Fatigue Delamination Growth In Composites Research Articles

Multiple local particle filter for high-dimensional system identification

Nonlinearity and high dimensionality emerge as two primary challenges in the realm of system identification within the context of structural health monitoring (SHM) applications. Particle filter (PF) has been demonstrated efficient for nonlinear identification, but it suffers from the curse of dimensionality and behaves poorly in high-dimensional problems. The idea of state and measurement partitioning has been used in many PF algorithms to simplify high-dimensional identification problems into the identification of several lower-dimensional subgroups, but with very few applications to SHM problems. In this context, by combining multiple particle filters (MPF) with the decay of correlations property, this paper develops a novel multiple local particle filter (MLPF) for high-dimensional identification problems. A whole state vector is partitioned into several state subgroups, each consisting of fewer state components and then estimated by one PF through a novel likelihood including the local state and measurement vectors. The feasibility and efficiency of the proposed method are tested through a benchmark toy example, a case study of a twenty-story Bouc-Wen frame structure under ground motion, and an experimental study of fatigue delamination growth in composites.

An XFEM-VCCT coupled approach for modeling mode I fatigue delamination in composite laminates under high cycle loading

In this study, virtual crack closure technique (VCCT) and extended finite element method (XFEM) are coupled to each other as XFEM-VCCT approach to simulate mode I fatigue delamination growth in composites, employing the direct cyclic method in Abaqus. Both two-dimensional plane strain and three-dimensional finite element models under force and displacement control are considered. Numerical simulation results are compared with the existed experimental test data for double cantilever beam (DCB) specimens and validated. Finally, challenges ahead of VCCT and XFEM-VCCT are discussed in detail and the appropriate method for modeling fatigue delamination growth in laminated composites under high cycle loading is suggested. It is found that simulation of the DCB fatigue delamination via the displacement control loading leads to more accurate results in comparison to the force control. VCCT was found as a suitable method for simulation of fatigue delamination growth in 2D and 3D-shell models. While XFEM-VCCT shows high accuracy and low computational time in 3D-solid finite element models. The key conclusion is that the XFEM-VCCT coupled approach is independent of time increment, whereas the time increment is more effective on the results of VCCT analysis, and it affects the run-time significantly.

Finite element modeling of mode I fatigue delamination growth in composites under large-scale fiber bridging

In this study, the Turon et al. and Kawashita-Hallett fatigue damage models are extended with a trilinear cohesive law to simulate mode I fatigue delamination in the composites undergoing large-scale fiber bridging. The trilinear cohesive zone model (CZM) is constructed by superposition of two bilinear CZMs. In order to implement the developed models, a user defined element (UEL) subroutine is prepared in the Abaqus software and finite element analyses are conducted based on the envelope loading technique to simulate the 3D double cantilever beam (DCB) specimens under high-cycle fatigue loading. The finite element models are assessed through the experimental data from the literature and also five available fatigue damage models based on the bilinear CZM. It is concluded that the trilinear CZMs has more accuracy than the bilinear ones in prediction of fatigue delamination with fiber bridging effects. A parametric study is performed on the two extended models and sensitivity of models to the fitting parameters and quasi-static CZM parameters are comprehensively investigated.

Modelling fatigue delamination growth in fibre-reinforced composites: Power-law equations or artificial neural networks?

This paper discusses two alternative modelling approaches for describing fatigue delamination growth (FDG) in polymer-based fibre-reinforced composites, i.e. semi-empirical equations having a power-law form and artificial neural networks. Barenblatt's self-similarity principles are applied for identifying a suitable expression of the delamination driving force in terms of the square-rooted energy-release-rate range and the associated peak values. The general dependency of pre-factors and exponents of FDG power-laws on the stress-ratio and mode-mixity is discussed in detail. Single-hidden-layer neural networks (SHLNN) with the support of self-similarity principles are here proposed as an alternative to semi-empirical power laws for describing FDG in composites. A example application of SHLNN to mixed-mode and variable stress-ratio FDG is provided for the carbon/epoxy system T800H/#3631. The SHLNN predictions are compared to a semi-empirical fit based on a modified Hartman-Schijve power-law.

Thickness effects on fibre-bridged fatigue delamination growth in composites

This paper provides an investigation on thickness effects on fibre-bridged fatigue delamination growth (FDG) in composite laminates. A modified Paris relation was employed to interpret experimental fatigue data. The results clearly demonstrated that both thickness and fibre bridging had negligible effects on FDG behaviors. Both energy principles and fractography analysis were subsequently performed to explore the physical reasons of this independence. It was found that the amount of energy release of a given crack growth was not only independent of fibre bridging, but also thickness. Fibre print was the dominant microscopic feature located on fracture surfaces, physically making the same energy dissipation during FDG. Furthermore, the present study provides extra evidence on the importance of using an appropriate similitude parameter in FDG studies. Particularly, the strain energy release rate (SERR) range applied around crack front was demonstrated as an appropriate similitude parameter for fibre-bridged FDG study.

