Open Access
Subscription or Fee Access
Sensor Capability Enhancement of PPP-BOTDA for Dynamic Distributed Strain Monitoring of Bridge Structures
Abstract
This article presents a signal-processing method for developing dynamic strain measurements of using distributed fiber optic sensing technique. To address the dynamic structural responses under random vibrations and shorten the nonlinear measuring error of Brillouin optical time-domain analysis (BOTDA)-based strain measurement, several analysis techniques are compared. The presented method combines the distributed sensing ability of BOTDA-based measurement and the discrete wavelet transform (DWT)-based signal decomposition to construct a low frequency band of time history vector of each spatial arrangement on the measured fiber optic sensor. The time history vector of high-level approximation coefficients of on the strained and unstrained sensing part receive similar systematic measurement errors, which is suitable for integrating a critical model of systematic errors. The critical model is used to index and modify the possible systematic measurement errors on the strained sensing part. Furthermore, the experimental results illustrate the effectiveness of the presented single-processing method. The analysis results show that the de-noised signal maintains desirable distributed measuring ability and frequency spectrum information. Finally, an existing bridge is continuous monitored by a 20-m-long fiber optic sensor and a long-gauge packaged FBG sensor under moving traffic loads. It is found that the minimum strain resolution is improved to 15 με, while the sampling rate is 0.017 s. The effectiveness of the signal process is verified.