This investigation treats the damped free vibrations of a sandwich beam having a viscoelastic core and elastic layers with a concentrated mass that are elastically constrained at both ends. In the analysis, the elastic constraints are considered as translational and rotational springs, and the Young's modulus of the viscoelastic core is assumed to be small compared with that of the elastic layer. The frequency equation is derived and effects of the magnitude and the position of the concentrated mass on the damped natural frequencies and loss factor of the system are discussed. Numerical results show that the loss factor becomes maximum at a particular value of the positions of the concentrated mass.

抄録全体を表示

A numerical method for the analysis of the free vibration behavior of twisted thick cantilever plates by using the Rayleigh-Ritz approach is presented in this paper, together with some numerical results. The method is based on the three-dimensional elasticity theory by considering rotating inertia and shear deformation. Numerical results are compared with the available ones in the literatures. The comparison shows the method presented in this paper to be useful for analyzing the free vibration of pre-twisted thick and thin cantilever plates. The effects of pre-twist and aspect ratio of plates on the free vibratory characteristics of the twisted thick plates are also investigated.

抄録全体を表示

Theoretical analyses are presented for free flexural vibrations of eccentrically stiffended rectangular plates with a small initial deflection. Clamped boundary conditions with movable or immovable in-plane edges and the initial deflection with double curvature are assumed. The modal equation is established on the basis of the Galerkin method combined with a multiple-mode approximation and the solution for free nonlinear vibration is obtained by applying the method of succesive approximation. Numerical results for stiffened square plates illustrate the influences of stiffening parameters, initial deflections, amplitude and in-plane edge conditions on the natural frequencies.

抄録全体を表示

This paper presents an experimental investigation for the higher modes of vibration of orthotropic stiffened square plates with a small initial deflection. The stiffened plates are made of tough-pitch copper plates by photo-etching. The plates are excited by an acoustic speaker and mode shapes are measured by laser holography. The effects of an initial deflection and the orthotropy, induced by orthogonary placed stiffeners, on the natural mode shapes are clarified. The experimental results for natural frequencies and mode shapes are found to be in excellent agreement with the previously theoretical investigations.

抄録全体を表示

The problem of a block repeatedly sliding, jumping and being conveyed inside a vibrating spatial tube is studied. Analytical models and equations are established in considering three kinds of motion states : relative sliding motion, flying motion and the bumping process. Precise investigations are carried out for the inclined frictions with component velocities and for oblique impact, which are seldom discussed thoroughly and are easily misinterpreted. The detailed motion of the entire response, called the "subperiodic motion", is revealed through numerical solutions. Fundamental effects of some basic parameters on the motion of the system are discussed. Appropriate ranges for both the frequency and the amplitudes of vibration are discovered to be very narrow. Experimental studies confirmed the theoretical analyses presented.

抄録全体を表示

This paper deals with an optimum design method for connecting elements of adjacent structures which are excited by earthquakes. Considering the expansion of this study to an active or semiactive control of connecting elements, the structures are modeled by block diagram. In order to improve the efficiency of calculation, the substructure synthesis method is introduced. Time-domain optimization with inequality constraints is carried out using a nonlinear programming technique. A connecting element is modeled by a pair of a spring and a dashpot. Elasto-Plasticity is taken into account using the equivalent linearization method. These characteristics of connecting elements are optimized to minimize the seismic time-response of the structures, under the constraint for relative displacement between these structures. This method is applied to a fundamental lumped-mass model of a boiler plant structure which is composed of the boiler and its supporting structure. As an objective function, a total of the story shear force of the supporting structure is adopted. As a result of this investigation, the proposed optimization method has proven to be effective and practical for aseismic design of adjacent structures with connecting elements.

抄録全体を表示

This paper presents a stabilization method of synchronized active control for machinery vibration isolation. The stability of the synchronized vibration control can be lost when occuring changes in the vibration characteristics of the system. The stabilization is accomplished by active damping control using the actuators for isolation, which smoothes the curves of the transfer functions between the control signals and the residual vibrations. Three damping control methods are proposed : (1) Feedback of the velocity at the actuator upper part ; (2) Feedback of the velocity at the actuator lower part ; (3) Feedback of the integrated force through the isolator. For the typical modal models of vibration systems, the stabilization performances of these feedbacks are examined numerically, evaluating eigen values of the state transition matrix of the synchronized isolation control. The effectiveness of these methods is demonstrated experimentally, using a machinery model supported by 4 piezoelectric actuators. It is also found that the isolation effects are increased by the active damping control.

