The stress intensity factors (SIFs) for an interfacial corner, which are compatible with that for a crack, can describe the singular stress field around the corner. However, due to the difference of the singular order with the angle of a corner and the combination of materials, we cannot compare the SIFs of interfacial corners to that with different singular orders. In this study, we estimated the fracture toughness (critical SIF) of jointed interfacial corners with different opening angles and different combinations of materials using the molecular statics method and cohesive zone model. The critical SIFs were normalized using those of interfacial cracks with the same combinations of materials. Regarding the relationships between the normalized critical SIF and the singular order for the numerical experiments using both the molecular statics method and cohesive zone model, the normalized SIF increased exponentially with decreasing singular order. The relationship between the singular order and the critical SIF obtained by the molecular statics or the cohesive model was almost unique regardless of the combination of materials and jointed angles. In other words, it is possible to translate the critical SIFs of a jointed interfacial corner to that of interfacial crack. However, the slopes of the relationship obtained by the molecular statics and the cohesive zone model did not correspond. This might be because only the molecular statics method considers the influence of crystal structure factors such as slip and dislocation. The interactions between atoms of the molecular statics and cohesive zone model along the interfaces are different. The relationship between the singular order and the critical SIF of actual jointed interfacial corner must be investigated experimentally in further research.
Numerical simulation using DNS (direct numerical simulation) were performed for multiple pulsating jets consisting of two jets with different phases of pulsation. The calculations were performed for multiple cases where the jets with Re = 1500 were arranged at short or long distances from each other, and had pulsations with St = 0.3 and 0 (in-phase), π / 2, π (anti-phase) phase differences respectively. In addition to evaluating the instantaneous vortex structure of the jets and the average contour in the cross section along the mainstream direction through the centerline of the two nozzles, we introduced entrainment and mixing parameters to evaluate quantitatively the average characteristics of the jets over the entire calculation area. For the jets arranged at short distance, the jets with in-phase pulsations caused flow switching, and entrainment and mixing were suppressed compared to the case without pulsations, on the other hand, the jets with anti-phase pulsations were expanded in the jets arrangement direction and the entrainment and mixing were improved. The jets with π / 2 phase difference caused flow vectoring to the side where the phase was delayed by π / 2. On the other hand, when the jets were arranged at long distance, switching or expanding flow did not occur for the jets with in-phase and anti-phase. However, for the jets with π / 2 phase difference, the jet flow expanded in the jets arrangement direction, and improving entrainment and mixing. These findings suggest that adding pulsations other than in-phase or anti-phase to the multiple pulsation jet can further improve the flow entrainment and mixing characteristics. These phenomena are due to the interference between the vortex rings formed by the two jets, and it’s state seems to be determined by the strength of the vortex rings and the positional relationship between them. Therefore, by adding a stronger pulsation, it can be expected that even if two jets are arranged at long distance, a flow similar to that when the jets are arranged at short distance will occur.
The thermodynamic suppression effect of cavitation occurs in cryogenic fluid, hot water, and Freon, which suppresses cavity development. In this study, the purposes are to clarify the characteristics of the thermodynamic suppression effect in hot water up to 140 ℃, where the thermodynamic suppression effect is expected to be comparable to that of liquid nitrogen or liquid oxygen, and its relationship with the scale effect, which promotes cavitation. To detect the small variation of the temperature inside the cavity, the thermistor was calibrated by using the secondary standard platinum resistance thermometer. The cavity length and temperature depression inside the cavity were measured as indicators to evaluate the thermodynamic suppression effect quantitatively. By increasing mainstream temperature under constant mainstream velocity, the cavity length was unchanged up to 100 ℃. These results can be attributed to a mixture of the thermodynamic suppression effect and promotion effect due to the scale effect caused by the change of Reynolds number. As for temperature depression, it was found that temperature depression was increased as mainstream temperature increased even when the thermodynamic suppression effect and promotion effect due to scale effect are mixed. By increasing mainstream temperature under constant Reynolds number, the scale effect was taken away to some extent and the cavity length was shortened at 100 ℃.
