Integrating visual sensing and structural identification using 3D-digital image correlation and topology optimization to detect and reconstruct the 3D geometry of structural damage

Abstract

This paper describes a novel technique for detecting internal or unseen damage in structural steel members by combining measurements from full-field three-dimensional digital image correlation (3D-DIC) with a topology optimization framework. Unlike the majority of conventional methods that rely on specialized forms of surface-penetrating waves or radiation imaging, this work employs optical cameras to measure surface strains and deformations using the 3D-DIC technique followed by an optimization approach to detect the existing damage. This data-rich representation of the structural component’s behavior is then used to reconstruct the underlying subsurface abnormalities via an inverse mechanical problem. The inverse problem is solved using a topology optimization formulation that iteratively adjusts a fine-tuned finite element model (FEM) of the structure to reveal irregularities within it. Having recently demonstrated the feasibility of detecting and reconstructing defects in small-scale structural components, this paper expands on the authors' previous work to demonstrate the feasibility and performance of the proposed method through an experimental program in which a set of large-scale structural steel beams with and without buried defects tested using a full-field 3D DIC sensing approach. The structure’s initial FEM is first created to discretize the member into elements whose constitutive properties are treated as unknowns in the optimization problem. The goal of the optimization is to minimize the discrepancies between the observed full-field response measured experimentally using DIC and that computed numerically using the model. To that end, an objective function is first computed as the sum of residuals by mapping both responses onto a common grid, which is then pushed to a minimum via the method of moving asymptotes (MMA) as the optimization algorithm. This study demonstrates that the proposed approach can identify unseen damage with an average accuracy (ACC) score of 96.80% on the defined configurations, with relatively minimal false identifications.

Identification of Location and Geometry of Invisible Internal Defects in Structures using Deep Learning and Surface Deformation Field

“…The strain tensors acquired from the DIC measurements are susceptible to noise originating from lighting fluctuations, glare, irregularities, poor‐quality speckle patterns, sensor noise, and quantization. [ 33–35 ] This noise can be observed in the strain maps presented in the first row of Figure 2, which depict exclusively the noise distribution when the corresponding external load is zero. As several factors (as mentioned earlier) cause noise in DIC measurements, noise elimination remains challenging.…”

“…As several factors (as mentioned earlier) cause noise in DIC measurements, noise elimination remains challenging. [ 35 ] However, we are more interested in deriving the strain tensors than capturing the strain tensor images because acquiring a minimum amount of true strain data that responds to the underlying subsurface defects can significantly assist the OD algorithms in identifying and characterizing invisible subsurface defects of various shapes and depths at varying load limits.…”

“…Currently, DIC is being applied in topology optimization frameworks to detect invisible internal defects. [33][34][35] However, this approach requires a finely tuned finite-element model (FEM) of the structure, which can be computationally expensive. Moreover, the surface deformation measured by the DIC is prone to noise originating from lighting fluctuations, glare, irregularities, poor-quality speckle patterns, sensor noise, and quantization; therefore, solving the inverse problem for small deformations or for abnormalities located deep in the sample becomes very challenging.…”

“…Moreover, the surface deformation measured by the DIC is prone to noise originating from lighting fluctuations, glare, irregularities, poor-quality speckle patterns, sensor noise, and quantization; therefore, solving the inverse problem for small deformations or for abnormalities located deep in the sample becomes very challenging. [33][34][35] Additionally, plastic deformation around the subsurface defect can affect the accuracy of measurements.…”

Performance assessment of steel cantilever beams based on connection behaviour using DIC technique and improved hybrid PSO algorithm

“…DIC has a better performance compared to Linear Variable Differential Transformer (LVDT) sensors and strain gauges that can extract only values at a single point. The DIC method has gained significant popularity in recent years, and various studies have been performed using this technique, [32][33][34][35][36][37] as it can update the FEM in a more efficient way. The DIC method obtains the amount of displacement and strain of the desired surface by comparing the images taken from the specimen, before and after the deformation.…”