GNSS And RTS Technologies Based Structural Dynamic Deformation Monitoring Of Bridges

It is a key task to monitor dynamic deformations of bridges in structuralhealth monitoring. Use of both Global Navigation Satellite System (GNSS)and Robotic Total Station (RTS) sensors become main means for monitoringstructural dynamic deformation. Based on the field tests on the NottinghamWilford suspension bridge in United Kingdom and on the Changsha SanchajiXiangjiang bridge in China, the present studies focus on distinguishing noisecharacteristics of GNSS and RTS measurements， multi-sensor integrationsystems， network-based real-time kinematic positioning technique， datafiltering technique， finite element modelling of the suspension bridges andin-situ bridge monitoring. The major contributions are provided as follows.(1) By distinguishing the error sources in GNSS and RTS measurements,the means were proposed to mitigate the corresponding errors. Caused by theionosphere and troposphere delays, the errors in GNSS measurements can bemitigated using the error empirical model or the observation differencemethod. Other errors in GNSS measurements due to orbit positions, satelliteclocks, receiver clocks and receiver positions, can be reduced by thedifference method. Additionally, the data filtering and de-noising method arefrequently employed in mitigating multipath errors and instrument-self noisesin GNSS measurements. Regarding RTS measurements, the employment of anautomatic target recognition technique is capable of mitigating aiming errors,and axis errors are mitigated by calibrating instruments before use.Furthermore, the characteristics of background noise in both GNSS and RTSmeasurements were distinguished experimentally in the time and frequencydomains. Finally, the number of visible satellites and the Position Dilution ofPrecision (PDOP) were computed for the integrated GPS/Galileo navigationsystem in order to predict the measurement precision of the integratednavigation system.(2) A method integrating GNSS and accelerometer sensors to monitorstructural dynamic displacements of bridges was developed. First, a dataacquisition approach was designed to synchronously collect multimode GNSSdata in three modes, i.e. the conventional Real-Time Kinematic (RTK) mode,the Network-based Real-Time Kinematic (NRTK) mode and the Post- Processing Kinematic (PPK) mode. Second, a Multimode Adaptive Filtering(MAF) algorithm combining a Least Mean Square (LMS) adaptive filter withChebyshev highpass filter was proposed. Third, a precise time data logger wasthen designed to realize acquisition of time synchronization between GNSSand accelerometer sensors. Forth, a special cage was used to ensure a rigidvertical axis between the GNSS and accelerometer sensors. Finally, in-situexperiments were conducted on the Wilford suspension bridge in Nottingham(UK). The above approaches including the multimode data acquisition and theMAF algorithm were employed in these experiments to detect load-induceddynamic responses of this bridge. The vibration displacements and modalfrequencies of this bridge were accurately computed from these measureddisplacement and acceleration series.(3) A method of integrating RTS and accelerometer sensors to monitorstructural dynamic displacements of bridges was proposed. A standardizedprocedure was established, in which dynamic displacements and vibrationfrequencies of bridges could be measured by the RTS. An integrated RTS andaccelerometer system with an easily installed device was developed tomeasure bridge dynamic responses. By simulating experiments for assessingmeasurement accuracy of RTS sensors, an empirical formula to analyze staticmeasurement accuracy of RTS sensors was derived. Finally, the integratedRTS and accelerometer system was used to monitor the load-induced dynamicresponses of the Wilford suspension bridge in Nottingham. The quasi-staticdisplacements and dynamic displacements were accurately identified with sub-millimeter level accuracy; the modal frequencies of this bridge wereaccurately detected with relative errors of less than1.8%.(4) A network-based real-time kinematic (NRTK) GNSS technique wasproposed to monitor dynamic deformation of bridges. The characteristics ofbackground noise in NRTK GNSS measurements were understood by thesimulated and full-scale experiments. Based on noise characteristics in NRTKmeasurements, a wavelet filtering scheme was designed to process themonitoring data. The feasibility of using NRTK GNSS technique to monitorstructural dynamic deformations of bridges has been demonstrated. Themeasurement accuracy of NRTK GNSS technique is comparable for the bridgestructural health monitoring similar to traditional GNSS technique, with loweroperational cost. (5) Monitoring experiments were conducted on the Changsha SanchajiXiangjiang bridge, and the results can be verified by the finite element modelof the bridge, established by MIDAS/Civil according to the bridge structure aswell as the material characteristics. The dynamic characteristics of the bridgewere obtained by the static and dynamic calculations, and the modal analysis.The dynamic responses of the bridge, induced by the traffic loads, weremonitored using the RTS and accelerometer monitoring techniques. Themaximum deflection of the main girder was acquired, caused by moving loadsof heavy vehicles across the bridge. The empirical formula to analyze thedynamic measurement accuracy of RTS sensors was deduced according to therelated accuracies of RTS sensors due to different baselines.