Fatigue failure involves fatigue crack initiation and propagation processes. In bulk metals, the fatigue crack initiation process is strongly affected by a dissimilar surface layer, while the fatigue crack propagation process is less affected. In the case of the metallic thin films, which are a kind of small-scale metallic materials, have a large surface-area-to-volume ratio, and the fatigue crack propagation process is dominated by the formation of out-of-plane surface fatigue damage, intrusion/extrusion, which is observed in fatigue crack initiation process in bulk metals. Therefore, even a surface layer with a thickness on the order of nanometers would have a significant influence on fatigue crack propagation behavior of metallic thin films. In this study, in situ field emission scanning electron microscope (FESEM) observation experiments of fatigue crack propagation were conducted on freestanding copper thin films with a thickness of 500 nm with a thickness-controlled surface oxide layer in the range of ~0 nm to ~20 nm. The results showed that the fatigue crack propagation properties were influenced by the thickness of the surface oxide layer: acceleration of fatigue crack propagation, transitions of fatigue crack propagation mechanisms (from intrusion/extrusion mode to tensile-dominant mode) and unstable fracture occurred at smaller stress intensity factor range ΔK for the films with a thicker surface oxide layer than that for the films with a thinner surface oxide layer. Moreover, the fatigue crack propagation rate da/dN was smaller in the region near the fatigue crack propagation threshold and thus the threshold stress intensity factor range ΔK_th increased in the films with a thicker surface oxide layer. In situ FESEM observations showed that fatigue damage area around the fatigue crack became narrower in the films with thicker surface oxide layer, and the fatigue crack reached the threshold when intrusions/extrusions were rarely formed ahead of the crack tip. This suggests that the fatigue crack propagation threshold in metallic thin films is determined whether intrusions/extrusions are formed around the fatigue crack.

This research concerns a bridge condition identification method using a convolutional neural network (CNN). Currently, bridges are mainly inspected visually by humans, and it is difficult to detect damage that does not appear on the surface. Therefore, a condition evaluation method using sensors is required. In this study, a damage identification method is proposed by classifying the images, visualized by vibration analysis such as spectrogram or FFT of acceleration response of a bridge, using CNN. The effect of analysis methods, the presence or absence of a time component, the processing of the image itself, and frequency resolution on diagnostic accuracy are clarified. The overall Identification rate is higher for spectrograms containing more information, and for damage with less effect on vibration, the FFT has a higher Identification rate. Furthermore, a method to improve accuracy by combining these multiple CNNs using Bayesian estimation is proposed. Accurately identifying damage, the degree of which varies incrementally, was a complex problem for a single CNN. Combining multiple CNNs with various characteristics using attribution probabilities has reduced misclassification and improved identification rates over a single CNN.

The load on bogie frames for railway vehicles in service varies greatly, which requires careful consideration of both variability and uncertainty in load and material strength when assessing the fatigue life of welded joints. In this study, a new method was developed based on reliability engineering to clearly differentiate between reliability and confidence level. The results showed that a simple formula could be used to determine the probability of failure and safety factor when load variation was relatively small. For easy reference, we provided the tables of partial safety factors for the target reliability. In situations with large load variation, we found that Monte Carlo simulations were effective for calculating the probability of failure. Lastly, we compared our calculated probabilities of failure with the literature and discussed these results.

This paper discusses an approximate multiscale stochastic stress analysis of a randomly arranged unidirectional fiber reinforced composite plate under the transverse tensile loading. The multiscale stochastic stress analysis is desired for quantitative apparent strength evaluation of composites with random microstructure, and an efficient method has been attractive in terms of computational efficiency. In particular, since huge numbers of fibers are included in an actual composite structure, accurate and efficient estimation of microscopic maximum stresses for random fiber location variation and their probabilistic properties is required. For this analysis, the successive local approximation method is employed. The effectiveness of the approximate multiscale stochastic stress analysis has been discussed, while the square array is assumed in previous literature. For applicability confirmation of the method to evaluation of an actual problem, the usage of a hexagonal array model should be discussed. From these backgrounds, the above analysis with the hexagonal array model is performed, and validity and effectiveness of this approach are discussed with the numerical results.

The influences of surface corrugation with small amplitude on zero-pressure-gradient boundary layer instability were examined numerically by solving the three-dimensional compressible Navier-Stokes equations. Three-dimensional sinusoidal corrugations, with amplitude less than 10% of the boundary layer thickness and wavelengths nearly the same as that of Tollmien-Schlichting (T-S) wavelength, were considered. The results showed that the basic flow near the surface followed the corrugated surface without any separation and low- and high-speed streaks were formed in the middle of the boundary layer. The three-dimensional surface corrugation had little influence on the growth of T-S waves compared to two-dimensional surface corrugation, that is, the growth rate of T-S wave was almost the same as that of smooth wall case. Amplitude of energy production in the three-dimensional corrugation case was found to be about half of that in the two-dimensional case while energy diffusion remained nearly the same in both cases. It was also found that the transient distance of change in surface geometry from smooth wall to three-dimensional corrugation wall was only about one corrugation wavelength.

