Approaches Of Condition Assessment And Damage Alarming Of Bridges Based On Structural Health Monitoring

Structural health monitoring is one of present research hotspots in the field of civil engineering. With the rapid development of infrastructure construction, structural health monitoring systems have been permanently installed on the large bridges. That analysis of massive data obtained by the Structural Health Monitoring (SHM) system installed on the bridge is a key and urgent scientific problem. The well analysis results can provide the scientific evidences for safety assessment, degeneration rule and analysis theory as well as assumptions for bridges. In this dissertation, systemic research on condition assessment and damage alarming based on Structural Health Monitoring was carried out.The main research contents of this dissertation are described as follows:This paper investigates the dependency of modal frequencies, modal shapes and the associated damping ratios on temperature and wind velocity. The Nonlinear Principal Component Analysis (NLPCA) is first employed as a signal pre-processing tool to distinguish temperature and wind effects on structural modal parameters from other environmental factors. The pre-processed dataset by NLPCA implies the relationship between modal parameters and temperature as well as wind velocity. Consequently, the Artificial Neural Network (ANN) technique is employed to model the relation between the pre-processed modal parameters and environmental factors. Numerical results indicate that pre-processed modal parameters by NLPCA can remain the most features of original signals. Furthermore, the pre-processed modal frequency and damping ratios are dramatically affected by temperature and wind velocity. The ANN regression models have good capacities for mapping relationship of environmental factors and modal frequency, damping ratios. However, environmental effects on the entire modal shapes are insignificant.This paper presents the model updating strategy for uncertain structures by utilizing uncertain experimental modal parameters. When there are sufficient modal parameters, the mean value and standard deviation of measured random modal parameters can be achieved. The mean values and variance of measured random modal parameters are chosen to formulate the objective functions and standard deterministic optimization algorithm is employed to solve the optimization problems. Both the mean value and associated standard deviation of random structural parameters can be obtained. When there are no sufficient modal parameters, only the upper and lower bound of modal parameters are available. Then the model updating for uncertain structures can be transformed into two deterministic constrained optimization problems. The differences and indifferences of the proposed two methods are investigated. The subsequent optimization strategy is employed to updating the uncertain structural parameters of the finite element model. The updated uncertain finite element model can be used for the subsequent reliability analysis.This paper presents the integrity analysis of strain time histories measured by structural health monitoring system permanently installed on bridge. The strain and temperature data is collected at the following stages: closure segment reconstruction stage and operating stage. The section coefficients are also introduced in this study, which can provide evidence for the damage detection under regular use. The maximum distribution of vehicle and temperature induced strain as well as bending moment can be obtained from the integrity analysis.This paper presents the reliability and fatigue life estimation approaches based on structural health monitoring. The integrity analysis of strain time histories gives information about maximum distribution of vehicle and temperature induced strain as well as bending moment. The resistance attenuation is also considered in this study based bridge detection. Finally, the first order reliability method (FORM) is employed to estimate the reliability index of main compotes of the bridge. At the mean time, the strain response at certain time interval can be chosen as the typical burdening cycle block. The fatigue lifetime can be estimated by combining previous S-N curves and typical burdening cycle block.This paper presents the bridge damage alarming based on statistical process control theory. The modal parameters of healthy bridge are used to train the NLPCA model and the SPE is employed to demonstrate how much the NLPCA interpreting the dataset. If the SPE is higher than the threshold, abnormal situation occurs and the effectiveness of the proposed method is verified by the modal parameter dataset collected by SHM system.