Hybrid Sensing Systems For Bridge Health Monitoring Applications

In the long-time use, bridges will be damaged to a certain degree due to the traffic loads, climate, natural disasters, and human factors. In order to monitor the health of the bridges, and reduce the potential losses, this paper has investigated a monitoring scheme for the bridge health based on composite sensing technique.With the development of the fiber technology, fiber grating has been attracted more and more interesting due to the many advantages, such as small size, low cost, high sensitivity, and electromagnetic interference immunity. Firstly, with the demand of the bridge health monitoring system, this paper analyzed the advantages of the fiber grating sensor compared to the traditional electrical sensor. Successively, the temperature and strain were calibrated by using electric sensors. Finally, the design of the bridge health monitoring system has been completed based on composite sensing technique (optical sensor&electric sensor), which realized the simultaneous monitoring for temperature and strain.The main contributions in this paper are as follows.1. The temperature and strain are calibrated in the fiber grating sensing systems, and the temperature compensation scheme using two fiber gratings has experimentally demonstrated, which can realize the composite information monitoring for stain and temperature.2. The wavelength demodulation schemes of fiber grating have been investigated. In this paper, the tunable F-P filtering scheme is used to demodulate the wavelength of FBG sensor, because it can directly convert the wavelength information to the electric signals, has small size and low cost.3. The fiber grating sensor demodulation scheme of up to eight channels is proposed based on ARM. Firstly, the time sequence synchronization and data sampling have been realized using CPLD. Then the original data are sent to the ARM, where the moving average filtering and peak clipping have been realized. Finally, the post-process data are sent to the computer. Furthermore, the communication protocol between computer and demodulator is designed using C#, which realizes the remote control and graphical display.4. The filtering scheme and peak clipping have investigated in detail in this paper, and several peak detection algorithms for wavelength are compared. Taking the errors and system stability into account, the Gaussian polynomial fitting method is more suitable for this system. Furthermore, the installation of fiber grating sensor is introduced according to the data processing scheme.