In this study, the bending strength of lattice-filled rectangular tubes under three-point loading has been studied using a nonlinear finite element analysis. The failure mechanism of lattice-filled tubes can be classified in four modes: tube wrinkling, tube yielding, core shear and tube denting, which are basically similar modes for the lattice sandwich panels. Based on this fact, the analytical equations for predicting the failure load for lattice tubes are proposed, and their accuracies are discussed by comparing the failure loads yielded by a finite element analysis with their predictions for the tube wrinkling, yielding and core shear modes.

The thermal shrinkage technique, which uses shrinkage strain and shrinkage zone to calculate welding distortion, is expected as a simple simulation to predict the welding distortion of large weld structures. In previous studies, parametric study was performed for bead-on-plate welding and optimum input data settings was clarified, which were the shrinkage strain of −0.012 and a shrinkage zone in which the maximum temperature reached 500°C or more for structural steel. In this research, the thermal shrinkage technique with established input data settings was applied to a construction machinery structure to validate whether the input data settings obtained in the previous study for test specimens is applicable to large weld structures. The analytical model is a construction machinery structure with dimensions of 3.9 m in length, 2.7 m in width, and 0.9 m in height, and has a 40-pass fillet welded joint. The welding distortion analysis using thermal shrinkage technique was conducted with the optimum input data settings. Thermal elastic-plastic analysis considering moving heat source was also performed to compare the numerical results by the thermal shrinkage technique and thermal elastic-plastic analysis. Numerical results revealed that the displacement distribution after 40-pass welding of the construction machinery structure obtained by the thermal shrinkage technique was in good agreement with that obtained by thermal elastic-plastic analysis. A more detailed evaluation showed that the displacement in each welding pass obtained by the thermal shrinkage technique accurately reproduced that obtained by thermal elastic-plastic analysis. The calculation time of the thermal shrinkage technique was 1/7 for the construction machinery compared with thermal elastic-plastic analysis, which clarified that calculation time could be shortened. Our results show that the thermal shrinkage technique with optimum input data settings could be applicable to large weld structures to predict welding distortion.

An advanced thermal barrier coating development is demanded for achieving the next generation gas turbine with a high efficiency. As one of many candidates in TBCs, RTa₃O₉ in a perovskite-type oxide attracts attention owing to the low thermal conductivity. In this study, thermal and mechanical properties of YbTa₃O₉ deposited by atmospheric plasma spraying process as a typical GT coating process were investigated via a high-temperature exposure test. The results obtained in this study are as follows: (1) High temperature exposure tests were carried out on freestanding ceramic sample (TC sample) which was extracted chemically from TBC sample. It was found that porosity of the TC sample was not changed significantly with exposure temperature under the fixed exposure time 100h. (2) Young's modulus of the as-sprayed TC sample was 23 GPa, and the associated thermal conductivity was 0.6 W/mK. Those material properties were increased monotonically with exposure temperature under exposure time 100 h, in spite of the porosity having no influence of the exposure temperature. (3) X-ray crystallographic analysis revealed that the crystallization of YbTaO₄ caused significant shrinkage of TC material which leads to increasing thermal conductivity. It is therefore desirable to stabilize the microstructure at high temperatures in the future.

In the present study, we have attempted to investigate a feasibility of exhibiting a significant trapping performance by means of multi-pairs of magnetic poles. We here focus on the two representative types of configurations using two pairs of magnetic poles. That is, the first is a vertical configuration of two pairs of magnetic poles, where each pair of magnetic poles is arranged perpendicular to the flow direction, and the second is a parallel configuration, where each pair of magnetic poles is arranged along the wall surface of a circular cylinder. As in the previous study for one pair of magnetic poles, we have here investigated the behavior of magnetic rod-like particles in a Hagen-Poiseuille flow, by means of Brownian dynamics, in the above-mentioned two pairs of magnetic poles, which induce magnetic forces to function to trap magnetic particles around the magnetic poles. In the case of the vertical arrangement of the two pairs of magnetic poles, a largely-packed aggregate is trapped between the first pair of magnetic poles, and this trapping characteristic is reproduced between the second pair of magnetic poles. In this case, these large clusters are mainly formed in the center area of the cylinder perpendicular to the flow direction. In contrast, for the case of the parallel arrangement, a largely-packed cluster is formed between a pair of magnetic poles and located along the wall surface. Different from the previous arrangement of magnetic poles, this largely-packed cluster may sufficiently survive and be trapped between the poles by shifting the position nearer to the wall surface. From these results, we suggest that the parallel arrangement of magnetic poles gives rise to a better trapping performance.

