Iterative two-stage approach for identifying structural damage by combining the modal strain energy decomposition method with the multiobjective particle swarm optimization algorithm

Abstract

Summary An iterative two‐stage structural damage identification approach that combines the modal strain energy decomposition (MSED) method with the multiobjective particle swarm optimization algorithm is presented. The proposed scheme is inspired by the general noniterative two‐stage approach, which attempts to locate damage in the first stage and estimate the severity of damage in the second stage; however, the present approach differs by constructing an iterative MSED indicator to perform damage localization and severity estimation in a cyclic manner. Consequently, false positive indications of damage are gradually suppressed, and the precision of the damage severity estimation is improved as the iteration continues. In addition, a new element activation operation is embedded into the iterative process; in this step, possible false negative indications of damage excluded in the first stage can be activated and subsequently retested to identify damage, thereby reducing the false negative rate. Furthermore, a comparative study is conducted between the iterative two‐stage approach and the general noniterative approach by contemplating a 3‐D structure, and the effects of measurement noise and spatial incompleteness are considered.

Structural damage identification with limited modal measurements and ultra‐sparse Bayesian regression

“…The example used in this numerical study is a 3‐D frame structure provided by Xu et al 54 As shown in Figure 3, it is a three‐story frame structure, with the cross area of an equilateral triangle. Each structural member is simulated as a Euler‐Bernoulli beam element.…”

Robust Structural Damage Detection Using Analysis of the CMSE Residual’s Sensitivity to Damage

“…Dessi and Camerlengo [28] also focused on technique processing information about mode shape curvature or strain modes with or without knowledge of baseline data. The general modal strain energy (MSE), as an extension of mode shape analysis, is widely appreciated because of its excellent damage-sensitive features [29,30]. It is formed by the product of the stiffness matrix and the second power of mode shape: P * j,n = (φ * j ) T K * n φ * j j = 1, 2, · · · , N j (4) where N j is the number of measured modes; φ * j and K * n are the j-th mode shape and the n-th stiffness submatrix of the damaged structure, respectively; and P * j,n is the MSE of the n-th damaged structural element associated with the j-th measured mode.…”

Bayesian estimation approach based on modified SCAM algorithm and its application in structural damage identification

“…The deterministic methods treat the characteristic parameters of the damage as a deterministic quantity, study the deterministic mapping relation between the characteristic parameters of the damage and the damage itself, and identify the damage through deterministic calculations and reasoning based on the mapping relation 1 . Commonly used deterministic methods include the following: (1) static and dynamic fingerprint methods based on natural frequency, 2 mode shape, 3 modal strain energy, 4,5 and modal curvature 6 ; (2) damage detection methods based on model updating 7,8 ; (3) identification methods based on measured time‐domain signals, including the Hilbert‐Huang transform, 9 and wavelet analysis 10,11 ; and (4) artificial neural network‐based identification methods 12,13 . However, in actual engineering projects, there are uncertain factors, such as complex properties of the structure itself and the impact of ambient noise in the damage problems 14 .…”