Real-time Identification Of Time-varying Bridge Cable Tension Forces Based On Wireless Sensor And Sparse Time-frequency Analysis

Fatigue damage is the dominate factor causing the damage and failure of bridges. For cable-supported bridges, cable is the critical stress component and the cumulative fatigue damage of cable is a grave threat to the safety of bridge structures. The traditional vibration-based cable tension force estimation methods can only obtain the time-average cable tension force and not the instantaneous force. However, cable tension forces vary in real time with the change of the moving vehicle loads and environmental effects, and this continual variation in tension force may cause fatigue damage of a cable. The identification of time-varying cable tension forces is the foundation for limit state safety evaluation and fatigue cumulative damage evaluation. Therefore, the real-time identification of time-varying cable tension forces of bridges based on wireless sensor and sparse time-frequency analysis is studied.Main contents are included as follows:A new method to identify the time-varying cable tension forces has been proposed based on adaptive sparse time-frequency analysis method. The adaptive sparse time-frequency analysis method is a newly developed method in signal processing field, the principle of it is that looking for the sparsest time-frequency representation of the signal within the largest possible time-frequency dictionary to obtain instantaneous frequencies. First, time-varying modal frequencies are identified from the measured acceleration data of cable using the sparse time-frequency analysis method. Then the time-varying cable tension forces are calculated by the relationship between cable tension force and time-varying modal frequency based on the flat taut string theory. By utilizing the integer multiples relations between high order modal frequencies and fundamental frequency of the cable, the optimization variables of the method are reduced. This will reduce the effects of measurement noise. Finally, the time-varying cable tension forces are calculated by taking the identified time-varying modal frequencies into the cable tension formulation.Considering multiple cases of cable tension force variation, wired sensor experimental data of cable model are empolyed to investigate the effects of cable tension force variation level and rate on the identification accuracy of the proposed method. Then, considering that data loss may occur when using wireless sensor, acceleration data measured by wired sensors are used to simulate the possible data loss. The acceleration data with data loss are used to identify the time-varying cable tension forces using the proposed method in order to investigate the impact of different data loss pattern and loss rate on identification results.Vibration tests on an experimental cable model are conducted using Imote2 wireless sensors with considering multiple cases of cable force variation. Since wireless sensors are sensitive to external environment, the data loss is investigated in the experiment. The experiment data without data loss and with data loss are analyzed separately based on the adaptive sparse time-frequency analysis method. Then the time-varying cable tension force is calculated by the time-vaying modal frequencies. In additional, the comparison of the identified results without and with data loss are implemented to verify the accuracy and robustness of the proposed cable force identification method. Furthermore, field test on Xiamen Haicang Bridge is carried out. The Imote2 wireless sensors are used to measure the vibration acceleration of the suspension cables. Then, the time-varying modal frequencies are identified and the corresponding cable tension forces are calculated, which further verify the ability of the proposed time-varying cable tension force identification method for actual bridges.