Bayesian Parameter Inversion and Analysis of Planar Truss Structures Based on OpenSees

The safety performance of engineering structures is inherently influenced by diverse uncertain factors, including material property variability, geometric imperfection, and stochastic external loads. Quantifying these uncertainties' effects on structural responses forms a critical foundation for reliability assessment and risk mitigation. Effective structural performance evaluation not only prevents catastrophic engineering failures but also supports decision-making in maintenance strategies. This study proposes a probabilistic reliability assessment framework for planar steel truss structures through systematic integration of finite element modeling using OpenSees with Bayesian updating theory. The methodology establishes a bidirectional relationship between numerical simulations and observed structural behaviors through probabilistic parameter calibration. Case studies reveal two principal outcomes: (1) The Bayesian-based approach significantly reduces model parameter uncertainties, leading to enhanced accuracy in predicting structural displacement responses; (2) The implementation of combined horizontal-vertical loading conditions strengthens the nonlinear interaction between structural parameters and displacement characteristics, thereby improving the completeness of parameter identification. These findings demonstrate the method's capability of addressing multi-source uncertainties and complex loading patterns, offering valuable insights for structural health monitoring and performance optimization in practical engineering applications.