The objective of this study was to determine the deflected wing shape and the out-of-plane loads of a large-scale carbon-composite wing of an ultralight aerial vehicle using Fiber Bragg Grating (FBG) technology. The composite wing was instrumented with an optical fiber on its top and bottom surfaces positioned over the main spar, resulting in approximately 780 strain sensors bonded to the wings. The strain data from the FBGs was compared to that obtained from four conventional strain gages, and was used to obtain the out-of-plane loads as well as the wing shape at various load levels using NASA-developed real-time load and displacement algorithms. The composite wing measured 5.5 meters and was fabricated from laminated carbon uniaxial and biaxial prepreg fabric with varying laminate ply patterns and wall thickness dimensions. A three-tier whiffletree system was used to load the wing in a manner consistent with an inlight loading condition.

This paper presents a semianalytical method for the prediction of interlaminar stresses and displacements near the free edges and ply cracks in general angle-ply laminates subjected to biaxial extensions and/or in plane shear deformation. The method is based on a state space representation of the three-dimensional equations of elasticity. Numerical solutions are obtained by using layer refinement in the through thickness direction and Fourier series expansion in the other directions. By this approach, an angle-ply laminate may be composed of an arbitrary number of monoclinic layers and each layer may have different material property and thickness. This method guarantees continuous fields of all interlaminar stresses across interfaces between material layers. Numerical results are compared with those obtained from other methods. It is found that the theory provides a satisfactory approximation to the stress singularities near the free edges and ply cracks. Numerical solutions for antisymmetric laminates under extension and general laminates under shearing are new in the literature and can be used as benchmarks for validating new models.