Oxymethylene dimethyl ether (OME) is a promising carbon neutral fuel for reducing well-to-wheel CO₂ emission and expected to suppress soot formation due to the lack of C-C bond. The present study aims to develop a chemical reaction model to predict soot formation processes of OME₃. Initial reaction processes of OME₃ within 10 ns after the onset of reaction were traced using molecular dynamics simulation under various Φ-T conditions (Φ: equivalence ratio, T: temperature), and reasonably converted to a chemical reaction model. The model was coupled with a conventional soot formation model for paraffinic fuels, which enables to simulate soot formation processes of OME₃. The Φ-T map of soot yield for OME₃ was calculated and it appeared that OME₃ can largely suppress soot formation compared with n-hexane as a paraffin fuel. Reaction path analysis revealed that the formation of unsaturated C2 species for OME₃ is delayed and reduced compared with n-hexane, which leads to the soot suppression effect of OME₃.

Ground heat is a renewable energy source that is less affected by atmospheric temperature and maintains a nearly constant temperature throughout the year. Therefore, it can serve as a heat source for hot water and air conditioning systems, resulting in a significant reduction in carbon dioxide emissions compared to conventional systems. In Japan, the recognition of the use of ground heat is still low, and the prices of ground heat exchangers and geothermal heat pumps tend to be high. In addition, the excavation cost is very high, so it is very important to reduce the construction cost. One cost-saving method is to install a ground heat exchanger in the core material of the earth retaining wall, which has been shown to reduce construction costs by more than 50%. In this study, a numerical program was developed to clarify the performance and characteristics of a ground heat exchanger attached to an H-beam in a “Pre-installation method using the earth retaining wall”. Then, the factors influencing the heat extraction performance of two U-Tube heat exchangers connected in series, four U-Tube heat exchangers connected in series and two W-Tube heat exchangers connected in series were investigated. Firstly, the validity of the calculation program was confirmed by comparing the numerical results with the experimental thermal response results. Secondly, numerical simulations were performed using the calculation program, and the following results were obtained. Under the conditions defined in this study, we found that the rate of heat extraction increases as the flow rate of the heat extraction medium increases in all ground heat exchangers. However, it becomes difficult to increase the heat extraction when the flow of the heat extraction medium reaches a turbulent area. All three models confirmed that the H-beam expanded the area where the formation temperature changes and increased the temperature of the heat extraction medium. This led to an enhanced heat extraction performance of the ground heat exchanger. Finally, we compared the heat extraction performance of several ground heat exchangers using the 'coefficient of heat extraction or rejection'. The results showed that the four U-Tube model had the highest performance, followed by the two W-Tube model, and the two U-Tube model had the lowest performance, although the differences were small.

In building a human-robot cooperative system, it is an important issue to reduce the harm caused by contact with the robot. A protective measure for a robot with a sharp and highly rigid surface is to smoothen the surface shape and reduce the rigidity in order to increase the contact area, and protection with a soft covering is effective. In this research, we focus on the event when the edge of the robot contacts the human body, especially the outer arms or legs. In order to design an appropriate soft covering, it is necessary to estimate the contact pressure between each part of the human body and the edge covered with the soft covering. The contact pressure can be derived from a contact model that considers the curved surface shape of the human body, the radius of curvature of the soft covering and edges, and the stress balance that considers the elastic modulus that hardens the hardness of the human body and the soft covering according to the compressive strain. Then, the validity of the proposed method was confirmed by material experiments to grasp the material properties and experiments simulating contact, using appropriate materials for the contact model targeted in this research. The proposed method can contribute to understanding the tendency of contact pressure by arbitrarily estimating the curvature radius of the edge and the hardness and thickness of the protective covering that covers the surface when applying the soft covering to the edge of the robot surface.

This study describes the development of a body-powered upper-limb prosthesis for children with congenital digit defects that realizes pinching motions using their own residual digits and joints. Existing digit prostheses assume that the thumb remains intact. If there is no thumb, a conventional upper-limb prosthesis cannot sufficiently restore grasping function. The developed body-powered upper-limb prosthesis was designed to mechanically convert the flexion and rotation power of the residual digits into the power of the opposing prosthetic digit. This conversion enabled the reconstruction of opposing pinching motions at the fingertip. Consequently, this design enhanced the function of the residual digit in the absence of the thumb. In this study, we developed a body-powered upper-limb prosthesis applicable to a single residual digit or two parallel digits. The digit prosthesis, which includes the thumb, was designed to synchronize its movement with the motion of the metacarpophalangeal (MP) joint of the residual digit. The prosthesis joints enable two degree-of-freedom (DOF) motions: flexion/extension and adduction/abduction. The digit opposing the thumb can be selected through this joint by the adduction and abduction rotation movements of the residual digit. The mechanism performance was evaluated using tensile tests, which confirmed that the force applied to the residual digit was linearly transmitted and exerted as a pinch force on the prosthesis. This feature allows for a more intuitive use by the user. The evaluation of its application in children with congenital digit defects revealed that expanding the palm allows the user to throw and catch a ball and use scissors in daily life.

