In this paper, a series of molecular dynamics (MD) simulations was performed to identify the high-temperature tensile properties of nanostructured polycrystalline CoCrCuFeNi high entropy alloy (HEA). The mechanism of strength reduction at elevated temperatures was understood through nanostructure observation such as phase transformation and dislocation evolution in MD simulation. The applicability of this material from room temperature to 1200 ℃ as a high-temperature use of structural material was identified. The stress-strain curve was found to gradually decrease ultimate tensile strength and yield stress as the temperature applied to the material increases. The elastic modulus decreases rapidly at slightly high temperatures but decreases gradually as it goes to the extremely high temperatures. Face-centered cubic (FCC)→hexagonal close-packed (HCP) phase transformation, which is energetic process between atoms due to tensile loading, was revealed. From the vicinity of the yield stress to the quasi-plastic regime, it competitively contributes to tensile properties between FCC→HCP phase transformation and growth of voids. In dislocation analysis, the typical partial dislocations such as perfect, Shockley, stair-rod and others were measured, in which Shockley and stair-rod partial dislocations show its characteristics that contribute to tensile properties. The results in this study contribute to understanding the high-temperature applicability of nanostructured polycrystalline CoCrCuFeNi HEA.

Carbon Fiber Reinforced Plastic (CFRP) is a composite material consisting of a resin matrix and carbon fiber reinforcement. The material is also commonly used in a prepreg laminate, a unidirectional reinforcing material with high strength and stiffness in one direction. Damage to laminates is highly complex, including delamination, fiber fracture, and base matrix cracking, requiring highly efficient and accurate non-destructive testing. Infrared stress measurement is an example of a non-destructive testing method for CFRP structures. The infrared stress measurement is based on Kelvin’s theory to convert the surface temperature fluctuation under cyclic loading to the distributions of sum of principal stress on surface (DSPSS), so it may result in a stress distribution that differs from the actual distribution due to transient heat conduction. It is necessary to consider structural analysis and transient heat conduction in the numerical analysis to reproduce DSPSS obtained by infrared stress measurement. This study performs a finite element analysis with transient heat conduction on simple shaped CFRP specimens to reproduce the trend of DSPSS obtained by infrared stress measurement. Firstly, DSPSS generated by the forced displacement of a CFRP specimen is converted to a temperature distribution using Kelvin’s thermoelastic theory. Finally, a transient heat conduction analysis is performed, and the distribution trend is discussed using the obtained temperature distribution as the initial value. A sheet of Teflon is inserted into the CFRP specimen as a defect, assuming foreign matter contamination during the manufacturing process. Previous study predicts the internal defect information by a machine learning model using the DSPSS from numerical analysis. There is a potential for a high-accuracy defect prediction using DSPSS obtained by the infrared stress measurement if DSPSS obtained by the infrared stress measurement and the numerical analysis have similar tendencies.

The objective of this study is to improve the mechanical properties of self-healing carbon fiber reinforced polymer (CFRP) with microcapsules (MCs) incorporating healing agents, in order to achieve superior mechanical properties and healing efficiency. The approach to improve the mechanical properties is to hybridize spread carbon fiber (SCF) and woven carbon fiber (WCF) as reinforcements. The interlaminar shear strength and the healing efficiency of the self-healing SCF/WCF hybrid laminates were evaluated by short beam shear tests, and then compared with those of self-healing SCF/EP laminates and self-healing WCF/EP laminates. Moreover, the finite element analysis (FEA) using a representative volume element (RVE) model of the SCF/WCF hybrid laminates containing MCs was performed to evaluate the elastic properties and the internal stress and reveal the relation between them and microstructure. Furthermore, we performed the FEA of short beam shear test using the elastic properties predicted from the RVE models and verified the validation of the RVE model by comparing the analytical results with the experimental results. The experimental results showed that the hybridization of SCF and WCF can be effective for improving mechanical properties of self-healing CFRP laminates compared to those of self-healing SCF/EP laminates. However, the self-healing SCF/WCF hybrid laminates did not present sufficient healing efficiency due to the large transverse cracks in the WCF layers. Therefore, in order to improve healing efficiency, it is necessary to suppress the large crack propagation in the WCF layer. The analytical results indicated that the stress concentration occurred in the WCF layer and could be mitigated by increasing the thickness of the SCF layer. The experimental results and analytical results were in good agreement. Therefore, the RVE model used in this study effectively predicted the properties of the self-healing CFRP laminates and provided microstructure design guidelines for achieving superior mechanical properties and healing efficiency.

