Design exploration methodology for ultra thick laminated composites

Abstract

Existing test and analytical methods (theoretical and numerical) are normally restricted to thin laminate components, which cannot accurately represent the 3D stress state behaviour of the so called Ultra Thick Laminates (UTL) structures. Thus, it is necessary to expand the scope of application of the current numerical methods to accurately predict the out-of-plane delamination failure associated with these types of structures (mainly due to the transverse shear stresses and interlaminar stresses). The overall objective of this work is to address the following research objectives: • To assess the functionality, advantages and limitations of different solid element formulations, including layered solid elements that are available in commercial Finite Element codes, applied to the mechanical response prediction of UTL composite components (thicknesses up to 30 mm are considered). • To perform a design exploration and optimisation of constant thickness UTL composite component in terms of the orientation of a varying and repeatable stacking sequence of an eight ply Non-Crimped Fabric, in order to assess the design implications on performance. In order to achieve the above stated objectives a standard, flexible and expandable FE based design exploration methodology (at a ply level) for UTL composite components is proposed, which considers a commercial FE tool (ANSYS), and a data management system and optimisation tool (ISIGHT), through the use of layered solid elements (SOLID186 and SOLID191, 20-node layered solid elements). Application of manufacturing design rules (for reducing the number of feasible stacking sequences to be evaluated) is also considered, in order to reduce the computational cost of such a study, as well as to present a practical solution from the manufacturing point of view. Initially, in-plane and out-of-plane capabilities of various layered element formulations and modelling strategies where evaluated for thin and thick laminate applications against known analytic solutions (CLT, etc), in order to understand the key parameters and the accuracy limitations of each formulation. This led to practical recommendations for pre and post processing of thick laminate FE models, such as for the number of layered solid elements required as a function of the thickness of the UTL component to effectively predict the magnitude and variation in transverse shear stress across the thickness. The application of this research was demonstrated on the design exploration and performance optimisation of a UTL composite specimen (with constant thickness) under a 3-point bending test (linear static analysis), for which experimental results were available. The individual ply orientations are the design variables considered, and the performance was assessed through the vertical displacement of the component and the maximum transverse shear stress value. This exploration of the design space did identify other possible configurations that may have a better performance than the baseline (Biax), considering only the maximum transverse shear stress values as directly responsible for the delamination failure. However, these improved designs may present a higher number of plies failed or a higher failure index (Tsai-Wu failure criteria). Further experimental studies are required to further explore the design space, but this work represents the starting point and possible approaches for development of robustness and weight optimisation of UTL composites are proposed.