抄録全体を表示

Vibration control of a double pendulum is studied by actively varying the vertical position of its hinge. The pendulum is hinged to a truck in a vertical linear way. The truck is moved vertically by means of a belt, two pulleys and a DC servomotor. The motions of the truck and pendulum are governed by a bilinear form of the state quantities. On the basis of control theory of the bilinear system, the motion of the pendulum is controlled and stopped. This method may be applied to the vibration control of the string for tether satellites and the pipe string for mining manganese nodules.

抄録全体を表示

This paper presents the active vibration control of a cantilevered pipe conveying fluid, which is hung vertically. The response of the cantilevered pipe is controlled by the torque produced by two pairs of tendons attached to the pipe and two servomotors. Supposing that the amplitude of pipe vibration is small and that the deflections in the two directions are not coupled, the deflections are controlled independently. The equation of motion is solved by means of the Galerkin method. In order to reduce the number of system modes, a suboptimal control law considering the effect of spillover is applied to the system. The experiment was performed using a long flexible pipe conveying water, which had a nozzle at the tip. Numerical simulations and experimental results are in good agreement with each other and show that this method is effective for suppression of the flutter of the cantilevered pipe conveying fluid.

抄録全体を表示

For vibration control of a flexible structure under high-speed motion, motion control using velocity trajectory is often used. One of the techniques used to determine optimal motion trajectory, which suppresses the residual vibration most quickly when stopping the motion, is an approach using linear programming. In this paper, this approach is extended to restrict the relative displacement and the jerk value during the motion, and to minimize vibration magnitude over some defined time interval. The proposed method improves robustness against higher-order vibration modes and nonlinearity ignored in structural modeling such as rattles and system changes. The validity of these improvements is demonstrated by simulation experiment.

抄録全体を表示

Excitation tests were carried out for an experimental model of a mass damper with a 200 kg moving mass which was installed on the roof of a 5-story building model with a total mass of 6029 kg and a height of 3.58 m. The 1st mode (2.4 Hz) and 2nd mode (7.3 Hz) of the building vibration were controlled, using a variable feedback gain control law to keep the mass damper displacement within a limit. Under the control law, the mass damper can work against not only winds and weak earthquakes but also strong earthquakes. Two types of variable gain control law were tested, one of which used a crisp switching rule of the gain and the other a fuzzy switching rule. The test results demonstrated the effectiveness of the control laws. Furthermore, the influence of the reduction in the number of sensors and robustness of the control system, including the modal filter, were tested with satisfactory results.

抄録全体を表示

In balancing the rigid rotor, correction masses which are calculated by the least squares method using a performance function defined as a sum of squares of a residual vibration are often undesirable values. In the actual correction process, the magnitude of the correction masses which can be put on or cut off from the rotor is constrained by the structure and the shape of the rotor or the method of the process. In this study, an improved balancing method which can reduce the correction masses is proposed. In this balancing method, both residual vibration and correction mass are adopted as parameters of the performance function. The effectiveness of the proposed balancing method is confirmed by computer simulation and experiment using the rotating cylinder unit of the VTR.

抄録全体を表示

This paper describes the principle of an active lead screw mechanism # (ALSM) and the basic characteristics of its experimental mechanism. The ALSM is comprised of a lead screw, two nuts, piezoelectric actuators #(PEAs) and a movable table. The table is connected to the nuts through the PEAs so that the PEAs can control the amount of the backlash and the position of the table. The ALSM has following capabilities ; (1) when the table is to be driven fast and positioned coarsely by a motor, a sufficient backlash can be given between the screw and nuts, (2) after the table is positioned coarsely, the backlash can be eliminated by the PEAs so as to give the ALSM a sufficient stiffness, and (3) keeping the backlash at zero, the table can be positioned precisely by the PEAs. The experimental mechanism achieved : (1) the maximum table speed ; 130 mm/s, (2) the positioning accuracy ; better than l0 nm, (3) the positioning time for a 10 mm step input ; 369 ms.