Vibration related to several performances such as durability and noise performance is the one of the important performances on automotive product. The technique of reshaping eigenmode is a practical way to improve the performance on the product design phase. However, it is difficult to design desired mode shapes efficiently because the mode shapes are determined by the balance between mass and stiffness distribution on a whole structure. In this paper, a method for designing desired eigenmodes through changing internal forces between subsystems is proposed. Firstly, according to frequency based substructuring, the dynamic stiffness matrix of a whole structure could be divided into the matrices of subsystems and internal forces occurred from coupling components. Then, on a specified eigenfrequency as a design target, it can be derived that the mode shapes are controllable by changing to desired internal forces. Secondly, desired internal forces are calculated through designing spring constants between subsystems based on the kernel compliance analysis method which can analyze vibration when several subsystems are coupled on multiple degrees of freedom. Furthermore, when desired internal forces are calculated, a method to change a component of the internal force vector one by one is also proposed. This method can visualize the range which can be designed all spring constants as positive values in advance and can avoid selecting negative spring constants. Finally, this proposed method is applied to a numerical case study to reduce vibration responses with allocating modal strain energy to subsystems through reshaping an eigenmode.

In this study, we have developed a non-destructive inspection system for detecting external wall tile delamination by incorporating a pulse water jet impingement force into a suspended apparatus. The delamination inspection using a pulse water jet offers flexibility in adapting to variations in the distance between the inspection device and the inspection surface, a challenge inherent to the suspended apparatus. Furthermore, the delamination inspection using a pulse water jet exhibits exceptional safety and cost-effectiveness. Using this inspection system, we conducted a delamination detection test on tiles with delaminated mortar and were able to detect it. The inspection speed of this apparatus is approximately 10 centimeters per second in horizontal movement, allowing for an inspection rate of roughly 18 square meters per hour for tiles with a 50-millimeter spacing. This inspection speed is comparable to that achievable by human operators. Moreover, the system can perform inspections at heights of up to 30 meters above ground level. This inspection system has the potential to address the issue of labor shortage by facilitating automation.

Magnetic Resonance Imaging (MRI) is used in the medical field, and capable of precise examinations such as cancer detection. While MRI has the benefit of no radiation exposure compared to CT, it can cause noise sound from the magnetic field coils and stress on a patient due to the time required for the examination. Additionally, it is difficult for the physician and patient to communicate using a conventional microphone because the magnetic field in the MRI room prevents the use of devices that contain magnetic materials. In this study, we aim to achieve communication in a high magnetic field and high noise environment such as an MRI room, and then propose a communication method using an optical-body-conducted speech microphone. Optical-body-conducted speech is a speech that is conducted on a human body obtained by a contact-type optical microphone. It has noise robust characteristics against airborne noise. It is similar to conventional body-conducted speech, so it has low-quality sound because it is attenuated at 2 kHz and higher. In particular, optical microphones can be used in strong magnetic fields because they are made of non-magnetic materials. Therefore, we propose a sound retrieval method for optical-body-conducted speech which is noise suppression and sound quality improvement of the optical-body-conducted speech using a convolutional neural network to obtain clearer and more natural speech from optical-body-conducted speech. We confirmed improvement in the sound quality of the optical-body-conducted speech by objective evaluation using MCD with CNN-WaveGlow transformation for sound quality improvement.

From the viewpoint of preventive maintenance, the analytical ferrography method is widely used as a method to diagnose the wear condition of worn surfaces through wear particle analysis of lubricating oil, and production losses are prevented by replacing or renewing lubricating oil based on the analysis results. On the other hand, wear particle analysis by analytical ferrography requires determining in the wear classification and its cause by comprehensively judging the wear particle size, color, shape, and surface properties, so that the analysis needs specialist knowledge and often takes time. Therefore, attention was paid to convolutional neural networks (CNN), which are compatible with image analysis. Currently, studies evaluating the accuracy of wear particle image analysis models using CNN have been reported. Still, it is not clear what features of the wear particles are important in determining the wear mode from the wear particle image. Therefore, in this study, a CNN model that discriminates four types of wear modes was first created using the Neural Network Console, 100 single wear particles were analyzed, and it was shown that the CNN model in this study could distinguish wear modes. Next, visualizing the basis of AI's decision-making when identifying wear particles using Grad-CAM from the analysis results could demonstrate that AI would learn by focusing on the lightness and darkness of the color in the wear particles image analysis. Furthermore, concerning the lightness and darkness of the color, it was found that cutting wear particles and spherical wear particles were identified based on the black color in the image, flat wear particles were identified based on relatively light colors such as yellow, yellowish brown and light blue, and severe wear particles were identified based on relatively dark colors such as brown and dark blue.