Study On The Damage Detection Method Of The Simply Supported T-shaped Girder Bridges’ Transverse Connection

Girder bridges have small tensile concrete areas and wide compressive flange plateareas, whose load-bearing reinforcing bars are intensively arranged at the bottom of thecompressive areas, so as to reduce the structure dead-weight and exert the better compressivecharacteristic of the concrete and the better tension performance of the reinforcing bars. Theenlargement of the arm of the compressive concrete and the tensile reinforcement can greatlyincrease the flexural capacity of the section. Therefore, T-shaped girder bridges are widelyapplied in China. Relevant data show that, the small and medium span bridges account forabout80percent of the total bridges in our country, and the T-shaped girder bridges are theimportant component of the small and medium span bridges.As the extension of the service period and the improvement of the traffic load grades,the bearing capacity of the early built T-shaped beam bridges can’t meet the needs of themodern traffic. Nowadays, a great majority of the T-shaped beam bridges under service havealready had damage in different styles and under different levels, among which the failure inthe transverse connection of the T-shaped is one of the most important damage. Thetransverse connection failure has the typical characteristics such as the scaling of concrete onthe connected steel plate and the crack of the connected welded seams,etc. which keeps theload-bearing beams from effectively transferring to the non-load-bearing areas.The failure in the transverse connection of T-shaped beam bridges can decrease theunity of the bridge structure and affect the transverse distributing rules. It causes thedeflection increase of the load-bearing beams and the deflection decrease of thenon-load-bearing beams, which weakens the collaboration of the load-bearing beams and thenon-load-bearing beams and easily destroys the load-bearing beams. In some serious cases,single girder loadbearing may occur, which has great potential traffic safety hazard.Because of the university of the failure in the transverse connection, it needs greatmanpower and material resources to change all the transverse connections which is alsounreasonable in economy and unscientific in technology. Aiming at this case, it’s necessaryto find an efficient method to detect the failure of the transverse connection, fully excavatethe bearing capacity of the bridge structure, and provide reasonable measures for the bridge maintenance and reinforcement.At present, there are no effective and systemic evaluation methodology of the failure ofthe transverse connection of T-shaped simply-supported beam bridge. Therefore, it’smeaningful to conduct the study on the damage detection method of the simply supportedT-shaped girder bridges’ transverse connection.1. It builds the theoretical dynamic model of the T-shaped simply-supported girderbridge with damage in the cross slab and calculates the dynamic parameters under manydamage conditions by means of numerical simulation. According to the dynamic parameterdata, it uses mathematical statistics to draw the relation curve between the damage degreeand the change of dynamic parameter, establish the figure files of the dynamic parametersunder transverse connection damage, and eventually obtain the impact law of the transverseconnection damage to the dynamic parameters. It considers:(1) when the transfer capacity ofthe cross slab plays an important action in the vibration, the effect of the slab damage on thefrequency can be obvious. In the practical projects, to the mode shape orders with highexertion of cross slab transfer capacity, the frequency can change obviously.(2)With thedecrease of the damage degree, the frequency decreases in a up-convex curve.2. The damage detection method of bridge cross connection based on the geneticoptimization of neural networks is presented. This method uses the genetic algorithm tooptimize the weight and the threshold of the Bp neural network and by choosing, crossoverand mutation operation, combine the advantages of them to detect the structure damage. Theinput parameter of the genetic optimization of neural networks is the ratio of modalfrequency rate and the mode of vibration and the output parameter is the damage degree ofthe cross connection. Use the analogue computing method to detect the damage of thesimply-supported bridge’s cross slab. The results show that, the presented method can detectthe damages of both single location and multiple location, what’s more, it can identify thedamage degree of the cross slab.3. The project of the Xinkai river middle bridge proves the feasibility and theeffectiveness of the presented damage detection method of the cross slabs. Both of the staticand dynamic load tests are conducted to the bridge. The static tests measure the displacements of the main beams, based on which the damage condition is identified. In thedynamic tests, the measured dynamic parameters are used for detection by the presentedmethod. At last, compare the cross slab states based on the static and dynamic testing dataand verify the feasibility of the presented method of this thesis.