抄録全体を表示

Optimal control of a screw conveyer is discussed in this paper. It is known that there is the optimal transport conditions exist from the viewpoint of transport efficiency when granular materials are transported by a screw conveyer. This is defined by the filling ratio of granules inside the trough. The filling ratio can be identified by detecting the torque of the screw conveyer shaft. Therefore, the optimal transport conditions can be realized by controlling both the transport quantity and the torque of the screw shaft at the same time. As a control algorithm, a generalized predictive control (GPC) method extended to a multi-input multi-output linear system is applied to the screw conveyer process. Good response waves for a set-point change and a disturbance input are obtained by introducing a design polynomial into the GPC algorithm. Moreover, the effects of the control are examined by FFT analysis of the response waves.

抄録全体を表示

Although analytical and experimental studies on the bending propulsion mechanism of fish or cetaceans have been done by many researchers, the reasons for high speed and high efficiency of the animals' bending propulsion motion have not been clarified, probably because the dynamic interaction between mechanism and fluid has not been taken into account. This paper proposes an analytical method for a coupled system of mechanism and fluid dynamics in a bending propulsion mechanism, based on the discrete vortex method. We analytically and experimentally investigate the characteristics of the propulsion speed of a three-joint bending propulsion mechanism. It is found that the propulsion speed takes a maximum value when all phase differences of joints are about zero for four amplitude patterns in both the theory and experiment. The experimental propulsion speeds are found to be 65∼80 percent of the analytical ones.

抄録全体を表示

This paper deals with a mathematical model of flexible bodies connected by rotary joints based on the Lagrangian formulation. The equations of motion are derived in terms of angular velocities of the bodies and amplitudes of elastic deformation modes of the bodies. The kinetic energy, which is a quadratic form of the velocities, comes to be of a simple form when the elastic deformation modes are chosen to be orthogonal to the translational displacement modes of the mass centers of the bodies. This, in turn, causes the equations of motion to be formulated in a recursive form.

抄録全体を表示

An optimum design method is considered as a method whose results are the most reasonable from various standpoints in the prescribed mathematical formulation. However, its formulations present difficulties in choosing objective functions, setting allowances to constraints and so on, because they are greatly concerned with the results of optimization. In this study, we proposed concepts of qualitative sensitivities and qualitative optimality compared to quantitative concepts, and proposed a general algorithm of the qualitive optimization method. We newly defined three kinds of fuzzy language and assigned fuzzy sets to each of them and assigned them to each candidate of discrete design variable. We propose fuzzy reasoning rules by using these two kinds of fuzzy sets. We also extend the proposed algorithm to multiobjective optimization. Through the numcrical example of a single objective problem, we obtained almost the same solution as in quantitative methods, and efficiencies were confirmed for this case. Also through the example to multiobjective problem, the new method was shown to have more flexibility in deciding in deciding the preferred objective functions compared to the conventional method.

抄録全体を表示

This paper presents an integral equation formulation for design sensitivity coefficients of frequency responses in acoustic problems. The direct differentiation method is applied in the derivation of the sensitivity formula. The integral equation relating the pressure at an internal point to the boundary pressures and velocities is regularized and differentiated directly with respect to a design variable such as shape design parameter, frequency, mass density and impedance. The resulting integral equation for the design sensitivity coefficients has no Cauchy principal value integrals and can be applied continuously across the boundary of the acoustic field. The effectiveness of the present formulation is illustrated through some numerical examples for three-dimensional problems.

抄録全体を表示

Transient vibration occurs out when the force acting on a system suddenly changes. Overshoot of the response is one of the evaluation points of the transient characteristics. However, it is difficult in a multi-degree-of-freedom system to reduce the overshoot without increasing damping. That is because so many vibrational components appear at once and interfere with each other. Sensitivity analysis has been used to improve eigenfrequencies, eigenmodes or transfer functions, but few applications are known in the field of time domain response. In this paper, improvement of the overshoot in a transient vibration through the sensitivity analysis is discussed.

