中村 俊哉

Surface treatment is important to ensure adhesion properties of adhesive bonding or coating. Although laser irradiation or atmospheric plasma treatment has attracted as innovative surface treatment in aviation industry, silica deposition treatment could be another choice to improve adhesion properties of aircraft structural applications such as adhesive bonding joints or coating. The silica deposition treatment is a kind of flame treatment which uses liquid petroleum gas containing silane compounds as combusting feed, and could be quick, easy, clean and costless process. In this study, mechanical strength properties of the silica deposition treatment were investigated. Two kinds of testing were conducted. One is single lap tensile tests of adhesively-bonded joints using carbon/epoxy composites or titanium alloys. The results of the tests indicated lap shear strength increased as the number of passes increased. The 12 times passing coupons of both of composites and titanium alloys exhibited high strength comparable to the conventional treatment coupons. The other is coating adhesion pull-off tests using aluminum alloys. Just only one-time pass brought high adhesion comparable to the conventional treatment. This means, as for aluminum coating application, the silica deposition treatment might be beneficial treatment in actual production scene compared with conventional ways.

Out-of-plane impact loading causes both the internal damage including interlaminar delaminations and the visible deformation on CFRP laminates. The damaged specimens subjected to drop-weight impact were observed in detail by several inspection systems: UT, X-ray CT, and shape measurement systems with laser. The influence of laminate thicknesses, stacking sequences, impactor shapes were observed using cross-ply and quasi-isotropic CFRP laminates. The results showed the bulge with the fiber breakage affects the detectability of the impacted part. By quantitatively evaluating the dent depth and the projected dent area, it showed the possibility that more diverse discussions could be made about dent detectability.

Three different panels assuming a representative part of aircraft lower wing are designed, manufactured and evaluated to quantify the manufacturing capability of current state-of-art AFP technology as well as the potential performance of an optimized composite panel.

A method is proposed to estimate the dynamic load on a beam from measured strain data. The FE method is used to model the beam with the central-difference scheme for time-integration. Since we do not have enough data to input into the equation of motion, the problem becomes mathematically ill-posed and then a technique of inverse analysis is employed. Furthermore, because the deflction and slope cannot directly be measured in actual cases, we introduce an operation to transform the strain data, which is usually measured, into the deflection and slope in the numerical scheme of the load identification. The the validity and the stability of the present method will be numerically discussed.

Fiber Bragg Grating (FBG) sensors have widely been used to monitor temperature and strain distributions as a part of the structural health monitoring system. Since FBG has the sensitivity to the variations in both temperature and strain, a compensation is required to separate the strain or temperature data from the sensor output which is the shift of the grating's Bragg wavelength. The present study develops a computational inverse thermoelastic analysis method to separately identify the thermal and mechanical boundary conditions (loads) from the output of the FBG sensor. Numerical study has been made for a corrugate-core sandwich integral thermal protection system (TPS) to examine the method. The discussion is focused on the computational stability. The results reveal that the identification of the mechanical load is less stable than that of the heat flux. It is also shown that the condition number of a coefficient matrix serves as the index of the stability of the inverse analysis.

In this study, crack propagation tests were conducted to clarify property for the fatigue crack propagation of Friction Stir Welded (FSW) 2024-T3 aluminum alloy. FSW panel has residual stress around weld line and the longitude residual stress is higher around it. The peak tensile residual stress is about 180 MPa in this case. To understand fatigue crack growth property on FSW panel, crack opening stress measurement of the base material and FSW joint using an extensometer with the modified tool is also conducted during the crack growth test. The modified tool is jig to mount the extensometer to the specimen with magnets. The test results indicate that the accuracy of crack opening stress measurement is improved. In addition, the crack growth acceleration and decelerate around the FSW line under the low applied stress range (25 MPa) is bigger than that under high applied stress range (50 MPa). This means the effect of residual stress under low stress range is relatively larger than that under high stress range.

We equipped a composite wing structure fiber Bragg grating (FBG) arrays including 246 FBGs with 10 mm gauge length, eight and six long-length FBGs with 300 mm and 500 mm, respectively. The length of the wing was 6 m and it was made of carbon fiber reinforced plastics (CFRP). The sensing system based on optical frequency domain reflectometry (OFDR) was used in a series of load tests. The measured results by FBG arrays showed the overall deformation of the wing and good agreement with analysis results. Additionally, strain distributions of stress concentration zones were successfully measured by long-length FBGs. © 2012 SPIE.

