Structural Condition Alarming For Multi-Tower Suspension Bridge Based On GPS-RTK Monitoring

With the vigorous development of the construction of long-span bridges, scientifical and reasonable damage alarming and safety evaluation of long-span bridges with massive data collected by the bridge structural health monitoring system (BSHMS) has been widely concerned. For multi-tower suspension bridge, under the influence of middle tower, there are great differences in the mechanical behavior and deformation between multi-tower suspension bridge and traditional suspension bridge. In this thesis, the Taizhou Yangtze River Highway Bridge is taken as the engineering background, accurate early warning research is carried out for multi-tower suspension bridge combined with the analysis of GPS-RTK monitoring data from the bridge structure health monitoring system. The specific contents are as follows:(1) The relevant theory of GPS technology used in the bridge deformation monitoring and the applications are systematically introduced. The development of GPS and the system composition, the basic ideas, main characteristics, and basic principles of GPS-RTK positioning system are specifically presented; and the applications for GPS-RTK in the deformation monitoring of large-span suspension bridges are studied.(2) Gross error elimination and noise reduction preprocessing are carried out based on GPS monitoring data. According to the characteristics of Taizhou Bridge GPS-RTK monitoring data, Pauta criterion (3σ criterion) is applied to eliminate outliers. Then wavelet threshold de-noising method is used with different wavelet functions, different decomposition levels, different thresholds and threshold criteria, and the de-noising effect are compared so as to choose the optimal combination of parameters.(3) The effect laws of environmental temperature on the displacement of girder are researched. First, the relationship between the three dimensional displacement of bridge girder and the ambient temperature is studied. For the gider of Taizhou Bridge, the temperature has remarkable correlation with the displacement for the longitudinal and vertical direction of bridge, and have weak correlation with the transverse displacement. Then the displacement-temperature linear regression models are set up using the least squares method, and finally, the correlation model is used to normalize the environmental temperature, and the displacement trend caused by the temperature effect is removed, and the true displacement state of the bridge structure is obtained.(4) The structural condition alarming method based on mean-value control chart is established. First, the displacement residual alarming index is constructed, the according to the mean-value control chart within statistical controls, the condition alarming method of the gider of Taizhou Bridge is set up, and then with the sample test and damage simulation, the feasibility and effectiveness of the condition alarming method are verified. The analysis results show that the abnormal condition identification and damage alarming can be carried out from multiple sections and multi directions based on mean-value control chart.