A Comparison of Modal Data Matching And Dynamic Residual Optimization in Structural Damage Detection

摘要

Damage assessment of structures using vibration data can be viewed as a parameter estimation process. This process is often posed as a nonlinear optimization problem, which can be formulated mathematically based on modal data matching (MDM) or dynamic residual optimization (DRO). In MDM, the objective function is posed to minimize the nonlinear least squares error function between the analytical and measured modal properties. The MDM approach has been studied and used extensively. The main contribution of this work is the development of DRO methods, where a dynamic residual function is defined based on the norm of the reduced dynamic residual between the analytical model and the measured modal properties. Two DRO approaches, namely the gradient-based and dynamic least squares (DLS), are compared with the MDM approach using numerical simulations and experimental test-data.