抄録全体を表示

One-dimensional vibration intensity is measured at 4 points, or 2 points on the assumption of far field (called the 4- or 2-points method in the present paper). With regards to theoretical principles, the 4-points method is more accurate compared with the 2-points method, but measured value at each point includes errors. The influence of these errors upon the calculated value of vibration intensity depends on the employed method. In the present paper, errors due to difference and measurement error for both methods are examined, measurements for a beam are performed and following results are obtained. (1) The 4-points method suffers from measurement errors significantly when the product of wave number k and the space between sensors Δ is small. For precise measurement, it would be better to make kΔ large and to correct the calculated value by the factor of the finite difference error. (2) For the present apparatus, measurement can be performed well by the 4-points method, setting Δ so that kΔ is nearly 1.

抄録全体を表示

This paper describes the detection of vibration intensity vector using three accelerometers. With this method, the amplitude and direction of vibration intensity vector are obtained by processing three output signals from accelerometers set on beams or plates of a machine structure. Using this method, vibration intensity vectors around the exciting point on a steel plate are measured. The vibration power of bending waves calculated by integrating the vibration intensity vectors is compared with the excitation power measured by an impedance head. The result shows that the three -accelerometer method can be used to detect the vibration intensity vectors. Vibration intensity and reactive intensity vectors on gas heat pump package panels are also measured using this method.

抄録全体を表示

This paper discusses the control performance and the optimization of the parallel hybrid dynamic vibration absorber (PHADVA). In most previous studies, a PHADVA was said to achieve better performance than an ordinary ADVA irrespective of the control level. In this study, it is shown that the control cost of a PHADVA exceeds that of ADVA at a certain control level. This result indicates the significance of the optimization of the spring constant and the damping coefficient. The numerical calculations were carried out using a three-degrees-of-freedom structure subjected to a stationary white-noise disturbance. In order to investigate the transient characteristics of PHADVA, computational simulation was also carried out by using two types of measured seismic disturbances. These results show the usefulness of the optimal design of PHADVA.

抄録全体を表示

A seven-degrees-of-freedom manipulator actuated by a tendon-drive system has been developed. In order to reduce the weight and volume of the manipulator, each motor was installed on the base frame and the motor torque was transmitted to each joint through a tendon pulley system. The tendon drive mechanism, which uses a coupled drive, actuates the manipulator by controlling the tension of each tendon. In the developed system, eight direct-drive motors and tendon-pulleys were used for a seven-degrees-of-freedom manipulator. In this paper, the structure of the manipulator developed is first explained. Then, the tendon-tension-based control method for the coupled drive manipulator is formulated. Finally, through simple experiments using the manipulator in which the control system was installed, the effectiveness of the proposed structure and the control method is demonstrated.

抄録全体を表示

To reduce the components, six electromagnets are applied to construct active magnetic bearings. The radial bearing is composed of three electromagnets, whose pole faces are in the shape of a cone, to generate the axial force ; the combination of the two radial bearings also has the function of a thrust bearing. A theoretical linearized model of the rigid-rotor motion is derived so that five control inputs can be decoupled into five motions when the electromagnets have similar dynamical characteristics. A method for designing an approximate decoupling control system is considered, and a relatively simple structure is presented to give actual input into the six electromagnets. The one freedom in giving the actual control input is considered in terms of the rate at which the two bearings take part in the thrust bearing. The results neglecting the thrust control are applicable to the similar radial bearings in the shape of a right cylinder.

抄録全体を表示

The present paper proposes an instantaneous inverse kinematic solution for redundant manipulators based on virtual arms. The virtual arm has the same kinematic structure as the manipulator except that its end-point is located on the joint or link of the manipulator. When the appropriate virtual arms are used, the configuration of the manipulator can be represented by a set of end-points of the virtual arms. First of all, this paper formalizes the kinematics of virtual arms and derives instantaneous inverse kinematics. Then, the methed is applied to winding control for hyper-redundant manipulators. The winding control presented here is divided into two steps : 1) planning desired positions for virtual end-points, 2) integrating them into the joint trajectory of the manipulator. The desired positions of each virtual arm can be computed in a parallel and distributed way and it is not necessary to consider joint space of the manipulator. Finally computer simulations show that the winding control for a hyper-redundant manipulator can be perfomed in 3D-space.

