Substructure identification for plane frame building structures

• Substructure identification and damage detection is proposed for frame structures.
• A novel simplified model for frame structures is used to reduce the modeling error.
• An approximate error measure is derived analytically and used to maximize accuracy.
• Method demonstrated using simulation of a 5-story 2-bay structure.
• Damage is very accurately detected and localized even with very large sensor noise.

In previous studies (Zhang and Johnson, 2012, 2013) [1,2], the authors proposed a substructure identification method for shear structures. This method is extended herein to accommodate more complicated plane frame building structures. Via a model reduction technique, a multi-story multi-bay plane frame is converted into a beam-like model, having one translational and two rotational degrees-of-freedom (DOFs) for each floor. Based on this model, an inductive substructure identification method is formulated, using the cross power spectral densities between the frame floor horizontal accelerations and a reference response, to identify the equivalent story stiffness and damping parameters of the frame from top to bottom inductively. An identification error analysis reveals how structural responses affect the identification accuracy. Exploiting this result, a reference selection rule is proposed to select the best reference response and further improve the identification accuracy. Finally, a five-story two-bay frame structure is used to demonstrate the efficacy of the proposed substructure identification, very accurately identifying the structural parameters even under the influence of large measurement noise; based on the estimated parameters, a hypothesis testing method is used to detect and locate structural damage, simulated by stiffness reduction of some columns.