Liquid fuel-only injection into combustion air is widely used in gas turbine combustors for power generation, particularly in large capacity type. This combustion air goes by the name of ‘crossflow’. For the liquid fuel-only injection, breakup process, trajectory of liquid jet and droplet sizes have been extensively investigated. However, there remain much space to be devised for improvement in atomization. In the present study, twin-fluid injection is introduced instead of liquid-only injection into a gas turbine combustor, where atomization air is supplied through the side port and mixed with liquid in the mixing port at right angle, and two phases mixture is injected into the crossflow. A simple-structured twin-fluid atomizer is designed. An experimental study was conducted to compare the behaviours between liquid-only and twin-fluid injected transversely into a crossflow. High-speed photography technique is used to clarify the atomization characteristics of jet/spray. One feature of twin-fluid injection is that annular liquid film appears at the exit of the mixing port, and the film easily disintegrates by aerodynamical force of the crossflow. The jet/spray of the twin-fluid injection is not blown away easily by crossflow, and flow across the crossflow. Furthermore, the jet/spray of the twin-fluid spread and droplets disperse more widely in the crossflow. In addition, the twin-fluid injection produces numerous fine droplets and few coarse droplets.
Electronic equipment cooling devices need to be installed inexpensively in high-density packaged equipments such as 1U-height servers. As a cooling device, a thermosyphon has attracted attention from the viewpoint of its cooling performance and cooling power saving. For cooling CPUs, thermosyphons especially made of copper with water coolant have been studied so far. From the viewpoint of cost and weight reduction, however, a thermosyphon need to be made of aluminum. And by optimizing the boiling heat transfer surface profile of a thermosyphon, CPUs can be operated stably at a lower operating temperature by reducing the wall superheat. Therefore, in this study, we have made the aluminum boiling heat transfer surface that has a porous structure formed by micro-curl skived fins to form reentrant cavities in order to enhance a boiling heat transfer performance. We verified the effect to enhance the nucleate boiling heat transfer characteristic of fluorine-based refrigerant HFE-7000 on the aluminum structured surface with micro-curl skived fins up to a heat flux 100kW/m2 (typical condition of CPU cooling), especially dependence of saturated vapor pressure (0.10MPa, 0.14MPa, 0.18MPa) and number of micro pores (467[1/cm2]～1250[1/cm2]) on the wall superheat and boiling heat transfer coefficient. From these examinations, mainly we concluded the following. (1)When the pore number density of structured surface is 833[1/cm2], the wall superheat reduces to below 1K and the boiling heat transfer coefficient enhances to around 100kW/(m2・K) at saturated vapor pressure 0.14MPa. (2)Basically, the wall superheat decreases as the increase of the saturated vapor pressure, however, in case of the structured surface with skived-fins, especially at low heat flux (under 40kW/m2) and high vapor pressure (0.18MPa), it tends to increase. (3)At the pore number density over 833[1/cm2], this tendency becomes more remarkable and at 1250[1/cm2] an instability of the wall superheat occurs.
This paper presents a novel technique of an active control to suppress sheet flutter. In this technique, sheet flutter is actively and non-contactly suppressed by injection and suction air flows through a nozzle. A nozzle is combined with a speaker as an actuator, and this device is drove in a sheet displacement feedback system. PET sheet is used as a controlled object, and the sheet is supported upper end in a wind tunnel. To show the effectiveness of this technique for sheet flutter suppression, the suppression characteristics are experimentally examined. Then, the suppression effectiveness is evaluated at the increased rate of the flutter velocity with control as compared to it without control. Moreover, air flow patterns around the nozzle and the sheet during control are visualized by the smoke wire method. Lastly, the control effects by injection and suction air flows to the sheet are examined in terms of air flow patterns, frequency response characteristics, and unsteady pressure. As a result, this paper shows the effectiveness of the proposed device and technique for sheet flutter suppression, and the flutter velocity is increased by 50 % with control as compared to without control. In this technique, the injection and suction air flows from the nozzle change the airflow around the sheet, not directly apply momentum on the sheet. Therefore, this technique is an air flow control.
The energy efficiency of the conventional flow control of hydraulic systems using the pressure loss of throttle valves, which are widely used in construction machinery, is very low. If the switching inertance hydraulic system which is a kind of PWM control hydraulic system is used, a significant improvement in energy regeneration efficiency can be expected. In this study, we first show a modeling and calculation method for understanding the effects of differences in valve switching time and pipeline configuration near the valve. We also confirm that the accuracy of the model is good and experimentally show that PWM control hydraulic system can regenerate energy to the high pressure source. Next, we examine the effect of the energy regeneration efficiency of PWM control hydraulic system due to the differences in valve switching time and pipeline configuration near the valve using the previous model. As a result, we confirm that when a relatively long pipeline is added near the valve, waste of kinetic energy and reduction of natural frequency affect the energy regeneration efficiency and average flow rate.