抄録全体を表示

This paper shows that our fictitious set point can be applied to a variable structure controll (VSC). The fictitious set point is simply realized in the control algorithm for a VSC and this control system has the feature that the manipulated variable is reduced as compared with the case where it is not used. The VSC is applied to the constant desired-value control of the two-link manipulator having uncertainties of the load and friction coefficient variations. The computer simulation shows that this system can accomplish the control purposes under such uncertainties and that the input chattering can be reduced remarkably by introducing this fictitious set point.

抄録全体を表示

This paper discusses robot motion planning in dynamic environments, that is, generation of a motion pattern to achieve a task under restrictions of time and space. We especially focus on "kendama" as a typical example of such a task. In this task, no a priori knowledge of a ball (mass & length of a string) is assumed to be possessed beforehand. To perform this "kendama" task, we propose how to generate the motion pattern of the manipulator with a force sensing device attached to the end tip. Experimental results obtained using a 2 d. o. f. drect-drive manipulator are also shown.

抄録全体を表示

In this paper, we discuss the lay-down operation, which is one of the tumbling operations using two-fingered hands. We explicitly derive an analytical solution for the minimum fingertip force at a given fingertip position. We also propose an algorithm which is able to determine the optimal fingertip position. The optimal fingertip position is obtained by a contour line of minimum fingertip force.

抄録全体を表示

For controling aqua robot manipulator working in water, we must consider the influence of fluid force on its motion control. Particulary, fluid flow seems to have a strong influence upon the motion of an aqua-robot manipulator when it undertakes a highspeed, precise motion. To reduce the fluid force influence on the motion of aqua-robot is quite useful for not implenting large capacity actuators into the aqua-robot manipulator. Thus we made experiments for investigating the condition of reducing fluid force acting on an aqua-robot manipulator's arm under the fluid flow. By changing fluid flow velocity, air bubble ejecting patterns and ejecting pressures from the surface of the manipulator's arm, we measured the fluid moment acting on the arm, and as a result, it was found that there exists a specific condition of air-bubble ejection for reducing the fluid influence even under the existence of fluid flow. Fine air-bubble ejection from the arm's surface has achieved about 40% reduction of the fluid force influence compared to no-air ejection.

抄録全体を表示

This paper proposes an image extraction technique of the human face by color image processing, fuzzy inference, and neural network. The image processing technique provides a method for extraction of skin color and measurement of detailed information of the human face. The fuzzy inference estimates the face direction based on the configuration of the extracted face candidates, while the neural network distinguishes faces and other objects by detailed information and face direction. Compared with the previous works which extract data on the human full face, this method makes it possible to recognize not only a full face but also profiles. Further, we show how this method works for the illusion problem. The extraction accuracy of this system is 80 percent at this moment. According to the experimental results, this system will be useful for the vision systems of autonomous robots.

抄録全体を表示

A 1-microphone system is usually used to detect the signal of a noise source in an active noise control system. In this case, unstable oscillation, called howling, will occur according to the gain of the transfer function. In a long duct, where few reflected waves exist, a 2-microphone system has been used to eliminate the signal from the secondary source to detecting microphones. However, in a short duct, the reflected waves must be considered. The authors studied the acoustic conditions in the 2-microphone system theoretically with the existence of the reflected waves. The elimination effect from the secondary source was confirmed by simulation and experiment. The 2-microphone system is valid when sound absorption material is installed at the opposite side of the opening end.

抄録全体を表示

There are a number of unclarified points concerning the impact phenomena and the optimum design of tennis rackets. This paper investigates the effect of string tension on the impact phenomena on the basis of a simple impact model where the contact duration is determined by the natural period of a whole system composed of the mass of the ball, the stiffness of ball and strings, and the reduced mass of racket at the impact point on the racket face considering the strong nonlinearity of the resforing force of the ball and strings. It also deals with the effect of string tension on the structural vibration of the racket frame, which is supposed to reduce the coefficient of restitution of the tennis racket due to the energy loss of the frame vibration. The calculated results explain fairly well the experimental results.

