Search results

The purpose of this paper is the development of a multiscale model which simulates the effect of the dispersion, the waviness, the interphase geometry as well as the agglomerations of multi-walled carbon nanotubes (MWCNTs) on the Young’s modulus of a polymer filled with 0.4 Vol.% MWCNTs.

For the determination of the homogenized elastic properties of the hybrid material representative unit cells (RUCs) have been used. The predicted homogenized elastic properties were used for the prediction of the Young’s modulus of the filled material by simulating a finite element (FE) model of a tensile specimen. Moreover, the model has been validated by comparing the predicted values of the numerical analysis with experimental tensile results.

As the MWCNT agglomerates increase, the results showed a remarkable decrease of the Young’s modulus regarding the polymer filled with aligned MWCNTs while only slight differences on the Young’s modulus have been found in the case of randomly oriented MWCNTs. This might be attributed to the low concentration of the MWCNTs (0.4 Vol.%) into the polymer. For low MWCNTs concentrations, the interphase seems to have negligible effect on the Young’s modulus. Furthermore, as the MWCNTs waviness increases, a remarkable decrease of the Young’s modulus of the polymer filled with aligned MWCNTs is observed. In the case that MWCNTs are randomly dispersed into the polymer, both numerical and experimental results have been found to be consistent regarding the Young’s modulus.

The methodology used can be adopted by any system containing nanofillers.

Although several studies on the effect of the MWCNTs distribution on the Young’s modulus have been conducted, limited results exist by using a more realistic RUC including a periodic geometry of more than 20 MWCNTs with random orientation and a more realistic waviness of MWCNTs with aspect ratio exceeding 150.

The purpose of this study is the investigation of self-healing materials containing encapsulated healing agents embedded in a polymer matrix with dispersed catalysts. In recent years, the high performance and design flexibility of composite materials have led to their widespread use in the aeronautics, space, automotive and marine fields. Simultaneously, as the need for advanced material properties has increased, many studies have been conducted on multifunctional materials, focusing on different fields of their desired capabilities.

A multiscale model was developed to simulate the effect of microcapsules on the mechanical behavior of the polymer matrix. Furthermore, the effects of microcapsule diameter and microcapsule concentration on the mechanical behavior of the composite were studied. Digimat and Ansys software were used to simulate the self-healing composites.

There is a trade-off between the efficiency of the microcapsules and the degradation of the properties of the composite material.

The generated model simulated an encapsulated healing agent in a polymeric matrix.

Compression is critical loading condition for composite airframes. Compression behaviour of structures with or without damages is a weak point for composite fuselage panels. This is one of the reasons for need of continuous in-service health monitoring of composite structures. The purpose of this paper is to characterize the compression panel behaviour on the base of a developed and implemented structural health monitoring (SHM) system.

The SHM system based on fibre optic Bragg grating (FOBG) sensors and standard resistance strain gauges (SGs) was placed onto/into (embedded or bonded) three stiffened carbon fibre reinforced polymer (CFRP) fuselage panels. The FOBG sensor system was used to monitor the structural integrity of the reference, impacted, and fatigued panels under compression loading. Both barely visible impact damage and visible impact damage were created to evaluate their influence on the panel behaviour. The functionality of the SHM system was verified through mechanical testing.

Experimental data showed the presence of impact damages significantly changes the buckling modes development and deformation behaviour of the panels. Some differences between the optical and SG sensors during buckling were observed. The buckling waves and failure development were very well indicated during loading by all sensors located on the panel surface but not by the embedded sensors. Good agreement between the data from the SGs and FOBG sensors was achieved for all sensors placed on the stringers, which did not buckle. The good reliability of FOBG sensors during the fatigue and static testing up to panel failure was verified.

The paper gives information about different buckling behaviour of CFRP fuselage stiffened panels in compression. The paper gives detailed information about measured signals from different sensors based on their location on/in the panel structure for realistic loading scenario of composite aerostructures. The paper gives an integrated overview of sensors placement considering possibilities to predicate structure behaviour.

The purpose of this paper is to develop a multi-scale modeling approach for simulating the tensile behavior of corroded aluminum alloy 2024 T3.

The approach combines two FE models: a model of a three-dimensional representative unit cell representing a pit and a model of the tensile specimen. The models lie at the micro- and macro-scales, respectively. The local homogenized mechanical behavior of the corroded material is simulated for different pit configurations. Then, the behavior of the pits is introduced into different areas (elements) of the tensile specimen and final analyses are performed to simulate the mechanical behavior of the corroded material. The approach has been applied to six different exposure periods of the exfoliation corrosion test.

