Identification of embedded interlaminar flaw using inverse analysis

Abstract

The integrity of a composite laminate can be greatly affected by an existence of embedded interlaminar flaw. In general, identification of such a flaw often requires expensive tools and tedious processes. The aim of the present work is to develop a novel method with the aid of an intelligent post-processing scheme, thereby not relying on those sophisticated experiments. Essentially the proposed procedure utilizes an inverse analysis to estimate unknown delamination parameters from limited measurements. The procedure first constructs approximate functions relating the delamination parameters to measurement parameters. Then, a multi-dimensional minimization technique is adopted to search for the best estimates of unknown parameters corresponding to the lowest value of error objective function. In the present verification and simulation analyses, surface strains at discrete locations on a composite laminate under three-point bending are selected as the input measurements. Although reasonable estimates are obtained with these measurements, to increase their accuracy, the deflection at load point is also included as measurement input. Additional improvements are observed when those measurements under multiple loading conditions are included. A detailed error sensitivity analysis is also carried out to confirm the method's robustness. These results suggest the current method to be one of the alternate identification approaches for detecting a single embedded delamination in composite laminates.

Damage detection of CFRP laminates via self-sensing fibres and thermal-sprayed electrodes

“…More detailed description can be found in a similar procedure carried out in Ref. [25]. The outline of the inverse analysis to estimate the unknown parameters is shown in Figure 3.…”

Assessment of Impact Damage in Composites via Self-Sensing Fibers

“…This method is known as a one of the most popular methods for multi-dimensional optimization especially in case of derivatives of objective function are either unavailable or discontinuous. It was introduced by Nelder and Mead [2] and previously utilized to find the embedded delamination in composite [3]. The method is based on the concept of a simplex, which is a polytope of N+1 vertices in N dimensions.…”

Modelling and identification of delamination in double-layer beams by the virtual distortion method

“…Delamination zones close to specimen boundaries as well as internal ones were successfully detected. Ramanujam et al [11] apply classical optimization to identify delamination via inverse analysis. Their numerical considerations included a simply supported 4-ply laminate beam with a single embedded delamination.…”