(SSI-COV)basic theory and fuzzy clustering algorithm are used to obtain a stable graph with numerical mode extracted.The multiple linear regression method is applied to establish a statistical model of dynamic parameters and the corresponding environment al factors.The residual matrix which calculated by the optimal statistical model,is applied on outlier analysis to obtain the novelity index which is sensitive to the structural change.Thus the overall performance of the structure is identified.Based on the basic requirements of data acquisition and automatic modal parameter extraction,Lab VIEW software environment which is suitable for national instrument is used to develop data continuous acquisition and automatic mode parameter extraction program.Data acquisition program achieve continuous data collection,data preprocessing and automatic timing storage function through a modular programming way.Through the pre-progressing module of the automatic modal parameter extraction procedure,the data whic h is detrended and decimated is obtained to establish a stable graph based on SSI-COV theory.The automatic modal parameter extraction is realized by removing the numerical mode in the stable graph.The applicability of the automatic acquisition program and the automatic modal parameter extraction procedure was tested by the 9 days continueous experiment on the Binzhou Yellow River Bridge model.Numerical model of Dasha River Bridge in Shen Zhen is established based on ANSYS software.By analyzing the numerical simulation and modal experiment results,a continuous dynamic monitoring system was established.The variance of frequencies in constraction phase are obtained by 42 days data.And the correlation between the structure frequency and the temperature da ta is analyzed to identify the structural performance changes during the construction phase.The 5-year monitoring data of the Pedro e Inês Bridge in Portugal is used to identify structural changes in the operational phase.The nonlinear relationship between temperature and structure frequency was established by using multiple linear regression method.A reference model which is applied to eliminate the temperature effect is establushed by using the data of first year,and the index that is sensitive to the structural changes are extracted.The change of the novelty index that reflects the structural change within 5 years is carried out to identify the early performance changes of the structure.