The numerical results show that the presence of pits and exfoliated areas reduces the yield strength of the material. The comparison of predicted elongation to fracture with the experimental of each exposure period value allows for the indirect assessment of the effect of hydrogen embrittlement.

Since the characteristics of corrosion damage evolution with exposure time are constant for the specific material, the model could be applied for the simulation of the mechanical behavior of any corroded structural part (e.g. a mechanically fastened panel) made from the aluminum 2024 T3 alloy.

The prediction of fatigue crack growth behaviour is an important part of damage tolerance analyses. Recently, the author’s work has focused on evaluating the FASTRAN retardation model. This model is implemented in the AFGROW code, which allows different retardation models to be compared. The primary advantage of the model is that all input parameters, including those for an initial plane-strain state and its transition to a plane-stress-state, are objectively measured using standard middle-crack-tension M(T) specimens. The purpose of this paper is to evaluate the ability of the FASTRAN model to predict correct retardation effects due to high loading peaks that occur during variable amplitude loading in sequences representative of an aircraft service.

This paper addresses pre-setting of the fracture toughness K _R (based on J-integral J _Q according to ASTM1820) in the FASTRAN retardation model. A set of experiments were performed using specimens made from a 7475-T7351 aluminium alloy plate. Loading sequences with peaks ordered in ascending-descending blocks were used. The effect of truncating and clipping selected load levels on crack propagation behaviour was evaluated using both experimental data and numerical analyses. The findings were supported by the results of a fractographic analysis.

Fatigue crack propagation data defined using M(T) specimens made from Al 7475-T7351 alloy indicate the difficulty of evaluating the following two events simultaneously: fatigue crack increments after application of loads with maximum amplitudes that exceeded J _Q and subcritical crack increments caused by loads at high stress intensity factors. An effect of overloading peaks with a maximum that exceeds J _Q should be assessed using a special analysis beyond the scope of the FASTRAN retardation model.

Measurements of fatigue crack growth on specimens made from 7475 T7351 aluminium alloy were carried out. The results indicated that simultaneously evaluating fatigue crack increments after application of the load amplitude above J _Q and subcritical increments caused by the loads at high stress intensity factors is difficult. Experiments demonstrated that if the fatigue crack reaches a specific length, the maximal amplitude load induces considerable crack growth retardation.

The purpose of this paper is to present, the global behaviour of sandwich structures comprising cellular cores is predicted by finite element (FE) analysis. Two modelling approaches are investigated, providing different levels of accuracy; in both approaches, the sandwich structure is idealised as a layered stack with the skin modelled using shell elements; while the core is either modelled with fine detail using beam micro-elements representing the cell struts, or is modelled by three-dimensional solid elements after an appropriate core homogenisation.

The applied homogenisation methodology, as well as the all important modelling issues are presented in detail. Experimental tests performed using a mass-drop testing machine are used for the successful validation of the simulation models.

It was concluded that the core microscale models having detailed FE modelling of the core unit cells geometry with fine scale beam elements are suitable for the analysis of the core failure modes and the prediction of the basic core stiffness and strength properties. It was demonstrated that the homogenised core model provides significant advantages with respect to computing time and cost, although they require additional calculations in order to define the homogenised stress-strain curves.

Special microscale material tests are required for the determination of appropriate materials parameters of the core models, as steel selective laser melting (SLM) microstrut properties differ from the constitutive steel material ones, due to the core manufacturing SLM technique. Stress interactions were not taken into account in the homogenisation, as the applied core material model supports the introduction of independent stress-strain curves; however, the predicted load deflection results appeared to be very close to those obtained from the detailed core micromodels.

The paper is original. The dynamic behaviour of conventional sandwich structures comprising conventional honeycomb type cores has been extensively studied, using simple mass-spring models, energy based models, as well as FE models. However, the response of sandwich panels with innovative SLM cellular cores has been limited. In the present paper, novel modelling approaches for the simulation of the structural response of sandwich panels having innovative open lattice cellular cores produced by SLM are investigated.

Impact and fatigue are critical loading conditions for composite aerostructures. Compression behavior after impact and fatigue is a weak point for composite fuselage panels. The purpose of this paper is to characterize experimentally the compression behavior of carbon fiber reinforced plastic (CFRP) stiffened fuselage panels after impact and fatigue.

