No abstract

This study addresses the issue of structural damage identification and location in carbon fiber reinforced polymer plates using electrical measurements. Electrical resistance tomography is presented as a method for structural damage localization in composite parts. A set of electrodes is fixed on the edges of the part and combinations of DC current injections and voltage measurements are applied to the system. The change of voltage between different times in the part’s service life (e.g. start and degraded) are monitored. These sets of measurements are used as input to inversely calculate conductivity maps for the complete composite part and thus indirectly assess its structural health. Such processes are inherently ill-posed. Data post-processing approaches are proposed here to diminish this uncertainty and to conclude to an optimally converge solution of the inverse problem. To assist the process, a material-originating mathematical constraint is introduced. The method is applied on carbon fiber reinforced polymer plates for different damage modes. Experimental recordings show that the analysis of electrical fields allows detecting the presence of damage. Discontinuities as small as 0.1% of the inspected area can be sensed. The proposed data post-processing techniques were applied and conductivity maps were calculated. The results show that using these techniques locating damage is possible with less than 10% error. Material-based constraints greatly enhance the prediction of the data post-processing techniques. It is believed that by overcoming certain implementation issues, electrical resistance tomography could evolve in the direction of a non-destructive evaluation or a structural health monitoring technique for composite structures.

We show that the replication process can be extended towards the production of functionally graded porous structures by fabricating and testing structures in which outer layers of dense metal encase a central part made of foam with graded porosity. Samples of this kind are produced by pressing individual layers of NaCl powder of granulometry 60-90 m, and then stacking these layers between two skins of dense aluminium. The stacked preforms are then infiltrated with pure aluminium and solidified before dissolution of the salt in water. Specimens containing up to five layers of porous Al of different density between two dense outer skins of pure Al are produced; selected samples are tested in three-point bending. Data show good agreement with analysis based on sandwich beam theory and the Deshpande-Fleck yield criterion. Results of this work indicate that whereas lightweight graded metal/metal foam beams show little promise from the standpoint of stiffness-limited design, they may be of interest from the standpoint of load-limited design.

An approach for damage inspection of composite structures utilizing carbon nanotubes (CNT) networks is investigated. CNT are dispersed in an epoxy using a processing technique compatible with commonly employed composite manufacturing techniques and subsequently used as matrix for a structural glass fiber reinforced composite. The developed electrical conductivity of the composite system is verified experimentally. The electrically conductive CNT network within the GFRP is exploited through distributed electrical voltage measurements to sense and, ultimately, locate damage in the plane of the composite plate. Damage in the form of cracks or delamination interrupts the continuity of the CNT network separating and isolating regions of the conductive network. Employing electric potential fields these changes can become measurable and can provide information for inversely locating the damage. Electrical Resistance Tomography (ERT) is formulated and experimentally applied to measure changes in the potential fields and deliver electrical conductivity change maps which are used to identify and locate changes in the CNT networks. These changes are correlated to capture the damage in the composite. Different damage modes are studied to assess the capabilities of the technique. The technique shows sensitivity to very small damages; less than 0.1% of the inspected area. The solution of the inverse ERT problem delivers a conductivity change maps which offers an effective localization with nearly 10% error and an inspection area suppression of around 75%. The proposed methodology to create CNT networks enables the application of ERT for Non-Destructive Evaluation of composite materials, previously not possible due to lack of conductivity, thus offering damage sensing and location capabilities even in-situ.

No abstract