In the fatigue damage assessments in component design, the elastic-plastic strain range is estimated from the elastic strain by using the so-called K_e factor. The K_e factor corresponds to the ratio of the strain range obtained by elastic analysis to that obtained by elastic-plastic strain analysis. The design code of the Japan Society of Mechanical Engineers (JSME) prescribed the K_e values for component design. The values were determined so that analysis results for representative component geometries and loading patterns could be predicted conservatively. Since most of the loads assumed for determining JSME K_e values were caused by thermal transients such as plant start-up and shutdown, the values may not be valid for seismic analyses, for which the load is given by acceleration of boundaries. In seismic analyses, even if the plastic strain becomes significant, the hysteresis property of the stress-strain property may bring about a damping effect. The main objectives of this study were to calculate K_e values, which are referred to as K_s values for seismic loads, and to understand K_s characteristics. The single elbow model or the piping system, which consisted of elbows, straight pipes and a weight, was assumed for seismic analyses. It was shown that, the K_s values were much less than the K_e values prescribed in the JSME code. It was revealed that the damping caused by the hysteresis property suppressed the increase in the strain range. It was found that the damping effect was not the same for each element. The strain becomes relatively large at particular elements such as the elbow and pipe near a fixed end. Since such elements acted as a damper, the strain range and K_s value tended to be relatively large.

Craze cracking is a typical cracking morphology caused by thermal fatigue loading; however, the reason for its cause by thermal fatigue loading is not fully understood. It is known that craze cracks do not grow deeply. In this study, evolution of craze cracking was simulated by Monte Carlo simulations using three-dimensional finite element analyses. Cracks were modeled using the X-FEM technique in Abaqus in order to consider the interaction between multiple cracks on the crack initiation and growth. Craze cracking was observed only when the growth in the depth direction was arrested. Since cracks kept the same depth for a long duration, many cracks could be initiated and craze cracking occurred due to the equi-biaxial stress state. Next, to investigate the change in failure load due to craze cracking, the limit load of a straight pipe subjected to a tensile, bending or internal pressure loading was analyzed. The simulated craze cracking was assumed to be located inside the pipe. It was revealed that the change in the limit loads due to craze cracking could be predicted well by replacing the craze cracking with an equivalent single crack. Although the replacement rule prescribed in the fitness-for-service code of the Japan Society of Mechanical Engineers predict reasonable limit loads, it did not result in conservative prediction. Conservative limit load could be predicted by setting the equivalent crack length to be the maximum extend of distributed cracks.

To better understand high Reynolds number zero-pressure-gradient turbulent boundary layers detailed hot-wire measurements of the mean and fluctuating streamwise velocity are made in RTRI ’s low-noise wind tunnel in Maibara Japan. The momentum thickness Reynolds number is up to approximately 8×10⁴. Using the friction velocity measured by oil film interferometry, mean velocity profile shows a clear log-law region. The turbulent intensity profile also shows the log-law scaling predicted by attached eddy model. However, the log-regions of mean and turbulent intensity do not match with each other. Kármán constant κ, additive constant B, and wake parameter Π are evaluated as 〈κ〉 = 0.385 ± 0.003, 〈B〉 = 4.26 ± 0.15, 〈B₁〉 = −1.077 ± 0.089, 〈Π〉 = 0.455 ± 0.025, and they relate closely with each other by simple equation.

To design a heater that controls a predetermined temperature distribution, the distributions of the heat generation densities for the heater should be designed. Generally, designers adjust the heat generation densities and calculate the temperature distribution by 3D thermal fluid simulations repeatedly. However, this process leads to a long design period because it incurs high simulation cost to achieve the optimum result. To shorten the design period, we have proposed a more efficient thermal design technique that combines the superposition calculation of the reduced order model (ROM), which is constructed from the results of the 3D thermal simulations, and the optimizing algorithm. The ROM decreased the calculation time to estimate the temperature distribution in the heater. Furthermore, we applied the Artificial Bee Colony (ABC) algorithm, which is a type of swarm intelligence, to estimate the optimum combination of the heat generation density distributions of the heater. Our proposed technique has the advantage of being able to calculate the total trial number of the 3D thermal simulations, which was the same as the number of divisions of heater areas. This enabled the designers to estimate the time to achieve the optimum thermal design in advance. Therefore, the proposed technique could be applicable to the thermal design for real products. In this paper, we used this simulation process to design the temperature control unit for a capillary electrophoresis DNA sequencer as an example. The optimizing calculation was completed within only 500 s by using the ROM. The results of the optimization demonstrate that the designed heater controls the three designated and uniform temperature distributions. The measurement results of the temperature distributions of the prototype heater agreed well with the target temperature distributions.

