This paper investigates changes in modal properties of a steel plate gider bridge due to artificial damages at the girder end. The modal properties are identified from vehicle-induced vibration data measured during the damage experiment on the bridge. Stochastic subsupace identification (SSI) is adopted to identify modal properties of the bridge, and dominant modes are extracted by means of stabilization diagram (SD). Observation demonstrates feasibility of damage detection from changes in modal properties. In order to examine the mechanism of the change, eigenvalue analysis utilizing finite element model with artificial damages is carried out. The mechanism of increasing frequency for the first bending mode is discussed by means of sesnsitivity analysis. Observations showed that the first bending mode is more greatly affected by changes in boundary condition due to damage than those changes in stiffness due to the damage. It is noted that in vibration-based bridge health monitoring it needs to consider the influence of change in boundary conditions at the supports.

View full abstract

A three-dimensional analysis method is developed for subway tunnel/soil dynamic interaction problems. Saving of computational cost is achieved by applying Floquet transform and Fourier expansion in the track direction to the railway track and tunnel/soil subsystems. This formulation needs the finite element discretization only in the tunnel cross section. The infinite soil domain is represented by an impedance matrix derived from analytical solution. Through numerical analyses of vibration reaction due to a stationary harmonic loading, the influence of track structures on the vibration reduction is discussed.

View full abstract

The mechanical behavior of corroded steel members is of interest due to the increased age of steel structures around the world. Especially, for the assessment of safety and development of the better maintenance plan, there is a compelling need for evaluating the residual strength. FEM analysis with solid elements is expected to evaluate the residual strength with high accuracy, however, it is computationally expensive. For this reason, FEM analysis with shell elements of which thickness is average thickness is often employed, but there is a problem that the result tends to be on danger side. This research proposes the method to evaluate the effective thickness by the convolutional neural network. The accuracy and effectiveness are verified by comparing FEM analysis with solid elements and shell elements of corroded steel plates and H-beam.

View full abstract

Unsupported sleepers which are found at railway ballasted tracks, cause various problems for track geometry management. However, because of the difficulty of grasping the situation, enough research has not yet been carried out. In this study, we examined a method for detecting unsupported sleepers easily and accurately based on general track data. Using the 2D FEM model proposed and track irregularity data of commercial lines, we calculated the amount of gap between the undersurface of the sleeper and the upper surface of the ballast, and compared those calculated values and the measured values on site. As a result, it is confirmed that they are in good agreement and this mothod is valid as far as the continuous welded rail section is concerned. Also, we showed the possibility of improving the accuracy of detection by reflecting the shape of the rail joint and broadening the bandwidth of the wavelength for restored track irregularity.

View full abstract

In this paper, we have developed a method to evaluate time dependent fatigue strength of structures based on train operation simulation using the railway operation information big data such as standard time table, run curve, actual riding ratio and running vehicle type. As a result of application of the method to the actual Shinkansen line, a double-track loading probability where trains on both tracks mostly intersect is 5% in the vicinity of the station and 0.2% to 1% in the middle section. In addition, the probability where trains on both tracks simultaneously intersect is 0.5% and 0.02 to 0.1%, respectively. At the position where the occurrence probability of the double-track loading is high, the probability is about 0.5% of the whole train passage at maximum; however, the proportion of double-track loading to the equivalent repetition number of fatigue strength is about 20%. The proposed method enabled to simulate the time dependent fatigue strength of the PC steel of the target bridge based on the assumption that the current operation conditions will be continued in the future.

View full abstract

The authors have been developing an anomaly detection system for road signs. As monitoring the acceleration responses, the temporal variations of vibration characteristics are contaminated by daily fluctuations due to the surrounding conditions. Therefore, in this research, a forced excitation experiment using a small vibrator was carried out for the road signs constructed in our campus, and the frequency transfer functions were accurately estimated with ARX method. And, a numerical model which is able to reproduce the frequency transfer function obtained by the forced excitation experiment was developed. The model parameters were determined to represent the averaged vibration characteristics. In addition, the effects of damage, temperature and base motion on natural frequency were studied using the numerical model developed in this research.

View full abstract

In this research, a small six-axis seismic sensor equipped with a MEMS acceleration & gyro sensor is developed and an estimation method for the maximum interlaminar deformation angle of a structure is studied. The attitude angle can be estimated by integrating the angular velocity measured by the gyro sensor. However, in general the estimated attitude fluctuates due to the bias error. In this paper, the method which estimates the attitude as correcting the bias error of the angular velocity by using the Kalman filter is studied. In this method, the bias error is estimated to fit the attitude with the output of the acceleration sensor. To verify the precision of this method, a single-axis vibration test is conducted. And, the precision of attitude angle is investigated in case that the acceleration data is contaminated by seismic motions.

