In the present study, we address a suspension composed of magnetic cubic particles that flow in a Hagen-Poiseuille flow in a gradient magnetic field generated by multi-pairs of magnetic dipoles. Brownian dynamics simulations were performed in order to investigate the dependence of the trapping characteristics on a variety of factors such as the magnetic particle-particle interaction strength, the strength of non-uniform magnetic field, the strength of the flow field and the separation distance between two pairs of magnetic poles. The main results obtained here are summarized as follows. For the case of a larger separation distance, an arch-type cluster with more cubic particles is formed between the magnetic poles. In this case, the central part of the large arch-type cluster is located in the area where there is less influence from the magnetic field. This in turn leads to an instability of the cluster and as a result it tends to be carried downstream by the stronger flow field. These results suggest that an increasing the pole separation facilitates the trapping of a larger arch-type cluster between the poles. However, the stability of the trapped cluster decreases with increasing separation, and beyond a certain distance, the cluster is carried downstream by the flow field. This behavior implies a trade-off between the trapping capacity and the structural stability of the cluster. If the separation distance between the two poles is sufficiently near, a small arch-type cluster is located in a vicinity area nearer to the wall surface where the flow field is weaker. Hence, we expect that the smaller pole separation may give rise to good trapping performance in the situation of a strong flow field.

In this study, we developed a high-speed sample sorting system by integrating a previously developed Cartesian-arm-based high-speed sample sorting mechanism with an inlet for bulk-loaded sample tubes. The system enables automated sorting of bulk-loaded sample tubes into dedicated containers. To evaluate the impact of this system on test results, we first measured the acceleration applied to the tubes using sample tubes equipped with built-in accelerometers. We then compared the analytical results of blood samples from the same volunteer when the tubes were manually sorted and placed into analysis racks versus when they were sorted and placed using the developed system. This comparison was conducted to determine whether the sorting method affects the accuracy of the test results.

This study investigates the factors contributing to the formation of irregular troughs occurring in nonwoven fabrics during the Roll-to-Roll transport process and proposes a quantitative evaluation method. Unlike uniform films, nonwoven fabrics exhibit localized stiffness variations due to fiber density distribution, potentially leading to irregular buckling. First, a correlation was observed among the grayscale values of the fabric image, constant-pressure thickness, and basis weight, proposing a method to estimate stiffness distribution from optical properties. Next, the three-dimensional shapes of troughs during transport were quantified using the light-section method. Experimental results indicated that irregular troughs tended to concentrate in local high-stiffness regions flanked by low-stiffness regions. It is presumed that the formation of these troughs is partly attributed to the following process: significant strain in low-stiffness regions induces compressive stress in adjacent high-stiffness regions via the Poisson effect, triggering buckling. This hypothesis was supported by Finite Element Method (FEM) simulations modeling the spatial thickness distribution, which showed compressive stress concentration in thicker areas. Furthermore, a novel line-scanning system utilizing Digital Image Correlation (DIC) was developed to visualize fiber texture during transport, demonstrating effective monitoring capabilities. These findings may assist in predicting defect risks based on thickness distribution, contributing to advanced process control.

The critical wind speed of overturning, evaluated using the RTRI detailed equation, is an important index for assessing railway vehicle safety under crosswinds. In this equation, the lateral vibration acceleration of a car body is used to determine the lateral inertia force acting on the vehicle. However, the conventional assumption of the lateral vibration acceleration, derived from running tests under calm conditions, does not necessarily correspond to that at the car body’s center of gravity just before overturning under strong winds. This study reexamines the assumed equation for the lateral vibration acceleration for a typical commuter-type vehicle, based on vehicle dynamics simulations using track alignment data under various maintenance conditions at a representative running speed of 120 km/h. The simulations clarified a strong correlation between the standard deviation of 10 m-chord versine and the maximum lateral vibration acceleration, enabling evaluation according to the track maintenance condition. Based on these results, new assumed equations expressed as quadratic functions of speed were proposed for each maintenance rank. The critical wind speeds of overturning calculated with these equations were generally higher, by up to about 3 m/s, than those obtained using the conventional assumption under the same vehicle and track conditions assumed in this study, indicating that the proposed equations tend to give higher critical wind speeds under many conditions while remaining consistent with the framework of the RTRI detailed equation and with the assumptions adopted in this study.

High-manganese cast steel is widely used in railway crossings due to its high toughness, wear resistance, and exceptional work-hardening capability. However, the mechanisms of work hardening and crack initiation under low slip ratios have not been fully clarified, as most previous studies focused on higher slip ratios that do not reflect actual railway operations. This study aims to elucidate the fundamental mechanisms of work hardening and damage formation in high-manganese cast steel under low slip conditions by employing a twin-disc rolling contact fatigue tester. Experiments were conducted using SCMnH3 rail specimens and SSW-QS wheel materials under a contact pressure of 900 MPa with slip ratios of 0% and 0.22%. Rolling cycles were varied up to 500,000. Results revealed pronounced wear and microcrack formation under low slip, whereas no-slip conditions produced negligible wear. Microstructural observations showed deformation bands in both conditions; however, low slip produced dense deformation twins extending deeper beneath the surface, increasing in density and depth with rolling cycles. EBSD IQ and IPF maps confirmed significant strain accumulation and crystallographic reorientation near the surface, accompanied by plastic-flow-like features that governed crack initiation and propagation. Hardness measurements indicated substantial work hardening under low slip, reaching approximately 590 HV with saturation beyond 300,000 cycles. XRD analysis revealed broadened diffraction peaks and localized α’-martensite formation within the extreme surface layer, suggesting transformation-induced hardening. These findings demonstrate that work hardening under low slip is primarily driven by the accumulation of deformation twins and dislocations, complemented by localized martensitic transformation. Crack initiation occurs along regions of concentrated plastic flow, consistent with damage patterns observed in actual manganese crossings. The study provides essential insights into the deformation and damage mechanisms of high-manganese cast steel under realistic wheel–rail contact conditions.

This paper proposes a non-contact method for detecting deviations in pantograph aerodynamic lift using contact wire uplift measured by general-purpose cameras installed along the railway line. Vertical displacements of the contact wire were simultaneously measured at multiple dropper points in running tests using a current collection test facility in RTRI. The maximum and time-integrated values of the vertical displacement were extracted as feature variables and analyzed using a Self-Organizing Map (SOM), an unsupervised learning method capable of visualizing the distribution structure of normal operating conditions. To quantitatively evaluate deviations from the normal state, a distance-based index was defined on the SOM by averaging the Euclidean distances between a target data point and the remaining data points, with multiple training trials conducted to reduce stochastic effects. The analysis simulated a practical scenario in which a large amount of normal-condition data is accumulated, followed by newly acquired data. The results show that pantograph lift conditions deviating by approximately ±20 N or more from the standard state exhibit the maximum evaluation distance, indicating clear deviation from normal behavior. Even when highly discriminative dropper-point information was excluded, large deviations remained detectable. These findings demonstrate the potential of the proposed approach for continuous, ground-based monitoring of pantograph lift conditions without direct sensor installation on vehicles or overhead contact lines.