Abstract:Modal curvature is more sensitive to structural damage than directly measured mode shape, and the standard Laplace operator is commonly used to acquire the modal curvatures from the mode shapes. However, the standard Laplace operator is very prone to noise, which often leads to the degraded modal curvatures incapable of identifying damage. To overcome this problem, a novel Laplacian scheme is proposed, from which an improved damage identification algorithm is developed. The proposed step-by-step procedures in the algorithm include: (1) By progressively upsampling the standard Laplace operator, a new Laplace operator is constructed, from which a Laplace operator array is formed; (2) by applying the Laplace operator array to the retrieved mode shape of a damaged structure, the multiresolution curvature mode shapes are produced, on which the damage trait, previously shadowed under the standard Laplace operator, can be revealed by a ridge of multiresolution modal curvatures; (3) a Gaussian filter is then incorporated into the new Laplace operator to produce a more versatile Laplace operator with properties of both the smoothness and differential capabilities, in which the damage feature is effectively strengthened; and (4) a smoothened nonlinear energy operator is introduced to further enhance the damage feature by eliminating the trend interference of the multiresolution modal curvatures, and it results in a significantly improved damage trait. The proposed algorithm is tested using the data generated by an analytical crack beam model, and its applicability is validated with an experimental program of a delaminated composite beam using scanning laser vibrometer (SLV) to acquire mode shapes. The results are compared in each step, showing increasing degree of improvement for damage effect. Numerical and experimental results demonstrate that the proposed novel Laplacian scheme provides a promising damage identification algorithm, which exhibits apparent advantages (e.g., high-noise insusceptibility, insightful in damage revealment, and visualized damage presentation) over the standard Laplace operator.

Bayesian structural damage identification using fraction function regularisation model based on natural frequency and mode shape curvature

A fraction function regularization model for the structural damage identification

An lp Regularization Model Based on Weighted Natural Frequencies and Strain Modes for Slight Damage Identification

A joint fraction function regularization model for the structural damage identification

Structural Damage Detection Based on L(1) Regularization Using Natural Frequencies and Mode Shapes

A Robust Sparse Bayesian Learning Method for the Structural Damage Identification by a Mixture of Gaussians

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

Structural damage detection based on fundamental Bayesian two-stage model considering the modal parameters uncertainty

Baseline-free Multidamage Identification in Plate-Like Structures by Using Multiscale Approach and Low-Rank Modelling

Fractal Dimension Based Damage Identification Incorporating Multi-Task Sparse Bayesian Learning

L1 Regularization Approach to Structural Damage Detection Using Frequency Data

A structural damage identification method based on sensitivity analysis of modal parameter error function

Structural Damage Detection Based on Variational Bayesian Inference and Delayed Rejection Adaptive Metropolis Algorithm

Structural Damage Detection with Two-Stage Modal Information and Sparse Bayesian Learning

Bayesian Probabilistic Inference for Nonparametric Damage Detection of Structures

Structural Damage Identification Based on Substructure Sensitivity and L 1 Sparse Regularization

Structural Damage Localization and Quantification Based on Additional Virtual Masses and Bayesian Theory

Novel Laplacian Scheme and Multiresolution Modal Curvatures for Structural Damage Identification

Damage identification based on Baysian fusion of global and local information

A Fast Friction-Model-inspired Sparse Regularization Approach for Damage Identification with Modal Data

Model-calibration-free Damage Identification of Shear Structures by Measurement Changes Correction and Sparse Regularization