| Abstract: | The objective of the study is to utilise a material's inherent electrical conductivity as means of damage quantification and damage location detection. After determining the percolation threshold for a carbon nanotube (CNT)‐epoxy mixture, an optimum concentration was chosen to infuse it into glass‐fabric reinforced panels to make them electrically conductive. Two different multiwalled CNT‐epoxy composites were manufactured for this study: CNT enhanced epoxy resin and glass‐fabric reinforced CNT epoxy resin. Epoxy resin‐based glass‐fabric reinforced composite panels enhanced with carbon nanotubes were manufactured with embedded electrodes and then subjected to damages. Rectangular panels of various proportions were studied. Disks made out of copper foil were affixed to surfaces of CNT epoxy panel, whereas in glass‐fabric CNT epoxy specimen, total of 64 electrodes (grid of 8 × 8) were embedded inside the composite panel under the top layer of the 10‐ply fabric. The disks acted as electrodes, enabling voltage measurements using in‐line 4‐probe technique, which minimises contact resistance between the electrodes and the material. Two different configurations of electrode network were employed to scan voltage change in the entire composite panel. The networks included evenly spaced (3 in. for inner ones) electrodes that spanned the surface of the panel. To further investigate influence of electrodes distribution, finite element simulations were used to solve the electrical potential distribution in the panel for various damage sizes and location. Predamage and postdamage voltage field was used as gauge in sensing the damage and its extent for quantification. The finite element method simulation results matched the experimental data closely. The results indicate that there is a consistent behaviour that can be correlated to the size and location of the damage. As spacing between electrodes is increased, they become less responsive to smaller damages. Forty‐eight electrodes (out of 64) were actively used and were enough to confirm that the method can be used as an alternative to electrical tomography method where fewer (boundary) electrodes per area are employed but at a higher cost of computational cost. One important aspect of this study with embedded and distributed electrodes is the fact that the method can be applied to larger panels increasing its utility in practical applications. |
