Enhancing the performance and lightness of different structures has already been achieved by the employment of fibre reinforced composite materials. Nowadays, a new challenging perspective is being given to these materials by the inclusion of non-metallic conductive components. This emerging technology will lead to multifunctional composites with possible applications in structural health monitoring and traffic monitoring. The aim is to avoid corrosion problems from metallic components, as well as to eliminate the need of expensive equipments used for the health monitoring of large infrastructures. In the present research, the strain-sensing capability of a core-reinforced hybrid carbon fibre/glass fibre braided composite has been investigated in order to develop continuous monitoring system. The characterization of sensing behaviour was performed with the help of an instrumental set-up capable of measuring the change in electrical resistance with mechanical stresses applied to the samples. The effect of core composition (carbon fibre/glass fibre weight ratio) on the strain sensitivity of the braided composites has been studied in order to find out the optimum composition for best sensing capability. Among the three compositions studied (23/77, 47/53 and 100/0), composites with lowest amount of carbon fibre showed the best strain sensitivity with gauge factors up to 23.4 at very low flexural strain (0.55%). Attempts have also been made in this research to develop a piezoresistive matrix for the braided composites in order to further enhance their strain sensitivity. For this purpose, the strain sensing capability of an unsaturated polyester matrix dispersed with chopped carbon fibres (1mm and 3 mm lengths) at various weight % (0.5, 0.75 and 1.25%) was evaluated in order to find out their optimum length and concentration. It was observed that chopped fibres with different lengths showed similar strain sensitivity, which however, improves with the decrease in their concentrations.
Connecting concrete technology and machine learning: proposal for application of ANNs and CNT/concrete composites in structural health monitoring
