For the purpose of evaluation of the flow filed and the stability of the groundsill, numerical simulations were performed for fixed bed hydraulic model experiments of river groundsills with a hydraulic drop. The numerical results are in good agreement with experimental data. With analyzing the numerical results, it is shown that the groundsill blocks are swept out due to hydrodynamic forces acting on the blocks, which are observed in 2D and 3D movable bed hydraulic model experiments. The computations also show that pore water pressure decreases with increasing permeability of materials under the structure, increases significantly with one-piece structure groundsill. With the result of numerical experiments, it is clear that sealing works installed at front under groundsill may contribute to dynamic stability in terms of flow forces, but sealing works installed at the end may benefit countermeasure against piping.

This study is to examine applicability of air-ion concentration to improve estimation accyracy of production rate of sea spray from the ocean. Concentration of air negative ions that is know to be influenced by generation of tiny sea spray was found to response to variation in production of film droplets induced by bubble bursting on water surface from our laboratory experiment. Also, concentration of air negative ions was observed in three ocean areas along Hokkaido, Japan; Sea of Japan, Sea of Okhotsk and the Pacific Ocean. The ion concentration was found to be correlated with the wind speed, wave height and whitecap ratio, all of which have been used as a model parameter of Sea Spray Generation Functions, indicating possibility of using air-ion concentration to accurate estimation of sea spray production.

In the coupled aero-hydro-servo-elastic analysis to predict the response of the floating offshore wind turbines, two models are widely used for the estimation of the hydrodynamic loadings; the Morison’s equation, and the potential flow theory. In this study, a semi-submersible type floating offshore wind turbine was analyzed with both hydrodynamic models with the same flexible structure, and obtained responsed were compared to each other. It is shown that while both models predicted the floater motion well, both underpredicted the pitch natural frequency components. For tower-base moment, the Morison’s equation overpredicted tower 1^st mode component, while the potential flow theory underpredicted wave excitation components for wind turbine operating condition. For the shear force and bending moment in the horizontal brace in air of the floater and the bending moment in the horizontal brace in water, the same frequencies with towerbase moments were excited, and the Morison’s equation predicted larger valules for all components than the potential flow. For the shear force in the horizontal brace in water, only the wave excitation frequency were excited and the two models showed similar results.

Three-dimensional sedimentary structures occur frequently in coastal and river environments. Brick-pattern ripples show a regular three-dimensional configuration composed of transverse crests and longitudinal bridges between the crests. Various studies have been made to examine the formation mechanism of brick-pattern ripples. However, these studies are insufficient to explain how the brick pattern appears on the surface of sand in an oscillatory flow. In this study, we make experimentally the detailed observation of the formation process by means of flow visualization and SfM-based 3D modeling to examine the brick-pattern ripple formation.

In recent years, technologies using big data such as deep learning have attracted attention for improving the efficiency of structure inspection. However, deep learning has a high degree of reliance on the quality and quantity of data, and the preparation of a data set on abnormal events used for supervised learning is a bottleneck in practical application. Therefore, in this study, for the purpose of complementing such problems, we examined and considered for the application to the detection of damage of concrete surface about the anomaly detection method by learning using the normal image. As a result of verification, it has been confirmed that the anomaly detection based on learning by normal images is effective for detecting the damaged area.

In the CFRP bonding method, it is necessary to satisfy the required bond length in order to obtain sufficient repair effect by CFRP. The required bond length is the length from the end of CFRP required to sufficiently transmit the force from the steel member to the CFRP via the adhesive. On the other hand, an energy release rate, ERR, is employed for the design of prevention of the delamination of CFRP when the required bond length is satisfied. However, the corrosion of steel bridges often occurs in narrow spaces, and it is difficult to satisfy the requiring bond length. In the case of insufficient bond length of CFRP bonded repair, ERR cannot calculate by the simple equation.

In this study, the authors indicate the calculation procedure of ERR for CFRP bonded steel plate with the insufficient bond length by using the shear stress in adhesive given by the shear lag theory. Furthermore, it was shown that the ERR given by the shear stress in the adhesive is almost the same as that given by FE analysis when the insufficient bond length is within the half of the required bond length.

Helicoidally laminated CFRP specimens replicated after a biological body are examined in plate flexure. The specimens were fabricated by stacking unidirectional prepregs with constant fiber angle difference between adjacent laminas. Point loading and distributed loading experiment in plate flexure were conducted, and flexural and fracture behaviors were observed. Compared to cross-ply laminate specimens, helicoidally laminated specimens show widely spread cracks and unique twisted delaminations. Under distributed loading, helicoidally laminated specimens show post-peak load recovery up to the peak load level.