A prediction model of the fluid force on a cylindrical blade driven by the Necklace vortex is developed and verified by experiments when the yaw angle is given. Based on the measurement results of the velocity ratio of blade and wind velocity and the fluid force coefficient at zero yaw angle, a prediction model for the relationship between the rotation angle of the cylindrical blade and the fluid force coefficient considering the yaw angle was derived from the analysis of the two-dimensional velocity components acting on the cylindrical blade. The validity of the predictive model was demonstrated by experimental verification. From the comparison, it was found that there was a time delay in the formation and disappearance of the Necklace vortex. Based on the experimental results, the characteristics of the fluid force acting on the cylindrical blade were classified into five angular regions and based on the characteristics of the effective velocity ratio considering the yaw angle, the classification was made into 12 angular regions. Depending on the yaw angle, wind speed, and rotational speed of the blade, all of these regions may appear or only some of them may appear. Therefore, the yaw angle characteristics of the cylindrical blade wind turbine driven by the Necklace vortex are very complicated.

Conventional single-effect absorption chillers require an 80~90 ℃ heat source to supply 7~8 ℃ chilled water for air conditioning applications. However, it is difficult to recover lower temperature heat sources such as 60 ℃ hot water with single-effect chillers. To improve the low-temperature waste heat recovery, we develop a low-temperature-driven absorption chiller that adopts a vapour exchange (VX) double-lift absorption cooling cycle. We developed a proof-of-concept prototype for this cycle and conducted its performance tests in this study. As a result, it delivered chilled water at 7 ℃ when driven by approximately 60 ℃ heat source hot water inlet and 30 ℃ cooling water. We also confirmed the reliability of this result based on the heat balance of the experimental data, according to the AHRI standard. However, the cooling heat quantity (Q_Ew) was approximately 12% lower than the calculation. So, this report also discusses a potential cause of the performance decrease under the assumption that there was internal heat loss in the prototype, considering its external heat balance was less than 2.0%. We compared Q_Ew with the heat discharge from the absorber (Q_Aw) and the condenser (Q_Cw), and found the internal heat loss from the absorber to the evaporator. Its heat transfer mechanism was confirmed by possibility estimating calculation. Then we assimilated this internal heat loss as a part of the cycle simulation program and computed the chilled water temperature characteristics of the prototype. The simulation results well agreed with the experimental results by considering the internal heat loss. This result proved the above assumption of the internal heat loss and thus the reliability of the experimental results.

This paper proposes a novel sensor-less force control method by an underactuated wire-driven robotic finger. This robotic hand contains Twisted round-belt actuator that yields contraction forces produced by twisting an elastic round-belt. This mechanism provides high-resolution control of both the force and joint angle in manipulation tasks. First, we explain the basic mechanism of the articulated two-fingered robotic hand having Twisted round-belt actuator, and mention its features by some experiments. Second, we show experimental results of round-belt contraction force on the basis of two parameters such as number of twist and tension of the belt, resulting in fingertip force estimation which is applicable to sensor-less force control. Finally, we apply this estimation method to the robotic hand, and experimentally demonstrate secure force control in sensor-less configuration at the fingertip.

This paper proposes a dynamic analysis method for evaluating the vibrational behavior of a parallel robot having spherical joints with clearance. By using variables that represent the relative displacement between the elements of clearance joints as generalized coordinates, it is enabled to implement a general algorithm regardless of the combination of spherical joints with clearance. The constraint equations are eliminated from the set of differential algebraic equations to be solved, and therefore the method does not need to use Baumgarte constraint stabilization method. The method was applied to a DELTA robot and were examined by measurement with a real machine. The acceleration of output link when using the special trajectories, which are likely to cause the impact only at a single spherical joint at the beginning of motion, was investigated. From the results of simulation, several motion modes of joint elements due to the clearance are confirmed; free flight, impact, and continuous contact modes. The acceleration of the output link fluctuates significantly at the time when the joint elements are in the impact mode, which shows the fact that the impact of joint elements at the spherical joints with clearance causes vibrational behavior of the robot. The appearance of the acceleration fluctuations are coincided with that of experimental results, which shows the effectiveness of the proposed method as a method for analyzing the vibrational behavior of parallel robots caused by joint clearance.

