Monitoring-based System Fatigue Reliability Assessment Of Orthotropic Steel Bridge Decks

The development of steel bridges is currently confronted with unprecedented opportunities with the increasing implementation of advanced concepts on green transportation and environmental protection in China.The issue on the applicability and safety of steel bridges during the service period has become a significant demand in the field of civil engineering.On the one hand,steel bridges suffer from the combination effects of material performance degradation and long-term external load during the life-circle service period,which will inevitably leads to structural damage and fatigue performance degradation.On the other hand,it is impossible to accurately determine the fatigue performance of steel bridges in the design stag due to insufficient consideration with respect to the uncertainty of stochastic external loads and the evolution of structural fatigue resistance.With the advancement of sensing technology and big data analysis theory,structural health monitoring(SHM)has developed rapidly as an emerging interdisciplinary in the field of fatigue assessment for steel bridges.Aiming at the improvement of long-term fatigue performance of steel bridges in consideration of the time-varying effect and evolution law of fatigue performance.This thesis is to carry out the research work in the field of fatigue monitoring and reliability assessment for steel bridges.The main research work and outcomes are as follows.(1)The short-grid fiber Bragg grating sensor for fatigue stress monitoring technology is developed to solve the problem of limited sensor placement on welded joints in orthotropic steel bridge decks,which implements the refined stress measurement for welded joints in steel bridges.The short-grid fiber grating sensor with a grating length of only 1mm is designed for different types of welded joints in orthotropic steel bridge decks.The grating length is one tenth of that of ordinary fiber grating sensors,which greatly reduces the length of the stress measurement area.The sensor parameters,optimization layout and monitoring system installation and commissioning process are described in detail in this paper.The effectiveness of the stress monitoring system is finally validated by comparing the data obtained from different monitoring locations.Meanwhile,the distribution forms of the stress concentration factors of the two types of welded joints under different statistical durations are determined(2)The uncertainty quantification approaches of fatigue monitoring data and stress spectrum probability modeling methods are developed to solve the computational efficiency and accuracy problems in the process of parameter identification,which effectively characterizes the multi-modal characteristics of fatigue monitoring data and stress spectrum.Probabilistic modeling analysis of stress amplitude,average stress and their joint distribution was carried out.Meanwhile,the statistical characteristics and probability distribution of stress concentration factors of two types of welded joints of ortho tropic steel bridge decks are determined.Comparative study is performed in probabilistic modeling of the daily stress spectrum of welded joints according to the optimal fitting criterion.The accuracy and effectiveness of three parameter estimation algorithms are checked and compared to select the optimal probability model from the mixed Normal distribution,mixed Lognormal distribution and mixed Weibull distribution.(3)A universally applicable fatigue model both in deterministic and probabilistic manners for steel bridges is established to solve the problems of insufficient consideration of stress concentration effect and limited selection of S-N curve,which will provide reference for the consistent fatigue evaluation of the welded joints in steel bridges.Base on the field monitoring approach of the structural stress,the fatigue life of welded joints is determined in combination with the Miner fatigue linear damage accumulation criterion.The probability-based formula of the master S-N curve and the probability distribution of each parameter are derived for further analysis.Case study on time-dependent fatigue reliability evaluation for welded joints of rib-to-diaphragm is carried out and then applied in engineering practice.(4)A system fatigue reliability evaluation method for orthotropic steel bridge deck system based on Bayesian network is proposed and is used to identifiy the key fatigue-prone dangerous areas of orthotropic steel bridge decks,which will improve the anti-fatigue design and fatigue life prediction level of steel bridges.Considering the interaction between weld joints and the importance of individual component to the overall orthrotropic steel bridge deck system,the system fatigue reliability evaluation approach is proposed based on the fatigue reliability analysis model for individual joints and the theory of Bayesian network.Case study on the system fatigue reliability of the orthrotropic steel bridge deck under three different stress states was carried out and then applied to the system fatigue reliability evaluation in engineering practice.The outcomes of this sduty will provide reference for fatigue design,maintenance and evaluation of steel bridges.