View full abstract

In numerical models for evaluating the load carrying capacity of bridges, connections between slabs and main girders are generally modeled as rigid even for bridges designed as non-composite. This is because it is thought that slabs and main girders deform as composite beams due to slab anchors and steel-concrete bonding under the design load condition. Measurement results on real bridges supporting this idea have also been reported. However, in the case of evaluating redundancy of bridges after damage, it can not be expected to exist composite action between slabs and main girders, unlike in intact case under the design load condition. In this paper, focusing on the nonlinear behavior of slab anchors, we have conducted a series of nonlinear finite element analyses of a non-composite multi-girder steel bridge. As a result, it can be concluded that the bending stress of main girders depends considerably on the behavior of slab anchors and that slab anchors can be yielded under the load smaller than the design load.

View full abstract

This study is aimed at proposing a fracture energy calculating method for ductile fracture analysis of welded joints subjected to cyclic bending load. Finite element analysis using shell elements is performed to investigate mesh dependency of fracture energy. As a result, it is revealed that equivalent plastic strain upon deleting the element is independent on element sizes. By performing solid element analysis without deleting elements, an estimated formula of equivalent plastic strain upon deleting the element is obtained. Based on these results, a new calculating method of fracture energy corresponding to element sizes is proposed. By comparing the results of the analysis with deleting elements using the proposed method to the experiment results, it is showen that the proposed method is mostly appropriate for practical use.

View full abstract

In this study, we investigated the effect of trajectories of bi-directional accelerograms on nonlinear seismic response of 3D soil column. Seismic response analysis for 3D soil column model were conducted at frequency domain and time domain, as input four kinds of bi-directional accelerograms matching the same elastic response spectrum.
As a result, in frequency domain, the maximum response of soil column is almost the same value, irrespective of the difference in acceleration trajectory.On the other hand, in time domain, the maximum seismic response values of soil column are different for each acceleration trajectories. In particular, in case of circular trajectory, seismic nonlinear response tends to increase.

View full abstract

As a construction method that can be constructed in even narrow space to suppress both liquefaction and lateral defomation of the ground behind the quay, referring to the previous study which was confirmed effectiveness, we proposed a method for arranging steel sheet-pile shear walls in the direction perpendicular to the quay wall. By analyzing using a soil-water coupled finite deformation analysis code, GEOASIA, we confirmed that the proposed method was useful in terms of suppressing liquefaction and lateral defomation. Furthemore, we showed the effectiveness of a new countermeasure method by revealing that the reinforcement effect was proportional to the distance between the sheet-piles and shape of steel sheet-pile. This is a result that can be derived by finite deformation analysis.

View full abstract

Numerical simulation based on the continuum mechanics is one of the potential evaluation methods for surface fault displacements. We have developed a parallel finite element method program for the evaluation. In this paper, we applied the numerical method to the simulation of 2014 Nagano-ken-hokubu earthquake in which the surface faulting was observed. We modeled a 5km×5km×1km domain around northernmost region of the surface faults including secondary faults. We applied the forced displacements on the bottom surface of the model based on the slip distribution on the primary fault and the elastic theory of dislocation. As the input slip increased, the surface slip appeared both on the primary fault and a secondary fault. Calculated surface slips were in good agreement with measured surface slip. Calculated surface slips on secondary faults were strongly dependent on the input displacements of the vicinity.

View full abstract

CFRP is a composite with lightweight, high stiffness, and high strength, however it behaves in a brittle manner with localized fracture. In the current study, helicoidally laminated CFRP specimens replicated after a biological body are examined in flexure for the purpose of mitigating a brittle behavior. Four kinds of CFRP specimens were fabricated by stacking unidirectional prepregs with constant fiber angle difference between adjacent laminas. Three-point and four-point flexural tests were conducted, and damage and fracture processes were observed closely. Compared to a cross-ply laminate CFRP specimen, helicoidally laminated CFRP specimens show the increased residual strength after the first peak, and their postpeak energy absorption ratio to prepeak is highly improved. Due to the unique damage processes of helicoidally laminated CFRP specimens, it is found that localized fracture is mitigated with reduced damage depth.

View full abstract

Generally, a multibody dynamics tool is used for railway vehicle running analysis, and the contact force between wheel and rail is generally given by a theoretical equation derived from contact theory. However, since the contact shape is assumed to be elliptical in this theory, the application of running on discontinuous rails such as rail joints is limited. In the running experiment, the contact force is obtained by converting from a strain in the vehicle wheel-set. When the discontinuous rail and the wheel contact each other, there is a possibility that the measurement accuracy may be deteriorated because the contact position and the number of contact points are different from the assumed.
In this research, we constructed a numerical analysis model using the explicit finite element method considering the dynamic interaction between the railway wheel and the track that can be applied to the running of the discontinuous part of the rail. As a result, the contact force analyzed by the constructed model is good agreement with verification tests and the validity of the constructed model was confirmed.