This study employs artificial neural networks (ANNs) to predict the structural compliance of randomly generated irregular geometries derived from Finite Element (FE) calculations. By imposing volume constraints, the scope of the study is confined to applying ANNs for learning from structural data generated by considering either multiple random walks of a circle or a set of randomly placed circles with allowed overlaps. Numerical results indicate that the learning outcomes of the former approach are more satisfactory than those of the latter. This suggests that the effectiveness of employing ANNs for predicting the structural compliance of irregular geometries is contingent upon how the random geometries are generated and the material volume ratio. The learning outcomes of irregular structures generated by the former approach with a higher volume ratio exhibit greater satisfaction due to a higher degree of structural connectivity.

Progressive damage behavior of woven composite structure is simulated, based on decoupled multi-scale damage analysis to consider microscopic woven structure. Since woven composites have complex microstructure, it is difficult to predict their damage behavior by only experimental approaches. Evaluating damage process is important for the assessment of safety, and numerical method of simulating damage behavior is demanded. In this paper, decoupled multi-scale analysis method was adopted. The constitutive law used in analysis of structural member scale (macroscopic scale) was identified by analysis of weave-structure scale (microscopic scale) and two-scale simultaneous analyses are no longer necessary. In microscopic scale, representative unit cell (RUC) of woven structure was considered, and in-plane periodicity was assumed to reflect inhomogeneity caused by woven structure. Microscopic analysis was carried out to identify the constitutive law which is used in the macroscopic analysis. Transverse crack of fiber bundle was modeled using enhanced continuum damage mechanics (ECDM) model. Fiber/matrix interface debonding was also modeled by cohesive zone model (CZM). In the macroscopic scale, structural member was modeled. Macroscopic analysis was carried out to investigate initial damage occurrence location and damage process. Damage progress was modeled by stiffness degradation, using CDM model and microscopic analysis results were reflected to determination of macroscopic damage variables. Four-point bending test of macroscopic specimen was carried out for validation of the analysis. Damage observation showed the macroscopic analysis effectively simulated damage occurrence and propagation. Localization analysis was then carried out to investigate damage behavior of woven structure, reflecting deformation history of macroscopic initial damage occurrence location to microscopic analysis. By comparing the analysis to cross-section observation, initial damage simulation showed good agreements with experiment.

In this study, a parallel three-scale homogenization analysis simulator that accounts for macro-, meso-, and micro-scale structures was developed for the carbon-fiber-reinforced plastics (CFRPs) used in high-pressure hydrogen storage vessels manufactured using the filament winding (FW) method. The developed simulator enables detailed analysis that accounts for fiber irregularities in the fiber bundle tapes of FW-CFRP. Because numerical simulations that consider fiber irregularities increase computational time and memory usage, we developed a parallel computation system for three-scale homogenization using the domain decomposition method. Numerical examples using large-scale computers verified that our parallel three-scale homogenization analysis has parallel computing performance close to the ideal acceleration ratio. We then investigated the effects of fiber waviness in fiber bundle tapes on the macro-scale properties of FW–CFRP by considering waviness as an initial irregularity in the carbon fiber arrangement. Our analysis of fiber irregularities revealed that fiber waviness has a significant effect on macroscopic stiffness and stress. Within the scope of this study, the macroscopic stiffness and stress were reduced by 40% and 57%, respectively, compared to the results without irregularities.

In the present paper, the multiple applications of high-density pulsed electric current (HDPEC) with current densities between 100 and 200 A/mm² made fatigue crack growth (FCG) slow, in which a current density of more than 150 A/mm² provided high efficiency to refrain from FCG. The application of multiple HDPECs under the current density of 200 A/mm² showed a significantly reduced fatigue crack growth rate (FCGR). The fracture surfaces without the HDPEC effect showed signs of FCG along the crystal structure elongated in the rolling direction, on the other hand, a series of welded joints were formed at the crack tips due to the effect of applying multiple HDPECs, which provided different patterns of FCG leading to crack arrest. When initial HDPEC is applied in a short crack under the same HDPEC condition, it shows partial welding at the crack tip, while more concentrated welding at the crack tip by moderate Joule-heating is shown as fatigue progresses and the fatigue crack length increases. Elemental analysis by energy-dispersive X-ray spectroscopy and crystallographic analysis by electron backscatter diffraction also provided the microstructure features at the crack tip subjected to the application of multiple HDPECs. It is most efficient to perform in a short crack with multiple applications, and applying multiple manners as the crack progresses contributes to the accumulated improvement of fatigue properties by slowing the FCGR based on each application.