抄録全体を表示

The operating characteristics and stress of a spring should be understood in order to design an optimum spring mechanism. This paper illustrates the calculation method of stress with a simple model and compares the calculation and test results on the operating mechanisms of SF₆ circuit breakers in order to prove that the stress vibrates with the surge time periods, and that the stress amplitude, and consequently the mechanical life of the spring, depends on the surge intensity which dominates dynamic operating characteristics. Calculations of the surge intensity of various kinds of springs are given. The stress amplitude can easily be obtained using the surge intensity even at the initial stage of design. A mechanism with a helical spring should be designed, taking account of the surge intensity because of its large value. Surge intensity is useful for optimum design in arrangement and selection of springs.

抄録全体を表示

The direction of the joint force acting on each pair of high-speed link mechanisms may change suddenly. This causes separation and collision of paired elements, and abrasion of the elements. These problems become more serious for joints moving in space, for their lubricant conditions tend to be poor compared with joints fixed to a base. Therefore, it is important in the dynamic design of the mechanism to control sudden changes of direction at the joints moving in space. This paper deals with two methods of dynamic design in order to control sudden changes of the direction of the crank -pin joint force on a crank-lever mechanism through the redistribution of the mass of each link. Examples of the design and the effect of this control are shown. A force sensor with strain gauges is made and the magnitude and the direction of the joint force are measured to confirm the theoretical results experimentally.

抄録全体を表示

Ta- and W-amorphous alloys, which are fabricated by the PVD method, have been rubbed against 8 pure metals (Ni, Fe, Ta, Cu, Al, Mg, Ti, W) in 1.0 × 10⁵ Pa dry air and in 1.5 × 10^-3 Pa vacuum. In vacuum, when rubbed against Ni, Fe and Ta, Ta-amorphous alloy is worn out much more than when it is rubbed against other metals. This is thought to be due to the high adhesivity between metals and amorphous alloy, because Ta-amorphous alloy contains mating metal (Ni, Fe, Ta) as a component. In all the rubbing systems, the coefficient of friction is higher in air than in vacuum. Rubbed against Cu or Mg, the amount of transfer of Cu or Mg onto the amorphous alloy surface is much greater in air than in vacuum. Oxygen atoms coexist with Cu or Mg atoms on the surface of amorphous alloy. From these results, it is deduced that atmospheric oxygen increases the adhesivity between metals and amorphous alloy, which enhances the transfer of metal onto amorphous alloy.

抄録全体を表示

Polyacetal gears have excellent characteristics, but there is one demerit in that they make as much noise as plastic gears. Therefore, for the purpose of reducing the noise of these gears, two kinds of composite polyacetal, filled with various additives, were developed. In this study, gears were made of these composite polyacetals, and the effect of reducing the noise was confirmed. Also, operation tests were carried out for long hours under various conditions, and various characteristics such as the wear of the teeth and the change of the tooth profile were examined and compared with those of polyacetal gears for which results of practical use were available. In this way, their practicality as power transmission gears was investigated.

抄録全体を表示

In the present study, the deformation mechanism of hot finish rolling of plastic gears is discussed. This mechanism is important for developing a simulation program of this finishing process. For this purpose, indentation tests of rolling rollers were performed so as to show the relationship between the plastic deformation and various conditions. As a result, it was found that the plastic deformation depends strongly on the Hertzian contact width between rollers. In hot finish rolling, the Hertzian contact width between tooth flanks of die wheel and work gear is generally not so large that the plastic deformation causes change in the work tooth form. Hence, it can be concluded that the change in the work tooth form in the hot finish rolling process is not induced by the plastic deformation of the work tooth flank. Thus, we calculated the temperature distribution in the direction of tooth thickness by finite element analysis of heat conduction. According to the result of this calculation, a melted layer of work tooth flank material exists on the tooth surface. Consequently, the change in the work tooth flank in the hot finish rolling process is induced by the flow of the melted layer.

