摘要: |
Many structural health monitoring (SHM)
techniques have been devised over the past
decades. However, there is no one-size-fits-all
solution that can be applied to all bridges for
structural assessments. Bridge-based weight-in-motion systems (BWIM) use the structure’s
response to estimate vehicles’ load distribution.
This technology is primarily used to obtain vehicle
axle weights efficiently in public. BWIM can be a
candidate that overcomes the shortfall of SHM.
The use of BWIM systems for SHM has rarely been
investigated. The objectives are (1) to study and
deploy low-cost BWIM sensors for accurate SHM,
(2) to evaluate the S-BWIM system, and (3)
assessment of the hybrid model capacity
combining physics-based mathematical models
(PSM) and practical machine-learning (ML)
models. A new low-cost BWIM system verified
with numerical results will be installed in the local
area, Dallas, and Fort Worth (DFW). This study will
help Region 6 communities, where low-cost
measurements are already used, and prediction
models are publicly available for monitoring both
traffic loads and bridge conditions. The
development of a hybrid model generally
adaptable for various conditions of bridges will be
a major contribution to the research community.
A comparative study of the proposed machine
learning algorithm (super learner) with low-cost
BWIM sensors will also be implemented in Region
6. |