The dynamic contact behavior of a minute liquid droplet upon the collision with a high temperature solid is investigated using total internal reflection imaging. An inkjet water droplet collides with a high temperature surface of sapphire and quartz glass prisms then splashes away. Contact behaviors captured from back using a nanosecond lighting stroboscope vary dramatically with the contact temperature T_c based on the heat conduction theory rather than the solid temperature T_s and are classified into four regions, (I) film evaporation, (II) nucleate boiling, (III) spontaneous nucleation and (IV) supercritical state regions. Contact area decreases significantly in the region (II) to show a minimum at a temperature close to the limit of liquid superheat, then increases in the region (III) to reach a maximum at a temperature close to the critical temperature before decreasing at higher temperatures. Even at a contact temperature so high as to exceed the critical temperature, liquid still contacts the solid surface over a significant area for several microseconds before drying up of the surface. The fine bubbles generated due to spontaneous nucleation hinders the contact due to the formation of the local dried area as the contact temperature approaches the superheat limit, whereas the contact is rather enhanced at further higher temperatures due to the dynamic action of spontaneous nucleation. Similar behaviors are observed for the quartz glass prism in the same range of the contact temperature.

Dual mass flywheel (DMF) dampers are conventionally used for torsional vibration suppression in automobile powertrains. The torsional vibration in powertrains has increased in recent years owing to the widespread downsizing of internal combustion engines to reduce their fuel consumption. Therefore, it is necessary to improve the torsional vibration suppression devices. Centrifugal pendulum absorbers (CPAs) suppress torsional vibrations better than DMF dampers. However, the mass production of the CPAs demands extreme accuracy in order to maximize their performance, and the CPAs are vulnerable to aging deterioration. We focused on a periodic reversal spring (PRS) with permanent magnets as a new torsional vibration suppression device. PRS with permanent magnets has a simple structure and no points of contact. Moreover, mass production of PRS with permanent magnets is easier compared to the mass production of CPAs. First, we studied the structure and operation of the PRS with permanent magnets. Next, we established the output torque and iron loss of PRS with permanent magnets using the three-dimensional electromagnetic field analysis. The characteristics of PRS with permanent magnets depends on the direction of the magnetic fields of permanent magnets. Furthermore, we examined the advantages and disadvantages of multipolarization of PRS with permanent magnets. We simulated the torsional vibrations in powertrains using the DMF damper, DMF damper combined with practical CPA, and DMF damper combined with PRS as the torsional vibration suppression devices. PRS with permanent magnets was made clear to suppress torsional vibration more effectively than the practical CPA.

A rational dimension reduction method based on a new type of complex modal analysis is developed in order to accurately analyze nonlinear vibrations generated in large-scale structures with local strong nonlinearity, non-proportional damping and asymmetric matrix at low computational cost. In the proposed method, first, the linear state variables are transformed into modal coordinates using complex constrained modes obtained by fixing nonlinear state variables. Next, a reduced model is derived by selecting a small number of modal coordinates that have a significant effect on the computational accuracy of the solution, and coupling them with the nonlinear state variables expressed in physical coordinates. In that process, the remaining modal coordinates that have little effect on the computational accuracy are appropriately approximated and integrated into the equations of motion for nonlinear state variables as correction terms. Furthermore, by using a method of estimating the effect of higher-order modes from lower-order modes, the computation of higher-order eigenpairs becomes unnecessary. From the reduced model constructed by these procedures, periodic solutions and their stability, quasi-periodic solutions and chaos can be computed with a very high accuracy and at a high computational speed.

The rotation speed of the ultrasonic motor depends on three parameters such as frequency, voltage and phase difference. In current mainstream drive circuits using a step-up transformer, ultrasonic motors are controlled frequency (rotation speed) by variable value, and voltage (rotation and stop) and phase difference (forward and reverse) by two values. The primary disadvantage of this drive system is that the control input value does not correctly correspond to the output value at extremely low rotation speeds, which prevents precise position control. In this research, we have achieved precise position control by controlling the frequency and voltage by variable values and the phase difference by two values as the control method. In order to realize this position control method, we developed a drive circuit that uses a direct digital synthesizer and a high voltage operational amplifier to generate a sine wave, in place of the step-up transformers typically used for drive circuits. By using these elements in the circuit, it is possible to control the three parameters (frequency, voltage and phase difference) of the output voltage applied to the ultrasonic motor with high speed response and accurate waveform. Using this control method and drive circuit, we performed precise position control within 2.424×10⁻⁶ rad. The features of the developed drive circuit and the control method for changing the three parameters are described.