Effects of large scale bridging in load controlled fatigue delamination of unidirectional carbon-epoxy specimens

Fatigue delamination growth in composites is accompanied by large scale bridging (LSB) that yields important toughening effects. However, the extent of this mechanism depends on the laminate geometry rendering its modeling a challenging task. This work presents a combined experimental/numerical study on characterization of specimen thickness dependence of LSB in fatigue delamination. Double cantilever beam specimens of different thicknesses (h = 2, 4 and 8 mm), equipped with arrays of multiplexed fiber Bragg grating sensors, are subjected to mode I fatigue loads. Measured strain data with the sensors are employed to identify the bridging tractions and subsequently compute the energy release rate (ERR) due to the bridging as well as the ERR at the crack tip. The obtained results confirm that fatigue delamination growth strongly depends on the specimen geometry when LSB prevails. It is shown that both the extent of bridging and critical ERR at failure increase by increasing the specimen thickness while the maximum bridging traction at the crack tip is found independent of the specimen geometry. The identified traction-separation relations serve to establish a power correlation, between the crack growth rate and ERR at the crack tip which is independent of the specimen thickness.

Fibre bridging effect on the Paris relation for mode I fatigue delamination growth in composites with consideration of interface configuration

Fibre bridging can significantly enhance delamination resistance making the use of a single Paris resistance curve to determine fatigue crack growth insufficient. An empirical Paris-type relation has been developed in a previous study to take fibre bridging into account in fatigue delamination growth. This relation was developed by correlating the Paris constants C and n to the amount of fibre bridging. This paper provides a further investigation on the interface configuration effect on fatigue delamination growth, illustrating the significance of fibre bridging. The results demonstrated that more bridging fibres can be generated in a multidirectional interface, making both log(C) and n significantly depend on fibre bridging. Thus, the method proposed in the previous study is further extended to take into account of the interface configuration effect.

The effect of fibre bridging on the Paris relation for mode I fatigue delamination growth in composites

Fibre bridging plays an important role in fatigue delamination growth in composite materials. It decreases the crack growth rate significantly with the increase of delamination length. In this paper, it is demonstrated how the Paris relation depends on the amount of fibre bridging. For this purpose, both parameters n and C in the Paris relation, are described as function of the crack extension a−a0, as a measure to quantify the contribution of fibre bridging. It appears that the parameter n only depends on the stress ratio and remains independent on the amount of fibre bridging. The parameter C depends on both the stress ratio and crack extension. The obvious difference in trends between both stress ratios indicates that a method incorporating the contribution of fibre bridging in the Paris relation requires extensive data sets.

Methods for the prediction of fatigue delamination growth in composites and adhesive bonds – A critical review

An overview is given of the development of methods for the prediction of fatigue driven delamination growth over the past 40years. Four categories of methods are identified: stress/strain-based models, fracture mechanics based models, cohesive-zone models, and models using the extended finite element method. It is highlighted that most models are phenomenological, based on the observed macro-scale behaviour of test specimens. It is suggested that a more physics based approach, focusing on elucidating the mechanisms involved, is needed to come to a full understanding of the problem of delamination growth.

Application of the Hartman–Schijve equation to represent Mode I and Mode II fatigue delamination growth in composites

This paper discusses the potential of a variant of the Hartman–Schijve equation to represent both Modes I and II constant amplitude delamination growth in composites. To this end we show that the delamination growth rate da/ dN can often be related to (Δ√ G − Δ√ G th) α , where α is approximately 2. As such the exponent in this relationship is considerably lower than the exponent in “Paris like” power law representations. We also show that this particular representation of delamination growth in composites is similar to that seen for crack growth in metals.

Simulation of Fatigue Delamination Growth in Composites with Different Mode Mixtures

A numerical model, obtained by implementing a cyclic damage model in the framework of an interface element, is here proposed to reproduce the crack growth in laminated composites subjected to cyclically repeated loads. This model, which differs from the few studies in the literature on the topic, applies not only to single fracture modes but also to mixed modes, and to constant or variable crack growth rates. The applied load (in terms of force or displacement) is assumed to oscillate between a minimum and a maximum constant value. The Paris curve can be reproduced with accuracy once some parameters in the numerical model are tuned with experiments. These parameters are preliminarily found by simulating fatigue delamination in mode I (DCB test), in mode II (ELS or ENF test) and with a fixed mode mixture φ = GII/G (MMB test). A non-monotonic curve is then used to interpolate these preliminary results. Furthermore, tests where the mode mixture changes with crack length could also be reproduced. With this model it is possible to predict the crack growth rate with cycle of a generic structure without knowing a priori how the mode mixture φ changes during the crack propagation. This novel procedure gives a new opportunity in the design of composite structures subjected to repeatedly applied loads.