The numerical inversion method of the Laplace transform, FILT (fast inversion of Laplace transform), which was developed by Hosono, yields an admirable accuracy when the singular points of the Laplace transform have finite values. In the forced vibration problems of a finite continuous system, FILT cannot estimate the effect of higher natural modes accurately because the Laplace transform of the displacement has infinite singular points that are distributed discretely and form an unbounded sequence. In spite of this fact, FILT is useful in such vibration problems. The results obtained using FILT agree with the ones obtained using the mode acceleration method except at an early stage.

抄録全体を表示

A semiactive seismic isolation system with distributed controllable friction dampers was studied, in which the damping force was controlled by varying the pressure between the friction elemens. Excitation tests were carried out for a base-isolated single-story building model supported by 4 rubber bearings and equipped with 2 controllable friction dampers on the bottom. The tests showed that the semiactive isolation system was equal or superior to normal passive seismic isolation systems in response acceleration reduction, and that the semiactive system could reduce the response displacement significantly, as compared with the passive system. The tests also showed that the small rotational response of the builiding model was induced by fluctuation in the small pressures of the friction dampers which was caused by their mechanism. However, simulation studies showed that the large rotational response could be controlled and reduced by the distributed dampers.

抄録全体を表示

A new model explaining the energy supply mechanism to self-induced sloshing in a tank with circulating flow is proposed. The oscillation energy is considered to be supplied by pressure fluctuations caused by the interaction between the circulating flow and the sloshing motion. When the direction of the under-water-surface stream of sloshing motion coincides with the steady circulation, the dynamic pressure increases and the ensuing static pressure reduction causes an imbalance between the fluid and the atmosphere. If the static pressure reduces more in the region where the free surface is descending, the work done by the imbalance supplies energy to the sloshing oscillation.

抄録全体を表示

Numerical and analytical studies are presented here which suggest that chaotic motion is possible from periodic excitation of a parametrically nonlinear mechanical system having Duffing type stiffness and van der Pol damping. Numerical integration is used to obtain phase plane portraits and Poincare maps for large-time motion. Period-doubling bifurcations and several types of limit cycles and chaotic behavior are observed. Chaotic motions are investigated using the Liapunov exponent, the invariant probability distribution, the Li-Yorke's theorem and Mel'nikov's method. Approximate analytical techniques are applied to analyze some of the limit cycles and trasitions of behavior. The results are used to predict the occurrence of chaos.

抄録全体を表示

In a horizontal rotating shaft containing a transverse crack, the vibration characteristics at the major critical speed are investigated. In a theoretical analysis, a model whose spring characteristic is approximately represented by a power series is treated. The analytic results are compared with the results of numerical simulation with a model of a piecewise linear spring characteristic. Experiments with a flexible rotating shaft system are also performed. The following points are clarified : ( a ) In the case of a comparatively large unbalance, the shape of resonance curves changes markedly depending on the relative angular position of the unbalance to the crack, and an unstable region appears for a certain range of angular position. This phenomenon is similar to the case of vertical shafts. ( b ) In the case of a comparatively small unbalance, the shape of resonance curves is not affected by the direction of the unbalance and an unstable region always appears. This is a phenomenon particular to the horizontal shaft.

抄録全体を表示

The effect of mechanical excitation on the contractility of left ventricle is simulated numerically on the basis of the model of Beyar and Sideman, in which the contraction mechanism of the myocadium is taken into consideration and the left ventricle is modelled by an ellipsoid of revolution. The model is improved by adding the effect of the reduction in the contractile force due to mechanical excitation and cardiac insufficiency. This simulation gives the same result as the experiment, in which the maximum pressure of the left ventricle with latent cardiac insufficiency is depressed remarkably by the mechanical excitation.