View full abstract

Cracks frequently occur around welded part in degradation problem of steel structure. The inspection of crack damage is carried out by visual inspection and magnetic particle examination as needed. However, it has disadvantages that depend on inspector's skill in visual inspection and is impossible to detect because of accuracy of detection and deal with internal cracks and defects in non-destructive inspection. Therefore, I investigated method of-confirming the presence or absence of cracks as a way that inspector with little experience and skill check cracks certainly by calculating rate of X-axis and Y-axis stress around cracks. In this paper, I carried out vibration fatigue test using actual size specimen, based on the result, I investigated by analysis models made with 3D FEM general-purpose analysis software.

View full abstract

In the railway bridge design standards, the allowable stress design method has been used. Thereafter, the limit state design method and the performance based design method were introduced in 1992 and 2009 respectively, for optimizing section design of railway bridge structures. For the purpose of reducing the construction cost, it has been a front-burner issue to investigate the different sections (e.g., compact section, non-compact section, or slender section) proposed in AASHTO and Eurocode etc. On this background, this paper investigated different design methods for steel-concrete composite railway bridges by using theoretical approach and finite element methods. Based on the results obtained in this study, when compact section was used,around 10% weight reduction of structural steel was confirmed.

View full abstract

A reliable estimation of surface fault displacement, including secondary fault displacement, is necessary for ensuring the safety of nuclear power plants. We developed a high performance computing finite element method (HPC-FEM) to evaluate the fault displacement. Probabilistic evaluations are required for estimaing the surface fault displacement because the solution of fault rupture process changes significantly depending on variability of input conditions. In this study, HPC-FEM is applied to the probabilistic evaluation with uncertainties of material properties. It is shown that HPC-FEM is able to compute the probabilistic distribution of the response in a practical time. The uncertainties of the material properties strongly in uence the critical base slip that makes the rupture reach the ground surface.

View full abstract

This paper reports seismic response analyses about the spherical gas holder which requires higher safety focusing on the ultimate limit state of the level 2 earthquake. In the analysis, a gas holder and pile foundation are modeled in 3D condition to evaluate their interactive behavior more realistically using a soil-water coupled finite deformation analysis code GEOASIA.
As a result, in the holder to be analyzed this time, all the piles lose the bearing function, and the piles have an insignificant effect on the response of the holder. Therefore, the ultimate limit state, which we have considered the maximum unequal settlement in the past examination, is assumed considerably safe. Also, under the greatest possible earthquake condition with this holder, it is possible to almost fully evaluate the seismic performance of the ultimate limit state by 2D analysis. On another front, 3D analysis should be applied for non-uniform condition such as a substandard earthquake input and irregular surface layers. Furthermore, by using this analysis method, it is possible to evaluate the structural stability of the gas holder integrally.

View full abstract

This paper discusses about on-site long term monitoring static and moving vehicle experiments on a composite twin girder bridge, and also investigates validity of an FE model constructed utilizing assumed structural properties from existing inspection documents as well as a visual inspection. The feasibility of a particle filter aided FE mode updating was also examined. In the model updating, the frequencies for the first and second bending modes were considered as design variables in the cost function.Those identified parameters from the model updating were compared with those from material experiments of which specimen were obtained formt the observation bridge. Observations showed that the roller support was not functioning due to deterioration. The updated FE model led to comparable results with those from field experiments.

View full abstract

The displacement measurement approach by high resolution / high speed video camera and image processing have the possibility to realize simple and detailed railway bridge deflection measurement without target. However, its measurement accuracy has not been verified in detail so far. This study verified the accuracy of multiple points measurement of non-target video measurement method for micro displacement through experimental tests of model bridge. As a result, it was clarified that displacement of approximately 0.03 pixel (full amplitude) can be measured excepting for the white spot. In addition, the applicability to practical were estimated based on the past actual measurement results and experimental results. As a result, it was suggested that deflection measurement will be possible up to a measurement distance of 50 m if the bridge has over 20 m span and it is not at night. Finally, by a field test was conducted and the results indicated that the non-target optical method is possible to measure the deflection to the train passing with approximately same accuracy as the optical one with target and U Doppler I.

View full abstract

In this study, we calculated with FEM analysis the changes in the frequency response of the vibration of the ballast layer and road bed due to application of the resilient sleeper in the frequency range of less than 200 Hz, which has a big influence on the ballast settlement. The results show that the use of the under sleeper pads (USP) with less than 0.23 N/mm³ of bedding modulus decreases the natural frequencies and the peak values of two modes: the one in which the sleepers and the ballast layer vibrate with the equal phase and the other in which they vibrate with the opposite phase. These changes lead to a vibration reduction of the ballast and road bed in the frequency band including these modes. Furthermore, we measured the road bed acceleration during a train passing in the section equipped with the resilient sleepers, and confirmed the reduction of the road bed acceleration by 3-8 dB in the frequency band of 60-150 Hz including the opposite phase mode.

View full abstract