Uncertainty Quantification in the Behaviour of Lap Joints in FRP Laminates

Abstract

Adhesive bonding and mechanical fastening have been continuously in use to fabricate fibre-reinforced plastic (FRP) laminated composite structures. The use of adhesive bonding has its roots in aircraft structures for more than 50 years, where durable and lightweight bonds are necessary. On the other hand, bolted connections/joints use nuts, bolts, and washers to join two adherends together, making it easier to install and disassemble the connections. The performance of joints can be affected by several uncertain preconditions, such as uncontrolled manufacturing, test procedures, material properties, loading, and environmental conditions. There could be uncertainty in the performance of the joints even if identical parameters are followed to fabricate the joints. Such variability in the performance of the joints must be assessed for the probabilistic analysis of various composite structures. In the present study, experimental investigations are carried out to study the performance and behaviour of single-lap joints in both connection categories, i.e. adhesive and bolted, between glass fibre-reinforced plastic (GFRP) adherends. The shear strength of the single-lap joints (SLJs) in GFRP adherends is evaluated under tensile loading in the longitudinal direction. Carbon nanotubes (CNTs) are incorporated into the adhesive matrix to attempt the shear strength enhancement in the SLJs. Uncertainty in the shear strength is quantified and modelled using Goodness-of-Fit (GOF) tests, such as Kolmogorov-Smirnov (K-S) and Anderson-Darling (A-D), with statistical distributions. SLJs are also investigated to predict their performances numerically using finite element methods. An acceptable agreement between the experimental and numerical performance is observed. The peel stress in the adhesive is obtained from finite element simulation. A sensitivity analysis is carried out to determine the material property that influences the peel stress the most. Uncertainty in the peel stress due to the induced uncertainty in the material properties is modelled using GOF tests. Similarly, single-shear joints in GFRP adherends with a single bolt are investigated experimentally. The uncertainty in the ultimate load capacity of the bolted joints under tensile loading is quantified using GOF tests. Numerical simulation is also carried out for the joints to obtain the performances such as the joint stress, stiffness and maximum out-of-plane displacement. Sensitivity analysis is performed to obtain the most affecting parameter, among the material properties and bolt tightening torque. The GOF tests are used with statistical distributions to quantify the uncertainty in the load capacity caused by the uncertainty induced in the sensitive material properties. The use of sandwich composite, in the recent past, has gained attention for their investigation in aerospace industries, as well as other fields such as marine and automotive. In the present study, the dynamic performances, such as fundamental natural frequencies and mode shapes, of the sandwich composite plate made of GFRP face sheets and PVC (polyvinyl chloride) core are evaluated through free vibration analysis. Sensitivity analysis is carried out to identify the most influential material property to the natural frequency of the sandwich plate. The uncertainty in the natural frequency of the sandwich plate is modelled using GOF tests. An elite statistical distribution, obtained by GOF tests, is recommended to describe or model the uncertainty. The performance of the recommended distribution is studied using various probabilistic plots, such as probability density function, cumulative distribution function, survival, hazard, P-P and probability difference plots. The outcome and knowledge of the present study will enlighten the industry to control the manufacturing procedure of the connections/joints in FRP structures, as well as the sandwich composite. Furthermore, the present study will form a basis for the probabilistic analysis of FRP and sandwich structures.