A Rapid Evaluation Method for the Vertical Condition of Bridge Substructures Using Bayesian System Identification and Output-Only Measurements

Abstract

Evaluating the bearing capacity of bridge substructures is very important for bridge maintenance and management. However, existing studies that rely on static load tests (SLTs) or transient response methods (TRMs) have limitations that are difficult to apply to operational bridges or require knowledge of the relationship between static stiffness and dynamic stiffness. This paper proposed a novel Bayesian system identification framework for rapid assessment of the vertical condition of bridge substructures. In the first step, a simplified analytical model was formulated to interpret the vertical dynamics of the soil–foundation–bridge pier system with lumped parameters. A Bayesian joint–input–parameter–state procedure was introduced to simultaneously identify unknown input and structural parameters, including stiffness and damping coefficients. After that, the proposed framework was numerically demonstrated, and the influence of extensive random initial errors was methodically examined. Finally, a full-scale in situ test involving TRM and SLT was conducted to further test the engineering compatibility of the methodology. The achieved results indicated that the simultaneous identification framework is effective and robust for estimating the vertical stiffness of piers and foundations, structural states, and unknown excitation using output-only measurements. The proposed framework can be effectively employed to assess the vertical condition of bridge substructures during construction or operation, particularly for rapid damage assessment of bridge structures after natural disasters.