In the previous study, one of the authors investigated stationary response characteristics of linear systems subjected to non-Gaussian random excitation. It was revealed that the probability distribution of the stationary response varies significantly depending upon the bandwidth of the excitation power spectrum. In real engineering problems, on the other hand, the transient response before reaching the stationary state is also crucial. Therefore, in this paper, we explore transient response characteristics to non-Gaussian random excitation by Monte Carlo simulation. The non-Gaussian excitation is prescribed by both a probability density function and a power spectrum with bandwidth parameter. In order to find the response characteristics that appears commonly for various non-Gaussian excitation, we consider four kinds of non-Gaussian distributions for the probability distributions of the excitations. Their shapes are much different from a Gaussian distribution and also distinct from each other. For such non-Gaussian random excitations, we look into the relationship between the transient response characteristics and the excitation bandwidth. Specifically, first, the transient response distribution is examined. Then, we observe the variance, skewness and kurtosis of the transient response to discuss the time variation of the response distribution in more detail.
A method to retain a cask by a retention plate has been developed. In this method, a small fitting gap between the bottom of the cask and the circular hollow of the retention plate controls movement of the cask in the range of the circular hollow area at an earthquake. The fitting of the cask and the retention plate absorbs the seismic energy by the motion of the cask and the collision between them. Characteristics of the vibration behavior of the cask with the retention plate were analyzed by dynamic analysis. As a result of the analysis, the cask does not get out of the retention plate or fall down during an earthquake. The movement of the cask mainly consists of three basic motions, slipping motion, rocking motion, and rotating motion, and reaction force between the cask and the retention plate becomes the maximum in rocking motion. The strength integrity of the retention plate was also confirmed under condition of the maximum reaction force, and the function of the retention plate, which is to keep the cask as standing at an earthquake, was maintained. In addition, characteristics of the vibration behavior of the cask and the design of the retention plate were validated by 1/4 scale mock-up seismic test.
In this paper, we propose a pedestrian trajectory prediction method for autonomous mobile robots. In many cases, there are many pedestrians in the environment in which the autonomous mobile robot runs. In such an environment, the robot needs to run safely for pedestrians. In order to avoid collisions with pedestrians and drive safely, it is important to predict future movements of pedestrians. In the conventional prediction method, the trajectory of a future pedestrian is often predicted from the position of a pedestrian in the past. However, in such cases, it is difficult to predict the movement to avoid obstacles such as walls and pillars around the pedestrian. In this study, point cloud data acquired by LiDAR is used to predict the behavior of pedestrians avoiding surrounding static obstacles. Based on the point cloud data, distances between the pedestrians and the static obstacles is calculated at each time. Then, input it into the network together with the transition of the pedestrian’s position to predict the future pedestrian’s position. In addition, we use attention mechanisms to model interactions between pedestrians. This makes predictions that consider not only static obstacles but also the effects of other pedestrians. The usefulness of this study is shown by performing accuracy evaluation using the dataset created in the simulation environment and the publicly available dataset.
Modal properties such as natural frequencies, modal shapes and damping ratio are useful to understand structural dynamics of mechanical systems. To use the modal properties for structural health monitoring, they need to be estimated under operational conditions. Therefore, operational modal analysis (OMA), extraction of the modal properties without input signals, has been proposed to easily extract the modal properties under operational conditions. Recently, OMA for underdetermined systems, i.e. number of measurements is less than that of active modes, has been paid attention to reduce the number of sensors. This paper proposes the OMA framework for the underdetermined systems based on Bayesian tensor decomposition of second-order statistics data. The proposed method enables us to extract the modal properties from underdetermined systems without tuning the number of active modes because rank of the tensor data corresponding to the number of the active modes is automatically determined via Bayesian inference. To show advantage of the method, the modal properties are extracted from artificial vibration data obtained from a mass-spring system under the operational and the underdetermined conditions.
