Skip to main content



Reporting period: 2017-06-01 to 2019-05-31

The design of composites leads to challenging tasks since those competencies that stemmed from the adoption of metallic materials are often inadequate for composites. Insights on many different disciplines and tight academic/industrial cooperation are required to fully exploit composite structure capabilities.

The main objectives of FULLCOMP are the following:
• Development of advanced analytical and numerical models to design composite structures and improvement of the academic-to-industry transferability of such advanced tools.
• Improvement of the existing design and optimization tools to fully exploit the enhanced capabilities of the composite structure for industrial applications.
• Enhancement in the stat-of-the-art of the failure and damage analysis of composites by introducing novel physics-based component-wise approach that provides high accuracy with very low computational costs.
• Development of multidisciplinary approaches for health-monitoring, non-destructive testing (NDT), and repairing.
• Enhancement of multiscale models by introducing new modeling approaches that require lower computational costs and that can be easily coupled with testing and monitoring techniques that are currently used in real applications.
• The exploitation of existing industrial experimental and manufacturing approaches to tune the academic research activities and to provide validation tools.
• Provision of an entrepreneurial background to Ph.D. students to enhance their career perspectives and improve their skill transferability.
• Pursuing a comprehensive dissemination campaign with attention paid to the publication in top journals, organization of special issues in international journals and organization of international conferences.
• Provision of training activities that deal with all the main tasks that are involved in composite structure design and utilization; including, manufacturing, health-monitoring, failure, modeling, multiscale approaches, testing, prognosis, and prognostic.
The main results and their exploitation are the following:
• Development of an adaptable refinement approach based on node-dependent Kinematics (NDK) and hierarchical elements are accounting for fully coupled electromechanical effects.
• Introduction of the Hierarchical Legendre Expansion (HLE) to handle complex architectures, e.g. 3D woven composites and metamaterials., at the macro-, meso-, and microscale via 1D models.
• Analysis of stress concentrations at the free edges of laminated structures and lap joints via 1D refined models and simulation of Lamb waves for structural health monitoring purposes.
• Development of a nonlinear, concurrent multiscale scheme based on 1D models and providing 3D-like accuracy with four times less computational time than 3D solid models.
• Implementation of a contact formulation for higher-order one-dimensional elements to simulate the impact and cohesive elements for interlaminar damage and failure.
• Development of interfaces for the in-house codes with commercial codes such as ABAQUS for pre- and post-processing operations and integration of micromechanics capabilities with the NASA NASMAT code.
• Development of a global-local approach that includes skin-stringer debonding.
• Implementation of a special homogenization procedure that preserves dissipated energy between the local and the global level.
• Analysis of the non-linear response of structures with varying material properties to develop a mapping of the sensitivity of these structures to variations to improve the structural performance.
• Multi-scale structural modeling by bridging micromechanics and the advanced CUF one-dimensional/beam structural theories using the Multilevel Finite Element (also known as FE2) framework
• Investigation of the effect of microscale imperfections (not straight carbon fibers) on the macroscale response (instability).
• Implementation of Green-Lagrange geometric non-linearities within a 1D framework of a total Lagrangian approach, including shear and membrane locking correction strategy via the Mixed Interpolation of Tensorial Components (MITC).
• Analysis of pre-buckled bistable composite beam structures for different slenderness ratios, material properties, and boundary conditions.
• Extension of the existing Carrera Unified Formulation framework to perform non-linear and large displacement analyses of non-prismatic, 3D-like, beam structures
• Development of a methodology to analyze the effect of localized low stiffness regions in the buckling and post-buckling behavior of thin structures
• Development of a new, non-iterative and intuitive methodology for designing topologically efficient buckling resistant structures
• Characterization of the full viscoelastic set of the composite material properties and identification of the uncertainty of the material properties, at each pertinent scale.
• Implementation of a hyperelastic constitutive law in a commercial Finite Element code and validation for the case of a yarn subjected to transverse impact
• Analysis of the effect of stochastic yarn distortion over the margin of its cross-section and stochastic yarn waviness over the yarn’s length, to the stiffness and strength of the macroscale textile layer properties.
• Uncertainty quantification of the strength properties of braided composites
• Integration of probabilistic methods within the framework of an inverse optimization-based approach for the characterization of material properties in lower scales.

The dissemination comprehended
• Four FULLCOMP special sessions in international conferences.
• One book with the FULLCOMP proceedings of the special session at the AIDAA 2015 conference.
• One book for Springer on the main results of the project with chapters authored by each ESR
• 65 journal papers.
• 83 conference presentations.
On average, each ESR had more than two secondments. Twenty-one journal papers, 12 conference papers, and six book chapters stemmed from secondment activities. FULLCOMP ESRs attended hard and soft skill courses, both network-wide or organized by each hosting institution. The network-wide training events are the following
• Four workshops
• One spring school
• One research retreat with external experts
• Two soft-skill courses
• Seven hard-skill courses or seminars

The presence of OTC partners provided ESR with the opportunity of increasing their competences. ESR participated in the following events with OTC partners:
• Two secondments.
• Five courses or seminars held by OTC scientists.

The main progress beyond the state-of-the-art is given by
• Reduction of computational times for the nonlinear analysis of composites. Such a reduction ranges from four to ten times as standard 2D and 3D FE.
• Prediction of the complete 3D stress field, including transverse stresses and more accurate evaluation of failure.
• Capabilities to handle complex macro, meso, and micro-architectures with low computational costs.
• Global-local approaches with node-dependent kinematics and two-way capabilities to optimize the computational efforts without accuracy penalties.
• Multiscale frameworks for multiple purposes, including nonlinear analyses and characterization.
• Design of bistable and variable angle tow laminates via 1D models.
• Development of hyperelastic laws for impact problems.
• Analysis of the effect of imperfections at various scales and of different nature, e.g. manufacturing.
Modeling strategies via 1D models
Example of damage via 1D models
Shear stress in composite shells via refined models
Variable kinematics structural model