Robots are becoming more familiar to humans, and soft robot that light and flexible and safety are attracting attention. We designed self-deformation-robot(SDR) for exploration and watching using bending-type-pneumatic-artificial-muscle (BPAM) which capable bending. This robot has a frame composed of BPAM and can move by deforming itself. Since the whole body of this robot consists of actuators, it can move from any posture. Furthermore, it has excellent shock absorption and can act in unknown environments without advanced sensing. First, we developed a robot with a regular icosahedron structure to solve the problems of the conventional regular hexahedron structure. Next, we installed a system on the robot and built a control method so that the robot could move wirelessly. After that, we conducted a running experiment of the robot and investigated whether it could move with the system installed. As a result, we achieved to drive the robot with developed system and controller. However, due to the strength problem of the connection part of BPAM, the output of BPAM cannot be maximized. In addition, the center of gravity of the robot shifted by the mass of the installed system, which making it hard to move. These will be resolved in the future.

In this paper, we present a method for identification of defects in concrete using machine learning. The time history data of acceleration of surface vibration obtained by hammering test is employed as learning data, and a convolutional neural network, which is generally utilized in image recognition is applied to estimate defect shape. Since information of relative position can be held in the convolutional neural network, it appears that this method is suitable for the position estimation of defects in concrete using the information from multiple sensors as input. In addition, a concrete floor plate is represented by a non-dimensional density matrix, so that both defective and healthy areas are treated in a unified manner and can be easily handled by a machine learning model. By delimiting the non-dimensional density matrix to a certain size and estimating the columnar non-dimensional density distribution, the data set can be constructed using the same method from floor plates of different sizes in the x and y directions. We challenge the estimation of 3D topology (position and location) of the internal defects in concrete plates, and some numerical results and considerations are shown in this paper. We also perform preprocessing on the data set and discuss the change in accuracy with and without preprocessing.

On the extensions of the cubic Bézier curve with four control points, to connect multiple segments with required continuity has been strongly intended and for example, tangent and curvature continuity at the start and end points are guaranteed independently by adding extra shape parameters. Contrary to this research trend, κ-curves, which control one curvature extremum on each curve segment instead of the end points, are defined as a sequence of the quadratic Bézier curve with three control points. In this study, in order to extend κ-curves, we propose generalized trigonometric basis functions consisting of (sint,cost,1). Using these basis functions, we also define a new free-form curve named generalized trigonometric curve. We discuss its degree elevation, Miura’s triangle, Gobithaasan-Miura’s recursive algorithm, handwriting and spline.

We developed an “Insight CAD system” as one of the design support technologies. The purpose of the Insight CAD system development is to check design rules automatically on 3DCAD, and to provide information about the violation part or the violation list for designers. In this paper, we report the rule check on the evaluation of assemblability. First, we developed a weldability function and generated two digitized rules. One rule is related to the distance between the welding lines, and the other is related to the insertion space of the welding torch. Next, we generated two rules on the assemblability during bolt fastening. One rule is related to the diameter of the hole into which the bolt inserts, the other is related to the position of the hole. We evaluated the four newly developed rules using a simplified model and found that the violation parts can be identified automatically with respect to the four newly developed rules. This technology can help designers to shorten the check time, reduce rework from manufacturing to design, and improve product quality.

Sagittal split ramus osteotomy is the most commonly performed for jaw deformity patients in oral and maxillofacial surgery (OMS). In recent years, virtual reality-based surgical training simulators have been developed for reducing the cost of simulated bone, however, effective assistance methods for improving surgical skills have not been well studied. Therefore, the purpose of this study is to develop a manipulation assistance method for surgical training simulator with haptic rendering. The surgical training simulator system is composed of three modules: pre-processing module, interactive surgical training module, and evaluation module. In this paper, we newly implement the pre-processing module for correct answer data generating method, interactive surgical training module for enable users to feel the unique reaction force including cortical bone and cancellous bone, and improve the evaluation module for correct answer data. In the evaluation experiments, 4 testees of surgeons and 4 testees of engineering students performed the surgical training for 1 jaw deformity patient. The training consisted of 3 sets of 5 repetitions. As the results, it was confirmed that our surgical training simulator has the ability to simulate actual surgery from the trials of surgeons. It was also confirmed that bone drilling technics of students had improved by repeating trials. These showed the usefulness of our surgical training simulator. Future work will include additional force assistance and graphic stereoscopy with head-mounted-display.