抄録全体を表示

A vibration isolator which possesses stiffness and damping controllability was constructed as an application of electro-rheological (ER) fluid. ER fluid is known as a functional fluid whose apparent viscosity can be varied by the applied electric field strength. The attractive characteristics of ER fluid are wide variation of its apparent viscosity and quick response time. In the present paper, a prototype vibration isolator was constructed and its performance was investigated. The experiments were conducted under the condition of constant vibration amplitude, in which a detailed investigation was made as to the efficiency of the ER damper for small-amplitude vibration. The conclusions obtained are as follows : ( 1 ) The stiffness as well as damping characteristic of the present type of vibration isolator can be varied by controlling the applied electric field strength. ( 2 ) The characteristics of the ER damper are dependent on the vibration amplitude, and the present vibration isolator is applicable to control of vibration of small amplitude.

抄録全体を表示

In major low-frequency vehicle vibration phenomena, such as shake and idle vibration, engine mounts play an important. To reduce shaking, the engine mounts must suppress Vibration by increasing their damping. In addition, in an idle state, the engine mounts should be soft enough to isolate the engine vibration from the vehicle body structure. These two requirements are difficult to be achieved in conventional engine mount design. This paper presents a new design method for solving the above-mentioned problem. Two key elements are used in this new method. The first is the phase control on vibration transmission utilizing the fluid resonance in the hydraulic engine mounts. The other in the vector synthesis approach in treating multiple vibration input to the vehicle body.

抄録全体を表示

The chaotic vibration of a cantilever beam with a permanent magnet and an additional mass on its free tip was studied experimentally. The beam was installed vertically and its clamped end was excited by harmonic force. Another permanent magnet was fixed under the free tip of the beam. Experimental evidence for the existence of a strange attractor in this deterministic dynamic system was presented. It is found that chaotic responses of the beam are much larger than its resonant ones. The range of chaotic vibrations changes obviously with the decrease in the distance between two magnets. The bifurcation diagram shows that the response of the beam suddenly comes into chaos from the regular vibration with the increase in the exciting amplitude. Experimental Poincare maps agree well with the numerical results.

抄録全体を表示

The purpose of this study is to clarify the mechanisms of the kip motion, or the upstart, which is a technique of the horizontal bar exercise. The player's motion was experimentally analyzed using a movie camera. Under the assumption that the player who performs the kip is considered to be a triple pendulum, the equations of the kip motion were derived and computer simulation was performed with various parameters. Furthermore, the kip motion of the triple pendulum with two active joints was executed and measured. It was concluded that successive motion in the kip can be divided into three phases, i. e., the swing phase, the kick phase and the stop phase. In the kick phase, which is the most important, two exact physical conditions must be satisfied, that is, certain timing of the kick and a certain attitude at the peak kicking moment are necessary.

抄録全体を表示

Recently, several efforts have been directed towards developing a joint time and frequency representation for signal analysis. Especially, for a time-varying signal, the use of a discrete pseudo-Wigner distribution (PWD) is suitable. However, application of the PWD is often complicated by the occurrence of interference terms. Thus we proposed a newly revised WD, called RID, and verified that RID remarkably reduced the interference terms of the auto-WD. In this paper, the RID is applied to cross-WD Which analyzes two different time series in the time-frequency plane. One of the two signals is called the reference function, and RID applied to the cross-WD is called cross-RID (CRID). Both the simulated signals and such instrumented signals as impact sounds of a sandwich steel sheet and voices are analyzed by RID making use of the reference function. Analytical results of low S/N signals obtained from CRID are superior to those from auto-WD.

抄録全体を表示

A generally noninteracting control system is designed for active magnetic bearing systems, which are composed of magnet cores in the shape of a cone, to have both functions of radial and thrust bearings. The structure of the bearing system is symmetric with respect to the gravity center of the rotor : the magnet cores are made of solid steel and the electromagnets have similar dynamical characteristics. The dynamics of the magnetic control force are modeled by the sum of two second-order time-lag systems, and unknown parameters are estimated experimentally. Experimental results are obtained for frequency responses of the rotor motion to disturbance inputs and for responses to impact to the rotor. The former responses are compared with the numerical results using the dynamics of the control force with the estimated parameters. The results show the effectiveness of the model and the usefulness of the method for designing the control system.