In total, three panels were manufactured and tested. The first panel was tested quasi-statically to measure the reference compression behavior. The second panel was subjected to impact so as to create barely visible impact damage (BVID) at different locations, then to fatigue and finally to quasi-static compression. Finally, the third panel was subjected to impact so as to create visible impact damage (VID) at different locations and then to quasi-static compression. The panels were tested using ultrasound inspection just after manufacturing to check material quality and between different tests to detect impact and fatigue damage accumulation. During tests the mechanical behavior of the panel was monitored using an optical displacement measurement system.

Experimental results show that the presence of impact damage significantly degrades the compression behavior of the panels. Moreover, the combined effect of BVID and fatigue was proven more severe than VID.

The paper gives information about the compression after impact and fatigue behavior of CFRP fuselage stiffened panels, which represent the most realistic loading scenario of composite aerostructures, and describes an integrated experimental procedure for obtaining such information.

Fatigue crack growth rate data for 2024-T3 aluminum are found using three parameters d*, σ* and μ* for short and long cracks for Regions I-III in conventional fatigue. Asymptotic solution of a line crack with a micro-tip is found to yield a singular stress behavior of order 0.75 in contrast to the 0.50 order known for the macrocrack. The difference is due to the micro-macro interaction effects. The three parameters account for the combined effects of load, material and geometry via the tip region. Data for short and long cracks lie on a straight with a slope of about 3.9-4.8 for R values of 0.286-0.565. The results were based on an initial crack a1 mm where a is the half length for a central crack panel. The paper aims to discuss these issues.

The belief that specimen fatigue data could assist the design of structural components was upended when FAA discovered that the NASGRO FCGD are not valid for short cracks that are tight and may even be closed. The regular ΔK vs da/dN model was limited to long cracks. The issue become critical for short cracks connecting the long ones of a few mm to cm or even m according to da/dN for the same crack history. The danger of short/long fatigue crack growth (SLFCG) prompted FAA to introduce an added test known as Limit of Validity (LOV), a way of setting empirical limits for structural components. The dual scale SLFCG data from ΔK micro/macro provide support for the LOV tests.

Data for short and long cracks lie on a straight with a slope of about 3.9-4.8 for R values of 0.286-0.565. The single dual scale relation on ΔK micro/macro can switch from microscopic to macroscopic or vice and versa. The difference is fundamental. Order other than 0.75 can be obtained for simulating different microstructure effects as well as different materials and test conditions.

Scale shifting from short to long fatigue cracks for 2024-T3 aluminum is new. The crack driving force is found to depend on the crack tightness. The sigmoidal curve based on the regular ΔK plot disappeared. The data from ΔK micro/macro for short cracks may supplement the FAA LOV tests for setting more reliable fatigue safe limits.

Continuous on-line monitoring of structural integrity are in priority in many engineering fields such as aerospace, automotive, civilian structures, and industrial applications. Of all these possible applications, the aerospace industry has one of the highest payoffs. Possible damage can lead to catastrophic failures and costly inspections. On the other hand, processing a signal consists of important feature from sensors measurements to reach the considered target. Typically, the sensors translate a physical phenomenon from one or many sources in temporal variations or in spatial variations. The purpose of this paper is to investigate damages, in terms of suddenly screw removal or in a small cut, detection in vibrating (clamped-free) aluminum beam structures using the empirical mode decomposition (EMD) method along with the Hilbert-Huang transformation (HHT). The perspective is to identify very small defects in real aircraft structures.

The proposed method deals with a new time-frequency signal processing analysis tool, for damages detection in a vibrating plate. An experimental test ring is used in order to excite a clamped-free aluminum plate. Two types of excitations are used. The first one is a harmonic excitation and the second one is a random excitation provided by an impact hammer. A hole and its filled by a screw with mass of 0.2 g, and a small cut is created, simulating a cut creation, are produced afterword, and the HHT is used in order to arise the developed oscillations, and to reveal hidden reflections in the data and to provide a high-resolution energy-time frequency spectrum.

The major finding was the clear amplitude increment either for screw removal or for cut creation, using the EMD process with the HHT, giving the possibility to detect them.

The use of the HHT to detect, using an experimental procedure, two different defects: a suddenly screw removal and a cut creation, in a clamped-free beam, excited by non-stationary and non-linear signals.