抄録全体を表示

In recent years, spiral bevel gears have been used in various fields. Therefore, it is one of the most important subjects in the gear field to develop gears which are effective to reduce vibration and transmission noise. However, there have been very few studies hitherto which deal with vibration and noise of spiral bevel gears, and many problems still remain to be solved. Considering the meshing characteristics of the Oerlikon-type spiral bevel gears under various running conditions in a case when the module is changed, the authors examined tooth bearing, acceleration at the gear housing and stresses at the tooth root of drive and driven gears. From this study, meshing characteristics of the Oerlikon-type spiral bevel gears (drive and driven gears) could be clarified in the case in which the module is changed.

抄録全体を表示

The paper investigates the effects of surface flaws caused by the grinding of zirconia on strength reliability of the material. The following grinding conditions were selected, and grinding operations using a diamond wheel were performed on the tension side of each of five flexural specimens (50mm × 4mm × 3 mm) : (A) creep feed grinding normal to longitudinal direction (grain size ; # 120, depth of cut ; 2 mm, feed ; 30 mm/min), (B) surface grinding after the creep feed grinding (grain size ; # 600, depth of cut ; 4 μm, amount of removal ; 12 μm ), (C) lapping after surface grinding (amount of removal ; 3.5 to 4.7 μm), (D) creep feed grdnding in longitudinal direction. For all specimens of (A) and two specimens of (B), the flaw in the fracture source could qualitatively be detected. The flaws in both (C) and (D) could not be detected. Based on the flaw calibration, the flaw size of (A) was expected to be 12 μm. The crack length at the moment of unstable fracture was longer than the detected crack length. Considering that the crack might extend during loading, the slow crack growth was investigated using a double-torsion specimen. It was demonstrated that the crack had grown to a certain extent during loading since the expected flaw size before loading coincided with the detected value.

抄録全体を表示

TiC-TiN-cermet cutting tools are coming into wide use today in metalworking industries because of their excellent properties. Particularly, improved toughness of cermet with increasing content of nitrogen makes applications for milling or drilling cutters possible. However, it is quite difficult to grind cermet with high efficiency because of its high degree of hardness and toughness. This results in the rapid wear of the diamond grains, and the need for frequent dressing of the grinding wheel surface. In this study, three different kinds of cermet are ground with a surface grinding machine using a resinoid bonded diamond wheel. The measurements of grinding force, surface roughness and grinding ratio are carried out to obtain a guide for high-efficiency grinding of cermet. To achieve the stable cermet grinding, it is important to take full advantage of the self-sharpening phenomenon of the grinding wheel. With respect to the grinding wheel, the friable grinding wheel will be suitable to attain the stable grinding processes. However, the surface roughness obtained is large and, therefore, finish grinding with a finer grinding wheel should be carried out to meet the demand for the cutting inserts.

抄録全体を表示

Deviation of punch load center from the punch center, lateral force exerted on the punch head and eccentricity between punch and die are measured through the forward cup extrusion. When an initial eccentricity exists between the punch and die, lateral force initiates and bends the punch toward the die center ; then the eccentricity of the cup decreases with increase in punch penetration. This centripetence appears when both initial eccentricity and extrusion ratio are large. Conversely, when both initial eccentricity and extrusion ratio are small, the eccentricity induced by some disturbances increases with punch penetration. A coned punch surpasses a flat-ended punch in the centripetence, but is apt to become unstable as the process proceeds. A frustum-headed punch is concluded to be the most stable. An increase in the land length stabilizes the punch but tends to lead to galling. Unbalanced friction and the oblique end of the blank cause larger eccentricity for the coned punch than for the previously reported flat-ended punch.

抄録全体を表示

This paper deals with a newly developed underwater displacement sensor utilizing the acoustic resonant frequency of water in a submerged pipe. Both ends of the pipe are closed. A speaker to generate sound is fixed on the end of the pipe with a reflector at the other end. Piezoelectric ceramic is used as a speaker. A sweep signal is radiated from the speaker to the water column in the pipe. The sound pressure takes the largest value at the resonant frequency. The displacement of the reflector which is inserted from another end of the pipe by a plunger, corresponds to the length of the water column. The displacement of the reflector has good correspondence to the resonant frequency of the water column. It is found from the experiment that it is possible to accurately measure underwater displacement, using this sensor.