抄録全体を表示

A theoretical linearized model is presented for the dynamics of active magnetic bearing systems in which electromagnet cores are in the shape of a cone in order to have the function of both radial and thrust bearings. In the analysis of the dynamics of electromagnets, a new model with the effects of eddy currents is considered to give a unified formulation that is applicable to a case of solid magnet cores. The linearized equations of rotor motion are much simplified when the dynamics of the electromagnets which are located in the same radial direction are equal in the two bearings. In this case, the design of a decoupling control system is considered, and the structure of the control system is obtained to give the actual control input into the electromagnet coils so that the resulting control forces may not interact in the five directions of motion.

抄録全体を表示

The objective of this paper is to solve the minimum-time trajectory planning problem of a robotic manipulator under both the average heat generation restriction and the intermediate state variable restriction. We have already proposed a new numerical method to solve this problem in which the trajectory is approximated by the fifth-order Hermite polynomial expression. In analyzing a many-degrees-of-freedom manipulator, however, this method cannot be carried out within a short calculation time because of its duplicated iterative calculation procedure. To reduce the calculation time, we investigated the properties of the method and found two useful characteristics. The first characteristic is that the duplicated iterative calculation is not necessary when the robotic manipulator is a conservative system in a mechanical sense. The second one is that the polynomial solution can be derived when the robotic manipulator is governed by the constant coefficient linear differential equations. Using these characteristics, we constructed the trajectory planning method, which can generate the trajectory of the three-degrees-of-freedom SCARA-type robotic manipulator within a short calculation time. Numerical examples were shown to prove that the proposed method was practical enough to be applied to industrial robotic manipulators.

抄録全体を表示

Investment in miclo-moving devices has been carried out. In many cases, their driving principle is based on a small amplitude of vibration. But, using this principle, it is difficult to move smoothly if there are holes or differences in level. In this paper, I have tried to make a device which can jump over obstacles and move smoothly on a rough surface, and then have examined its characteristic obstacles condition. This device has some functions for moving adjusted to the change of, for example, jumping in all directions moving quickly and recovering when it falls down.

抄録全体を表示

In order to clarify the unstable vibration of a flexible disk coupled with a head-suspension system, the effects of friction force between head sliders and the disk and other parameters of the head-suspension system are analyzed using a disk-head coupling model. In this model, it is assumed that the transverse motions of upper and lower head sliders interact with disk motion through transverse contact springs at the coupling point. It is found that the dynamic component of friction force, which is proportional to varying load on each disk surface, generates a bending moment about the coupling point on the neutral axis of the disk, and that this moment destabilizes some natural modes of the head-disk coupling system. These instabilities which due to frictional moment are hardly affected, by contact spring stiffnesses on upper and lower sides, but are proportionally dependent on the friction coefficient on the side where the dynamic contact force is larger. When the friction coefficients of the upper and lower surfaces are different from each other, strong instability appears at the resonance frequency of the upper and lower head-suspension system, the degree of which is proportional to the difference of friction coefficients of the upper and lower surface.

抄録全体を表示

In order to clarify the unstable vibration of a flexible disk coupled with head-suspension systems, the effects of dynamic friction force between head sliders and the disk and other parameters of the head-suspension system are analyzed using a disk-head coupling model. In this model, it is assumed that the transverse and pitching motions of the upper and lower head sliders interacts with the disk motion through transverse and pitching contact springs at the coupling point. The dynamic components of friction force produce a bending moment about the coupling point, and this moment destabilizes some natural modes of the head-disk coupling system in the entire frequency range. When pitching motion of the head-suspension system is coupled, a reacting friction force on the head slider produces a moment about the center of mass of the head slider, and this moment strengthens instabilities. Unless the friction moment is applied to the head slider, the pitching moment of inertia of the head slider has the effect of suppressing the instability due to the friction moment on the disk, except near the pitching resonance of the upper and lower head-suspension systems. From these two effects of pitching motions of the head-suspension systems, the degree of instability of the natural modes increases dramatically near the pitching resonance. If the pitching resonance of the upper and lower head-suspension systems approach their transverse resonance, strong instability appears at this frequency.

