Research On Structural State Analysis Based On D-S Evidence Theory And Bridge Deflection

Since the beginning of the 21 st century,major bridge safety accidents have occurred in China,causing serious losses to people's lives and property safety.Therefore,it is necessary to establish a proper bridge structure condition assessment system for the condition assessment of bridges.At present,there are many methods for assessing the condition of bridge structures,and the scope involved is also quite extensive.However,the method of assessing bridge deflection as a structural condition can better reflect the actual operation status of the bridge.Under actual conditions,bridge deflection is the superposition of multiple action effect deflection signals,which can be separated into vehicle load effect deflection,long-term downward deflection effect deflection,and temperature load effect deflection through related algorithms.The D-S evidence theory has the ability to handle multi-source uncertain information fusion,and can form intuitive and reliable decisions.Therefore,this paper takes the issue of bridge deflection structure state assessment as the starting point,and carries out research on the construction of bridge deflection classification early warning mechanism and the improvement of D-S evidence theory,and forms a structural state assessment system based on improved D-S evidence theory and bridge deflection.The main research contents and conclusions are as follows:(1)Established a graded early warning mechanism for bridge deflection.First of all,numerical simulations were carried out considering the effects of actual vehicle flow loads and bridge design loads.The deflection and deformation laws of bridges under vehicle loads were mastered.Based on relevant specifications and extreme values of vehicle load deflections,a vehicle load effect deflection grading early warning mechanism was constructed.In order to grasp the long-term downward deflection characteristics of typical long-span beam bridges under the influence of internal and external factors,the data of the typical deflections of domestic and foreign bridges was calculated.Based on these statistical data,the characteristics of the long-term downward deflections of long-span beam bridges were analyzed.The longterm downward deflection effect deflection grading early warning mechanism was obtained.Finally,the finite element numerical simulation method of bridge structure deflection under the action of temperature field was deduced.The gradient temperature and the overall heating characteristics were considered respectively.The temperature deflection limit requirements according to the code were combined with the numerical simulation results.,A grading early warning mechanism for temperature load effect deflection was established.(2)An improved D-S evidence theory fusion framework is proposed.First,considering that the evaluation results of the traditional basic grading method have interval uncertainty,the interval number is used to represent the evaluation results of each basic grading method.Second,the Euclidean distance theory is introduced to add the basic probability distribution function of the traditional DS evidence theory.Improvement;furthermore,considering the problem of too large conflict coefficients in the actual evaluation process,the evidence correlation coefficient method was used to modify the conflict coefficients;finally,the entropy weight method was used to assign weights to bridge effects and evaluation objects.It is integrated into the new DS evidence theory evaluation system,and a typical example is given.The evaluation effect is better.(3)The improved D-S evidence theory evaluation method was applied to actual engineering and achieved a good evaluation effect.First,the bridge dynamic deflection monitoring system was established,and the internal hardware and software components and selection design of the system were analyzed,and the zero drift problem in the engineering application of the monitoring system was solved.Second,the deflection separation method was used to perform the deflection signal.Separation,the measured deflection signal is separated into three components of vehicle load effect deflection,temperature load effect deflection and long-term downward deflection effect deflection;then,the measured data is taken as an extreme value for normal fitting and 68.27% is taken as a probability guarantee.The characteristics of the long-term deflection of the bridge are used to construct the measured data deflection interval values.Finally,the DS evidence theory improvement method is used to evaluate the bridge deflection structure state.The final structural state comprehensive assessment level is B,and the evaluation results show that the bridge structure state good.