Thermal analysis is required in the design of the structure of atmospheric reentry vehicles, which are subjected to severe aerodynamic heating. However, prediction of aerodynamic heating is difficult owing to highly complex physical phenomena. Inverse heat conduction analysis, the methodology to estimate heat flux on boundaries and entire temperature distribution from limited number of temperature measurement, is expected to solve this problem and to contribute to improving structural integrity. Present study develops computational method of transient inverse heat conduction analysis using finite element method and pseudo-inverse matrix. The developed inverse analysis code is applied to a reentry vehicle in order to examine the present method and to discuss the computational stability and regularization methods. Sequential function specification (SFS) method and rank reduction are employed to improve the accuracy and stability of the inverse analysis. The results of the numerical simulation reveal that the present method works effectively in solving inverse problem stably by using the combination of SFS method and rank reduction. Especially, the rank reduction is quite efficient at regularization. SFS method requires proper number of future time steps. © 2012 American Institute of Physics.

In this study, cyclic loading experiments and metallographical observation with scanning electron microscopy were conducted to clarify the fatigue life and the origin of fatigue failure of friction-stir-welded (FSW) 2024-T3 aluminum alloy sheet (t = 2 mm) and base material. The observation revealed that FSW specimens were fractured by particles as well as the base metal. The fracture occurred away from the weld area, and the FSW specimen had a fatigue life that was comparable to that of the base metal. The metallographical observation revealed that bigger particles in the stirred area were observed much less frequently compared to the rest of the area. Particle distribution in other areas was similar, regardless of the distance from the center of the weld. On the basis of these findings, we conclude that polished FSW joints have a fatigue life similar to polished base metal. © 2011 IBF (RWTH Aachen) & IUL (TU Dortmund).

We have implemented strain measurements along three sensing lines of long-length fiber Bragg grating (FBG) sensors in a wing model. The sensing system based on optical frequency domain reflectometry (OFDR) can measure strain at an arbitrary position along long-length FBG sensors with the high spatial resolution. We estimated displacement from the measured strain data by applying the beam theory. In addition, in order to investigate the applicability of the load identification method based on the inverse analysis, we applied it to a simply supported beam. In this experiment, the strain distribution along the beam was measured by the long-length FBG and the displacement distribution was also estimated from the measured strain data. Then the applied load distribution was identified by the inverse analysis using the strain or the displacement data.

Elasto-plastic FEM is used to examine the crack opening stress of a 2024-T3 Aluminum alloy sheet and a friction stir welded (FSW) panel. To investigate the effect of plastic deformation around the crack on the crack opening stress, the effects of the following two parameters, the distance between the weld line and the center of the crack starter, and the magnitude of the tensile residual stress on the crack opening stress, are evaluated. The da/dN-ΔK curves and a-N curves for the FSW plate are obtained numerically using an experimental da/dN-ΔK curve for the base material and the calculated crack opening stress. In addition, the da/dN-ΔK curves where ΔK is evaluated by correction factor and its a-N curves are obtained analytically. Comparison of these numerical results with the results of empirical tests shows while that the FEM result conforms to experimental data, the calculations based on the correction factor show poorer correspondence. These results demonstrate that FEM can reasonably predict the da/dN-ΔK curves and a-N curves for an FSW panel.

The purpose of thisstudy is to evaluate the crack closure phenomenon of 2024-T3 aluminum alloy panel and the FSW panel joined by the alloy. Focus attention on crack opening stress, its calculation using elasto-plastic FEM is conducted to clarify the influence of residual stress on crack opening behavior. The calculated crack opening stress of base material and the FSW panel are in close agreement with the measured results. And the FEM results show that the part of crack faces around FSW weld line is opened by tensile residual stress despite no external load. This behavior conforms qualitatively to that of test result. The da/dN - aKeff curves for FSW panel are obtained using the crack growth test results of base material and the calculated crack opening stress for base material and the FSW panel. The obtained curve conformed to the experimental result, indicating the crack growth rate of FSW panel could be predicted quantitatively by evaluating crack opening stress in elasto-plastic FEM.