In the seismic response analysis using high-fidelity models of large-scale reinforced concrete structures, such as nuclear power plant buildings, the application of high-performance computing is indispensable owing to the number of the degrees of freedom. This study shows the development of a seismic response analysis method for large-scale RC structures with FEM utilizing HPC. We applied the developed method to a seismic response analysis of a nuclear power plant building with surrounding ground, which is approximately 1.5 million DOF problem. This paper shows the results and discusses the analysis performance and effectiveness of the method.

Steel railway bridges are said to be inferior to concrete bridges in noise and vibration. Although the vibration damping material is used as a countermeasure for the web plate, the vibration damping material measures have restrictions in the use range. In addition, the vibration damping material often does not match the reduction effect of the vibration of the steel plate and the vibration radiation noise, and the reduction effect by the vibration damping material installation cannot be grasped correctly. Therefore, in this research, we proposed a method to evaluate the countermeasure effect based on particle velocity, and verified the validity of this evaluation method. Furthermore, we proposed a countermeasure using polymer cement mortar (PCM) with no restriction on the place where it can be installed. Based on the particle velocity and vibration velocity obtained by the test, it was verified that the two are equivalent in performance by comparing the reduction effect of the PCM and the damping material. In addition, it was confirmed that the reduction effect of PCM can be grasped by analysis by comparing the reduction effect of vibration velocity obtained by analysis with the test result.

In case of large earthquake, it is important to grasp the damage distribution over a city because such information is very useful for taking emergency activities. Therefore, many researches have been carried out to develop a system for estimating the damage of structure. In many researches, acceleration responses or other motions are observed with sensors and the measured time series data are analyzed to find out the change of vibration characteristics due to the damage or to estimate the deformation of structure. On the other hand, we focus on the breaking sounds of structural members emitted in a large earthquake to assess the damage of structure. In this research, frist, breaking sounds of timber and other sounds were observed in a silent laboratory, and the recorded data were used for machine learning of NN (neural network). In the analysis, about 80% of samples were correctly judged by using the learned NN model. The echoes of breaking sounds were also investigated whether they could be categorized into the breaking sounds or not. The k-Shape, that is one of the clustering methods, were also applied to the recorded sounds to study the ability for classification. Comparison between the analyzed results of them show that the NN model gives better success rates of detecting breaking sounds of timber in this research.

Rail breakage due to transverse crack originating from squats and gauge corner cracks has been occurring in the rail head, therefore it is required to research on crack propagation.

In this research, we carried out measurements of residual stress inside the rail head, and FEM analysis taking into account the rail shape and the singular field of the crack tip. Then we evaluated the influence of the residual stress and the crack initiation position on the transverse crack growth rate. It was shown that the residual stress inside the rail head which were changed due to annealing and usage affects the transverse crack growth rate. It was shown that the FEM analysis accuracy is good agreement by reflecting the measured residual stress and rail shape etc. The analysis result without taking into account the measured residual stress is about half of transverse crack growth rate where it is taken into account. It was clarified that the depth of transverse crack leading to the rail breakage is different by the position of transverse crack being different in the rail cross section direction.

Unsupported sleepers which are found sporadically at railway ballasted tracks, cause various problems in track geometry management, one of which is track buckling in the section of continuous welded rail (CWR). Unsupported sleepers reduce lateral ballast resistance force as the buckling resistance and reduce the buckling stability of the track. However, sufficient research of unsupported sleepers has not been executed, and the definite influence was not clear.

In this study, we grasped the actual condition of unsupported sleepers in the section of CWR in commercial lines. Based on the results, we clarified the influence on the track buckling temperature due to the support state of sleepers using a track buckling analysis program. In addition, by focusing on the relationship between the distribution of lateral ballast resistance force and the track buckling temperature, we showed the possibility that the buckling temperature can be estimated from the strain energy value of the lateral ballast resistance force.

This study aims to propose an efficient modal identification method for bridges under operation. The noisy condition caused by the traffic loadings is one of the difficulties involved in the operational modal identification. To cope with the problem, this study quantifies uncertainty involved in the modal properties utilizing Bayesian statistics. The quantified uncertainty enables to determine the reasonable model order and to extract the stably estimated modal properties from the determined model. The proposed method is applied to traffic induced vibration measured from an actual truss bridge. The extracted modal frequencies well correspond to peaks in power spectral density curves. Twelve bending and tortional modes are efficiently extracted by the proposed method. Six modes in the twelve is possibly newly observed modes.