Research On Condition Assessment System Of Long Span Prestressed Concrete Cable-stayed Bridge

Along with the bridge design trend of complication and bridge accidents happening continually, more attention has been drawn to safety, durability of bridge structure.Regarding the features of long span prestressed concrete (PC) cable-stayed bridges, after summarizing bridge health monitoring and condition assessment theory all over the world, with WenHui Bridge as background, index system of bridge condition evaluation based on AHP was introduced and a multi-layer assessment model was proposed then. Finally, the algorithm was programmed based on Matlab, and it provided an academic method on condition assessment of long span PC cable-stayed bridge.In this thesis, a confined optimization method for cable stayed bridge was applied with the moment distribution of girder under dead loads as optimization objection variables, and the increment of cable tensions as optimization variables. Results show that the presented method is easy to apply, reasonable and effective. In respect that information inadequacy, relationship ambiguous, it is difficult to apply current methods to analyse and predict bridge working condition. Using prediction theory of sequence of number, according to the condition assessment results of earlier stage, degeneration of the bridge structures was predicted. Finally, neural network (NN) and fuzzy mathematics were integrated and applied in condition assessment of long span PC bridges. 3-layer fuzzy BPNN (back-propagation neural network) was proposed, DV (damage value) function and degree of grade membership model were applied then. The applications of fuzzy mathematics in condition assessment of bridges were discussed.