Building mass dampers (BMDs) are structures designed to mitigate the responses of super tall buildings to earthquakes. These BMDs contain a soft layer at relatively high points in such buildings, and the component above the soft layer acts as a tuned mass damper (TMD). Given that BMDs utilize a part of the buildings as the tuned mass, a large mass ratio can be yielded even in the case of super tall buildings. Therefore, BMDs are expected to demonstrate very good seismic performance against large earthquakes. However, for BMDs to be applicable to these buildings, the initial stiffness of the soft layer must be increased from the optimum level, such that excessive displacement of the layer owing to strong winds can be prevented. This stiffness is larger than the optimum stiffness applicable in the case of small and medium earthquakes. Therefore, a problem arises in that the seismic performance of these BMDs is reduced against small and medium earthquakes that occur more frequently than the large ones. From this standpoint, we installed an active mass damper (AMD) on top of a non-optimal BMD, in which the soft layer's stiffness and damping were increased above the optimal levels. We also designed this AMD such that the seismic performance of the non-optimal BMD was close to that of the optimal BMD, and the habitability during small and medium earthquakes was improved, while explicitly considering the constraints of the AMD through model-referenced predictive control. Ultimately, to demonstrate the effectiveness of the proposed method, we conducted a numerical analysis using input waves with various frequency components.

In a power assist robot driven with an artificial muscle, an accurate artificial muscle model is required to design a robot. It is desirable that the model can be calculated without adjustment parameters based on the actual characteristics of the artificial muscle. The characteristics of various artificial muscles mounted on the robot must be actually measured to identify the adjustment parameters. The above explanation is the reason why the accurate artificial muscle model without adjustment parameters is required. The purpose of this study is to improve the accuracy of the artificial muscle model by modeling the hysteresis characteristics of the artificial muscle based on the parameters such as material properties and geometric conditions. In this paper, we propose the hysteresis characteristic model of McKibben type artificial rubber muscle focused on the end shape of the artificial muscle. It is considered that the hysteresis is caused by the shearing force between the rubber surface and the fiber sleeve at the muscle end owing to friction. In the proposed model, the hysteresis characteristic is modeled based on the physical properties and shape parameters of the artificial muscle. The accuracy of the proposed model is confirmed by comparing with the actual characteristics of artificial muscles made with the same dimensions as the model.

Because mechanical parts will be broken by high acceleration caused by impact of collision, a shock reduction mechanism takes an important role for mechanism design. Springs and dampers have been often used for a shock absorber, however, because of their mass, the impact is transmitted to the body. Momentum exchange impact damper (MEID) and mechanical singularity will be effective mechanism design. The former transfers momentum from body to damper-mass to reduce the body velocity. The latter avoids impact transmission. This paper proposes an impact mitigation mechanism inspired by break-fall in Judo. The mechanism exchanges momentum without additional sensor nor actuator, and avoids impact transmission by singular configuration. It is shown that the proposed mechanism has 1-DOF by Kutzbach-Gruebler’s equation using a planer model and two singular configurations by singularity analysis. Dynamics simulation confirms the concept of the proposed mechanism (Momentum exchange). And a prototype is designed and motion captured by a high speed camera to evaluate the momentum exchange.

In order to clarify the effect of thermal treatment on wear resistance and on corrosion resistance of Ni based alloy 690 which employed to pipe material of steam generator in pressurized water reactors, impact fretting wear properties and polarization characteristics were investigated using an impact fretting test rig which is connected to a potentiostat and can control load and amplitude with high precision. As a result, it was confirmed that the thermal treatments improved the corrosion resistance on the polarization test, but it had almost no effect on the impact fretting wear. Since the current obtained by the potentiostat, the anodic dissolution (corrosion) at the fresh surface had a significant effect on impact fretting wear under anodic potential. Furthermore, according to Faraday's law, the dissolution volume of impact fretting test V_f estimated from the current obtained by the potentiostat almost agreed with the wear volume V_g calculated from the surface geometry.