抄録全体を表示

An Autonomous mobile robot system with a visual sensor for automatic pipe welding was constructed. The robot developed can move along a pipe autonomously and detect a weld line to be welded by means of its visual sensor. The robot can confirm automatically whether the line detected is a weld line to be welded or not. Moreover, the tracking of the weld line at a constant welding speed and butt welding of the pipes are executed automatically. Generally, the recognition of a weld line is difficult in welding because of the brightness of the welding arc. In this system, in order to recognize it stably, a special lighting system and some image-processing methods are applied. The effectiveness of the system constructed was confirmed through experiments on automatic weld line tracking.

抄録全体を表示

This paper presents the application of a neural network of the Hopfield model to a plant layout problem which is classified as an assignment layout problem. The allocation matrix which is the solution of the layout problem is expressed by neurons, and the conditions of constraint are described as synapse connections. Then, we can obtain the energy function of the neural network corresponding to the objective function of the layout problem. The layout method using this Hopfield model is utilized to obtain the suboptimal solution by simulation. The effect of the coefficient of the energy function is also investigated. Furthermore, a layout improvement procedure using the simulated annealing method is developed to avoid the local minimum of the energy function.

抄録全体を表示

It is very important in a design problem of automatic assembly lines to estimate the fluctuation of daily output which results from the irregular occurrence of trouble. On the other hand, the irregularity exists not only in trouble occurrences but also in the length of recovery times. In this report, the effect of fluctuations in the recovery time will be clarified in comparison with effects of the irregularity in trouble occurrences by simulations on a computer. The main results are as follows : ( 1 ) The effect of recovery time fluctuations is comparatively small relative to that of the irregularity in trouble occurrences, ( 2 ) Trouble of short duration with low occurrence rate has little influence on the dispersion of the production rate.

抄録全体を表示

A heuristic based on simulated annealing method is developed to solve lot sizing problem with setup times. The problem assumes that the unlimited overtime production is available for each planning period so as to generate a feasible neighborhood solution easily. Firstly the problem is modified into a zero-one programming problem in order to obtain the objective function value without complicated calculation. Then the simulated annealing heuristic is implemented to obtain the best solution. Finally this solution is fed into a simplex algorithm to obtain the revised best solution. In the numerical examples, the simulated annealing heuristic is compared with the Lagrangean relaxation approach and the following results are obtained ; ( 1 ) the heuristic produced better solutions on the average, ( 2 ) setting the production quantity equal to either zero or the cumulative demand performed well in the condition of many products, and ( 3 ) performance of the heuristic was relatively sensitive to the initial random seeds.

抄録全体を表示

A single-machine scheduling model is proposed for minimizing the expected total cost consisting of flowtime cost and asymmetric penalty cost of earliness-tardiness subject to machine breakdowns. All jobs have a common due date and an identical piecewise linear cost function. Since it is difficult to calculate the exact value of expected total cost, an approximate function of expected total cost is introduced to simplify the scheduling model. Dominance conditions necessary for an optimal schedule are derived for this scheduling problem. Efficient methods exploiting these conditions are proposed for obtaining an optimal schedule minimizing the modified expected total cost. Numerical examples are demonstrated to show the effectiveness of the proposed methods.

抄録全体を表示

From the point of view of environmental protection and limited resources, reducing the amount of package cushioning and properly designing its size are problems that need to be solved. In this study, we introduced the idea of linear programming to the usual cushioning design method. Consequently, various restrictive conditions (the strength of the part of the product which is supported by the cushion, the maximum area of the part that can be allocated to the product) could be considered in cushioning design. In addition, we could optimize the volume of the package cushion and the total material cost of the cushion and shipping container as well as the thickness of the cushion. The main results obtained through simulation are as follows. (1) It was possible to reduce the cushion volume by about 50 percent in the specifications of this report. (2) The ratio of fiberboard's unit price to the cushio's influences the size of the cushion in optimum cushioning design taking into account the total material cost. (3) There is a tendency for low-expansion foam to be suitable for cases in which the area of the cushion is limited, while high-expansion foam is suitable for cases in which the strength of the supporting part of the product is low.

抄録全体を表示