In recent years, the popularity of fiber-reinforced materials in the aerospace and automobile industries grows due to their high stiffness-to-weight ratios. Thus, a structural design methodology is needed to bring out the celebrated mechanical properties of these materials as much as possible. From this aspect, topology optimization has been considered as an effective method for such a method. As a natural consequence of the trend, intensive researches on topology optimization methodologies to handle the anisotropy of fibrous materials has been conducted. However, these researches were limited to the static problem, i.e., compliance minimization problem. In the present work, we propose a dynamic topology optimization method for fiber-reinforced materials which is capable of optimizing element-by-element orientations. Our method builds upon a framework of SIMP type topology optimization and can optimize each element’s density and orientation as design variables. Numerical analysis is done by Finite Element Method and Method of Moving Asymptotes is implemented as an updating scheme of design variables. The effectiveness of our method is validated through the eigenfrequency maximization problem and eigenfrequency gap maximization problem.

Distribution and change of roll stresses, slippage, web tension, and friction forces during winding rolls are investigated using commercial FEM software. The model used here is two dimensional and consists of an isotropic web and a rigid winding core. The rolls are wound up to 70 or 80 layers by the center winder with a constant line tension and this winding process is numerically simulated. The results show that the roll stresses decrease when the front part of the web is tapered compared to the non-tapered case and approach the result of Catlow or Hakiel. The roll stresses and slippage increase as the line tension increases or an imaginary coefficient of friction or Young’s modulus decreases. The slippage occurs as the inner layers of the roll move faster than the outer layers. In case of large slippage, the tension of a point on the web gradually increases largely after entry into the roll. The friction forces acting between the web layers change violently in the inner layers of the roll and gradually decrease to zero in the outer layers. The friction forces acting on the inner surface of the web layer mainly direct to the direction of winding. A few local areas with extremely high friction forces occur and disappear abruptly and some concentric circles with relatively large friction force are observed constantly in the inner layers of the roll. The friction forces acting on the inner surface of the previous turn largely change in areas where the friction forces of the new turn are significant.

Cerebral palsy (CP) causes abnormal gait pattern due to increased stretch reflex and contracture in specific muscles. Increased stretch reflexes in CP are caused by increased activation of muscle spindles due to increased sensitivity of gamma motor neurons. Furthermore, these neurological abnormalities may cause abnormal postures such as crouching and scissoring. This study aimed to construct a system for supporting treatment and orthosis development in patients with CP by simulation using a three-dimensional neuromuscular skeleton model and evaluate the validity of motor function due to CP reproduced by gait simulation. By reproducing motor function due to CP, the simulation system was able to obtain gait and muscle activity patterns similar to the actual ones. Therefore, motor function in CP can be quantitatively evaluated using this simulation system. In the model of children with CP, gait factors and joint angles of hip, knee, and ankle resembled those of the experimental results; however, the results of muscle synergy analysis were different. Hence, although gait pattern might be highly involved in the enhancement of stretch reflex and muscle contraction, muscle interaction and single muscles influence muscle control in CP.

Gymnastics are diversifying skills and advanced with the development of the competition. Therefore, the guidance and scoring of gymnastics are also difficult. In recent years, various sports aimed at improving the performance of players from scientific perspectives are increasing. Even in the gymnastics competition, a player's skeleton is estimated from a number of points with three-dimensional coordinate data in measurement using Light Detection and Ranging(LiDAR), and the operation analysis of the player is conducted. They identify gymnastics skills as a scoring support system and serve to accuracy and fairness of scoring. However, they have not been evaluated how players deal with the apparatus. The grasping the size and direction of the force from the amount of deformation of the apparatus leads to the improvement of the skills. In addition, players are changing the gymnastic apparatus to their own preferences by adjusting the load of cables. From this, the load of cables is an important factor in load condition identification of the apparatus. In this research, we will use the load cell to measure the load of cables and consider whether the load of cables when the quasi-static load is applied is useful for load condition identification of the gymnastics horizontal bar apparatus.

This paper investigates the utility of nonlinear optimal feedback control for a rotary pendulum based on deep learning. The neural network trained with supervised learning using solutions of the minimum-time optimal control problem provides the feedback controller, which generates control variables by current state variables. The utility of the proposed idea is verified by the numerical simulation and experiment considering the pendulum angular velocity constraint. Results demonstrate the feasibility of commanding the rotary pendulum with the proposed nonlinear feedback controller.

Japan has been a super-aging society since 2007. The authors proposed a small four-wheeled vehicle along a concept called “NNC” suitable for such a society as Japan. The abbreviation “NNC” is after the initials of the following words, Narrow, Near and Current. A vehicle along the concept is about up to 1m wide and 2.5m long, tandem for two passengers, only for neighborhood business and can join the road traffic flow of ordinary cars. Vehicles by the NNC concept will be a new mobility device not a car nor a bicycle, which can be helpful for elderly drivers, mothers in childcare and can improve the structures of urban cities and rural areas.