This work represents the first step towards the application of machine learning techniques in the prediction of statistical design allowables of composite laminates. Building on data generated analytically, four machine algorithms (XGBoost, Random Forests, Gaussian Processes and Artificial Neural Networks) are used to predict the notched strength of composite laminates and their statistical distribution, associated to the uncertainty related to the material properties and geometrical features. This work focuses not only on the so-called Legacy Quad Laminates (0°/90°/±45°), typically used in the design of composite aerostructures, but also on the newer concept of double-double (or double-angle ply) laminates. Very good representations of the design space, translating in low generalization relative errors of around ±10%, and very accurate representations of the distributions of notched strengths around single design points and corresponding B-basis allowables are obtained. All machine learning algorithms, with the exception of the Random Forests, show very good performances, with Gaussian Processes outperforming the others for very small number of data points while Artificial Neural Networks have better performance for larger training sets. This work serves as basis for the prediction of first-ply failure, ultimate strength and failure mode of composite specimens based on non-linear finite element simulations, providing further reduction of the computational time required to virtually obtain the design allowables for composite laminates.

@en

A micromechanics framework for modelling the longitudinal failure of unidirectional fibre composites, in both tension and compression, is presented. Appropriate constitutive models are used for the different constituents (i.e., fibre and matrix) and interfaces. An algorithm for the generation of the fibre misalignment is reported. It combines a stochastic process with an optimization procedure in order to match the numerical and experimental misalignment angle distributions. Two alternative strategies could be used for the optimization procedure, based on the minimization of the standard error of the likelihood or probability, respectively, whose pros and cons are discussed. Finally, micromechanical modelling in tension and compression are reported, and the influence of the fibre misalignment angle distribution is assessed.

@en

The effect of through-thickness compressive stress on mode II interlaminar fracture toughness is investigated experimentally and replicated numerically. The modified Transverse Crack Tensile specimen recently proposed by the authors is used, together with an experimental device designed to apply a constant transverse compressive stress on the surface of the specimen. Experiments are conducted using IM7/8552 specimens for different compressive stresses, ranging from 0 to 100 MPa, covering all the practical applications commonly encountered in the aeronautical industry (e.g., tightened filled holes or bolted joints). It is shown that mode II interlaminar fracture toughness increases with the applied compressive through-thickness stress. Finally, experiments are replicated using appropriate numerical models based on cohesive elements that take into account frictional effects. A good agreement between numerical predictions and experiments is found.@en

The effect of through-thethickness compressive stress on the mode II interlaminar fracture toughness is investigated experimentally and replicated numerically. It is shown that the mode II interlaminar fracture toughness increases with the applied compressive through-the-thickness stress. A good agreement between numerical predictions and experiments is found. @en

Several problems arise when measuring the mode II interlaminar fracture toughness using a Transverse Crack Tension specimen; in particular, the fracture toughness depends on the geometry of the specimen and cannot be considered a material parameter. A preliminary experimental campaign was conducted on TCTs of different sizes but no fracture toughness was measured because the TCTs failed in an unacceptable way, invalidating the tests. A comprehensive numerical and experimental investigation is conducted to identify the main causes of this behaviour and a modification of the geometry of the specimen is proposed. It is believed that the obtained results represent a significant contribution in the understanding of the TCT test as a mode II characterization procedure and, at the same time, provide new guidelines to characterize the mode II crack propagation under tensile loads.@en