System identification based structural damage aggravation detection in a large masonry building

System identification is widely used to estimate the dynamic characteristics of structures using either nonparametric or parametric approaches. Parametric system identification constructs mathematical models to estimate dynamic characteristics such as natural frequency, damping ratio, and mode shapes. Also, the same approach can be used to characterize damage if periodic measurements are available. In this paper, we report damage aggravation in a large neoclassical masonry structure using system identification. We took measurements in 2019 and 2023 in the Institute of Engineering, Dean Office building, which was damaged by the 2015 Gorkha, Nepal earthquake. The building is a neoclassical three storied construction with a central courtyard and an extension along the N-S direction. Without major interventions and strengthening, the building is still in use. Using the numerical algorithms for subspace state space system identification (N4SID), vibration frequencies of the structure are identified for both measurements. Also, mode shapes are identified, and damage aggravation is detected using natural frequency and MAC parameters. The first mode frequency along the N-S direction is reduced by 44% and that along the E-W direction is reduced by 1.15%, which indicates a major damage aggravation along the N-S (Y) direction.