This paper examines the applicability of the hot spot method using a small wireless system and the effect of data loss during data communication on fatigue evaluation. Fatigue failure is a phenomenon that progresses due to repeated loading. Since it is not practical to monitor all structural members, it is important to monitor the structural integrity of damage-prone areas. In addition, due to cost and installation space constraints, the use of inexpensive, compact IoT devices is effective. Compact wireless fatigue monitoring methods using strain gauges have been proposed, and systems based on the nominal stress method have been considered. However, there are cases where the nominal stress cannot be defined due to the complex structure of the weldment, making fatigue evaluation difficult. In such cases, the hot spot method, which measures the structural stress concentration, is effective. However, the application of the hot spot method to a small wireless monitoring system has not been sufficiently studied. In this paper, a small wireless monitoring system was fabricated and its applicability to fatigue evaluation of welded joints was confirmed. From the test results, it was confirmed that the data loss was caused by the data communication condition of the small wireless system and the accumulated fatigue damage degree was underestimated. It is necessary to select an appropriate small wireless monitoring system considering the information on fatigue and stress state required for the target structure. This selection should consider the trade-off between transmitting raw data and performing edge processing before transmission. The former provides richer information but can cause data loss and high-power consumption due to the large volume of transmission. On the other hand, the latter reduces data loss and saves power, but can provide less information.

The Japan Atomic Energy Agency (JAEA) is developing an evaluation method for a two-phase flow in the reactor core using simulation codes based on the volume of fluid (VOF) method. However, it is impossible to simulate boiling on the heating surface in large-scale domains, such as fuel assemblies, using this simulation method because simulating boiling based on the VOF method requires fine meshes to resolve the initiation of boiling. Therefore, the JAEA started developing a simplified boiling model (SBM) for the two-phase flow simulation in fuel assemblies. In the SBM, the motion and growth equations of a bubble are solved to obtain their diameter and time length at the detachment, the size scale of which is within/around the calculation grid size of the numerical simulation. This information is given as the vapor injection flow rate for the boundary condition on the heating surface. JUPITER calculates the bubble behavior with a scale of more than several millimeters. Using the developed SBM, this study simulates convection boiling on a vertical heating surface. A comparison between the simulation and experimental results showed good reproducibility of the heat flux and velocity dependency on the passage period of the bubble.

Quartz tuning forks (QTFs) are highly effective force sensors in frequency modulation atomic force microscopy (FM-AFM), owing to their inherently high quality factor (Q). This attribute significantly enhances force sensitivity and stability during surface imaging. Our study introduces a counter piezo driving scheme to restore the diminished Q of QTFs affected by mass imbalances, which may arise from geometric discrepancies or the attachment of a tip to one prong. The approach involves vibrating the two prongs of the QTF using piezo elements situated at the joint, applying differing voltages. This strategy adjusts the mechanical coupling between the prongs, thereby influencing the Q. Unlike traditional methods that focus on prong-mass tuning, this technique does not require intricate modifications to the prong structure. Experimental validation was achieved with asymmetrically structured QTFs featuring prongs of varying thickness. A notable increase in the Q was observed, several times higher than that achieved with single piezo element driving, depending on the degree of asymmetry. The findings were corroborated by finite element method simulations, which not only confirmed the substantial Q enhancement in tip-attached QTFs but also elucidated the underlying mechanism. It was demonstrated that precise tuning of piezo element voltages effectively aligns stress vectors at the joint, leading to highly efficient excitation of resonance oscillation, resulting in a significantly improved Q.

To investigate the crack growth behavior in a cladded structure, a numerical simulation method based on the extended finite element method (XFEM) was developed. This method employs 8-node hexahedral continuum elements enriched only with the Heaviside step function, which can model planar crack independently of finite elements. For a crack across the interface of dissimilar materials in cladded structures, stress intensity factors along the crack front are evaluated by the domain integral method. Crack front shapes are updated by using Paris’ law and then smoothed by cubic Bézier curves in each region. In addition, fatigue crack propagation analyses for cladded compact tension [C(T)] specimens were performed. The relationships between crack length and load cycles as well as the transition of propagating crack front shapes were compared with experimental results and validated. Furthermore, sensitivity analyses were performed to explore the influence on the crack propagation behaviors of analysis parameters and conditions. The developed method was shown to provide an appropriate approximation of fatigue crack growth behaviors in cladded C(T) specimens.