In the sale of steel materials such as tool steel, it is important to optimize the size of standard stocks to reduce end material and cutting costs resulting from number of cuts and cutting area. If there is difference between the size of standard stocks and the sales size of steel, excess cutting and end material occur. So, the size of standard stocks is important to reduce end material and cutting cost. However, in the conventional method, stock manager analyzes past orders and determines the size of standard stocks which is not the optimum. Therefore, we developed a method to determine the size of standard stocks that reduce cutting costs and end material by (1) creating candidates for standard stocks, (2) giving evaluation score to all combinations of past orders allocations to standard stocks candidates and (3) sorting out optimum standard stocks by mixed integer programing so that total evaluation score is high. As a result of the computational experiment, the development method reduced 2.7% number of cuts, 17.3% cutting area, and 6.3% end material volume compared to the conventional method.

The effects of surface roughness on the molecular gas lubrication characteristics have been studied for defining the application limits of Molecular gas film lubrication (MGL) equation against the roughness. The applicability has been evaluated by comparing the difference of characteristics calculated by MGL equation and direct simulation Monte Carlo (DSMC) method. In this paper, the evaluation for the two kinds of surface roughness was performed to check the influence of roughness shape. The one of roughness has the sinusoidal shape and the other has the triangular shape. It comes evident that the magnitude of transverse flow velocity in the gas film has close relationship with the slope of surface roughness, by comparing the flow pattern for two roughness shapes. The transverse flow, which is not included in MGL equation, causes the difference between the calculated results obtained by MGL equation and DSMC method and, thus, the slope of surface roughness can be the index of applicability of MGL equation. It is clarified that the applicability declines in the case of large roughness slope. The roughness slope, which gives the limit of application, is large for smaller Knudsen number.

If an abnormality occurs while walking and a fall occurs, elderly people with reduced muscle strength and bone density may become bedridden in the worst case due to a fracture of the neck of the femur. The purpose of this study is to develop an algorithm that detects late swing using only from the acceleration in the direction of travel for the purpose of preventing falls during walking. In this study, the late swing detection algorithm is proposed to detect from the rate of change of the traveling direction acceleration. We used motion capture system to confirm whether the discrimination late swing is detected the point between the maximum speed of the foot and the heel strike. From the results, the late swing in 88% to 98% of the gait cycle was detected using the proposed algorithm.

Focusing on the difference in wind speed due to different positions in a wind screen, we devised a simple method to recognize wind noise only by the difference in sound pressure level between two microphones placed in the wind screen. First, we conducted ventilation test with the wind screen, and confirmed that the wind speeds at the two different locations differed by 60 to 70%. Next, a loudspeaker test was conducted to examine the wind screen insertion loss and the effect of ground reflections on a microphone located 30 cm above from the ground. As a result, it was confirmed that the wind screen has almost no effect of 0.4 dB or more on the measurement of sounds below 50 Hz. Finally, a field experiment using a loudspeaker and natural wind was conducted to verify the discrimination performance of the devised method. We observed a sound with a certain sound power of 5 Hz using a loudspeaker, and tried to discriminate wind noise from the difference in sound pressure level between the two microphones placed in the wind screen. As a result, for example, when the discrimination threshold was set to a level difference of 2 dB, 89% of the data was discriminated correctly as wind noise, and 6% of the total number of data was recognized incorrectly. From those results, it was confirmed that wind noise can be easily and automatically discriminated by using the level difference from the results measured by the two microphones in the wind screen.

For improving safety and shortening the working time in payload transporting using an overhead crane without depending on the skills of the operator, we examined a control method for automatically suppressing payload sway. With the conventional control method, the residual payload sway, which is the payload sway after the crane stops, can be suppressed, but the overshoot, which is the moving distance of the payload after starting the stop operation, increases. We propose a payload sway suppression control method using simulated repowering control that simulates a repowering operation by a skilled operator. With this method, after the stop operation starts, the crane performs deceleration and subsequent trapezoidal waveform velocity drive. The parameters of the trapezoidal waveform are determined as follows. The payload sway caused by the deceleration is negated by the trapezoidal waveform velocity drive. In addition, the velocity and acceleration of the trapezoidal waveform to a range not exceeding the maximum value, to shorten the time width of the trapezoidal waveform. We conducted a simulation to verify the effectiveness of the proposed method. We confirmed that it can shorten overshoot compared with the conventional control method by suppressing the residual payload sway to various values of the pendulum length, which is the length from the rotation center of the rope to the center of gravity of the payload. We also confirmed that the proposed method does not worsen the residual payload sway and overshoot, even if the stop operation begins at various times, such as during acceleration, constant velocity movement, and deceleration.