This paper presents numerical solution to a shape optimization of viscous flow field for stationary fluid-structure-interactive (FSI) fields. In the FSI analysis, a weak coupled analysis is used to alternately analyze the governing equations of the flow field domain and the structural field considering geometrically nonlinear. A minimization problem for total dissipation energy is formulated for the shape optimization of viscous flow field in the FSI fields. Shape gradient of the shape optimization problem is derived theoretically using the Lagrange multiplier method, adjoint variable method, and the formulae of the material derivative. Reshaping is carried out by the H1 gradient method proposed as an approach to solving shape optimization problems. For shape optimization of the viscous flow field in the FSI fields, a new shape update method is proposed to overcome separation and interference of the finite element meshes on the common boundary between the flow field and the structural field. Numerical analysis program for the shape optimization problem is developed by using FreeFEM, and the validity of proposed method is confirmed by results of 2D numerical analyses.
A Scotch yoke mechanism is often used instead of a slider-crank mechanism to obtain high output power. The kinematics of the Scotch yoke mechanism are characterized by elements translating along two orthogonal axes. This is realized by two slider guides placed in parallel in each direction. As a result, the Scotch yoke mechanism achieves high power transmission with high rigidity. A drawback of this mechanism, however, is that it requires high precision in machining and assembly processes. In this study, an orthogonal double-slider mechanism was developed to solve problems caused by machining and assembly errors. With an L-shaped double slider, the mechanical structure is simple and this mechanism has the advantage of lower driving torque compared with a slider-crank mechanism. The developed mechanism is derived from a Scotch yoke mechanism, but the major difference between them is the arrangement of the sliders and slider guides. This study is the first to investigate the effect of the arrangements of slider guides on the driving characteristics of the mechanism. Through measurements of the force acting on slider guides and their deflection, the effect of the slider configuration was experimentally verified as the number of the slider guides was reduced from the maximum of four. In addition, when the mechanism was used in a water pump, the change in shear force acting on the slider guide in the direction of piston translation was investigated as the number of slider guides was reduced. The deflection of the slider guides for the piston was measured under static conditions. Also, the starting force from a stationary state was measured, revealing that the load acting on the mechanism was reduced by removing one of the slider guides arranged in parallel.
In a photovoltaic generation plant, the tilted angle, azimuth angle, and solar arrays’ interval are the main design parameters because they affect the annual amount of received light. Increasing the tilted angle of the solar array and the interval between the front and the back can reduce the shadow ratio, but the number of solar arrays that can be installed decreases, and the total amount of light received drops in the entire facility will decrease. Based on the solar orbit analysis, the amount of solar radiation with shadows on the solar array is determined using the ratio of shadows on the solar arrays. Photovoltaic generation plants will be built on complex shaped sites. It is considered to place the solar arrays at a site in a multiply connected area that contains land that cannot be placed. Because of the weather in the installation area affects the placement of the solar arrays the weighting factors of the amount of solar radiation and the sunshine duration are used. This study proposes a method to determine the placement of the solar arrays that maximizes the amount of received light on a site. This placement problem is formulated as a combinatorial optimization problem and solved by genetic algorithms. The proposed method is able to obtain the optimal value for the solar arrays placement.
The vibration generated at the contact point between a tool blade and a machined surface during cutting significantly influences the generation mechanism of regenerated chatter vibration. However, it is difficult to measure the vibrations generated at this contact point. In our previous research, we focused on the chatter mark formed on a machined surface owing to the chatter vibration during cutting, and analyzed the periodicity of the pattern using a two-dimensional discrete Fourier transform. Furthermore, we proposed a method for inversely analyzing the chatter vibration information generated at the contact point during cutting from the periodicity of the chatter mark, and demonstrated its effectiveness. The proposed method is based on the formation mechanism of the chatter mark estimated from the geometrical relationship between the tool blade and the machined surface, and the validity of this formation mechanism has not been directly investigated. In this research, the movement of the tool blade during cutting was reproduced based on the vibration displacement of the tool shank measured in the cutting experiment. A voxel model cutting simulation was then performed using the reproduced tool blade movement, and the chatter mark was reproduced in the simulation. By comparing this reproduced pattern with the chatter mark obtained by actual processing, the validity of the inverse-analysis model proposed was verified more directly than the previous research.
