Parameter Inference for Structural System Identification Based on Static State Estimation

Building on previous work that introduced an observability analysis (OA) based on static stateestimation (SSE) for structural system identification (SSI), this study extends SSE to performparameterinference by augmenting the state vector with structural parameters in addition to conventionalstate variables. Performing OA beforehand ensures that the selected measurements enable uniqueand robust parameter recovery. The estimation problem is formulated as a weighted nonlinearleast-squares optimization and solved through an iterative nonlinear process, with both structuralanalysis and parameter derivatives of the stiffness matrix obtained numerically using third-partyfinite element software.The framework unifies state and parameter estimation in a single formulation, enables the use ofhigh-fidelity models for response and sensitivity calculations, and incorporates robust handling ofheterogeneous measurements and faulty data. Numerical studies showaccurate recovery of stiffnessparameters under varied measurement layouts, uncertainty levels, and data quality, confirming themethod's robustness for model updating and structural health monitoring.