We develop an output simulator for a baseball-type acceleration sensor in which one three-axis sensor for low acceleration and three sensors for high acceleration are installed. The acceleration vector at any position inside this ball-type sensor in rotational motion is formulated on the basis of the assumption that all the sensors are in rigid configuration. Since this simulator can estimate kinematically the acceleration under the change in time of rotation axis on the sensor coordinate system, we investigate the behavior of the rotation axis suggested in the previous flight experiment. In order to confirm the accuracy of the sensor outputs, preliminary experiment is performed, in which acceleration outputs are measured by rolling the ball-type sensor along a slope. The comparison of obtained acceleration outputs with theoretical values shows sufficient agreement. At the same time, the moment of inertia of the ball-type sensor is measured. Sensor outputs are simulated under two different conditions, in one of which the rotation axis is changing in time and the axis is fixed in the other. As the result, the acceleration output in the former condition varies linearly while the output keeps almost constant in the latter. The measured outputs of the flight experiment successfully reappear in the results under the former condition. The error between the simulated and measured outputs is calculated to 18 trial data of straight and curve balls, and the errors in the former condition become smaller than those in the latter. The present result therefore concludes that the rotation axis of the ball-type sensor in flight changes with time.
Once fully automated driving has been implemented, passengers will be able to take advantage of not only the sitting posture but also standing and perching postures. Physical workload of these postures should be investigated in order to design car seats that provide passengers appropriate physical workload. This study aimed to investigate the feasibility of simulation-based physical workload evaluation of these postures by a musculoskeletal model. Twelve male students participated, and four electromyograms (EMGs) of the rectus femoris, tibialis anterior, gastrocnemius, and erector spinae were recorded with varying seat angle against the horizontal surface: 0° (i.e., sitting), 30°, 45°, 60° (i.e., perching), and 90° (i.e., standing). Results showed that the effect of the seat angle condition was significant for the EMGs of rectus femoris and erector spinae. In particular, the EMG of rectus femoris was relatively higher than the other three EMGs. The EMG of rectus femoris increased with the increase in seat angle in the range from 0° to 60°and decreased in the range of 60° to 90°. The musculoskeletal analysis was carried out with the same condition as the EMG measurement. The average height and weight of the participants were applied to the analysis model, and the muscle activities of the four muscles were estimated. As expected, the trend of muscle activity in the rectus femoris estimated by the musculoskeletal analysis agreed with the experimental result. The average absolute error of muscle activity in the rectus femoris between the experiment and the analysis was approximately 2.5%. From these results, it is suggested that musculoskeletal analysis can be utilized to estimate muscle activity of a spectrum of static posture from sitting to standing through perching.
The variation of the noise metrics of a group of sound sources due to its movement at a constant velocity along a straight track is discussed in the paper. When sound sources move, frequency and amplitude modulation is observed in the radiated sound field. The frequency and amplitude increase when sound sources are approaching and decrease when going away. We found that when frequency and amplitude modulation were considered, the calculated noise levels LAmax, LASmax, LAeq,T became larger than the calculated levels without their consideration. The difference of the calculated levels between the cases increased with respect to the sound source velocity. This difference was independent of distance between the track and the receiving point. Approximating discrete sound sources in line by a finite line source would underestimate the noise level. The level of this underestimation would increase by decreasing source division. The underestimated level was calculated for sources with reference noise spectrum for rolling stocks with a constant sound power level as we focused on the influence of frequency and amplitude modulation on the radiated sound field. The underestimated level was large for LAmax and LASmax compared with LAeq,T. The difference of the calculated levels between the cases decreased monotonically with respect to the source frequency.
It will take a long time to put autonomous driving into practical in city road. Because the reliability of the technology for collecting traffic environment information and making appropriate driving judgment plans is still insufficient. Therefore, the concept, which is called “Shared Control”, has been proposed. It means the cooperation between a driver and a steering assistant system, guides a driver to a proper driving operation by interfering with each other. In Shared Control, there is a case where interference is excessive. This is because there are two intelligent agents in Shared Control, a driver and a Autonomous Driving System (ADS), and each may have different operations. This state in which a driver and ADS have different target operations and the excessive interference is called " confliction ". When confliction occurs, a driver's acceptability decreases and a driver's annoyance increases, and in addition, a driver may be endangered. Therefore, we develop a steering assistant system that gives priority to human and suppress confliction between a driver and ADS. Specifically, by adding a new feedback loop to Shared Control, a target operation of ADS is modified according to the interference state and the operation of a driver. We report the effectiveness of the proposed Human-Centered Shared Control through actual vehicle experiments.