抄録全体を表示

To improve the accuracy of CMM (coordinate measuring machine), the numerical error correction of systematic error components ascertained by the calibration procedure is employed. One way to calibrate CMM is to take advantage of the three-dimensional gage made for experimental purposes by the author. It is necessary that the gage is inspected correctly and measured with high accuracy with respect to repeatability and reproducibility. Furthermore, the shape of the gage or the relative positions of steel balls used as measuring points should remain unchanged for a long time. This paper describes the experiments on the accuracy of the gage, repeatability and reproducibility of measuring the gage, and the long-term stability of the measuring points, and gives the results of tests.

抄録全体を表示

Achieving expected future exhaust emission and fuel consumption standards for passenger cars will require fundamental improvements in the combustion process for spark ignition engines. improvements must be made in not only fuel-injection but also spark-ignition devices based on precise studies of the engine-combustion process. Real-time measurement of light emssion of spark and igition kernel in an actual automotive engine is currently very few. In this paper, a real-time method using an optical combustion sensor that combines fiber optics with a conventional spark plug is presented. It is shown that intensity of light emission of spark is related to the spark voltage, and is changed according to the mixture temperature and engine load.

抄録全体を表示

In this paper we apply the robust stabilization problem using normalized coprime factor plant descriptions, which is proposed by Glover et al., to the robust control design for a flexible arm. First we design the feedback controller for this robust stabilization problem, which stabilizes the nominal plant including a rigid body mode and the first two flexible modes. The feedback system with this controller satisfies some design objectives, e. g. the bandwidth requirement. We show that by this controller we can reject the spillover effect of the remaining higher flexibility modes. Furthermore we consider robust stability analysis for the structured uncertainty. We investigate the dependence of the maximum robustness margin on the damping coefficient. Then we calculate a range of the damping coefficient in which stability is guaranteed by the controller which achieves the maximum robustness margin.

抄録全体を表示

We have constructed a leg-wheel mobile robot called Chariot which has mechanically separated wheels and legs which allows it to utilize the advantages of the two mechanisms. Chariot can move at a speed of 20 km/h on a flat plane. It can also climb steps. In this paper, one of the characteristic motions of this leg-wheel robot, that is, pitching of the body when accelerated or decelerated by wheels is analyzed. It has been shown that pitching is reduced by trajectory optimization. The results of the experiments are also shown.

抄録全体を表示

The manipulator/vehicle system floating on water consists of a vehicle and a manipulator attached to it. Similar to the space manipulator/vehicle system, the system on water is not fixed to an inertial coordinate system. Moreover, the external forces affect the motion of the system, and the interaction of motions among the vehicle, the manipulator and water causes vibration of the system. This paper discusses the control issues of a manipulator/vehicle system floating on water and proposes an algorithm to control both the endpoint trajectory of the manipulator and the vibration of the vehicle simultaneously using redundant actuators. We develop the basic kinematic equations and then derive the equation, which describes the dynamic relation between the system and the external forces/moments applied to it, then propose a simultaneous control algorithm of the manipulator endpoint trajectory and the vibration of the vehicle. Simulation results illustrate the effectiveness of the proposed control algorithm.

抄録全体を表示

In this paper, we propose a "center of gravity control" method to apply to the manipulator/vehicle system, controlling the center of gravity of the system, so that the system cannot fall down. Unlike in the case of the manipulator fixed to the floor/ground tightly, it is a very serious problem for the manipulator/vehicle system to avoid tumbling. Therefore, we consider the algorithm to control both the tip position of the end effector and the center of gravity of the manipulator and the payload being carried by the manipulator. We applied this control algorithm to the manipulator with a vehicle for man-robot cooperation. The trajectory of the manipulator is given by a human operator and the center of gravity is set in advance so as not to tumble. We carried out experiments to confirm the effectiveness of this control algorithm for the manipulator/vehicle system. In this experiment, we applied this control algorithm to a manipulator for the man-robot cooperation. This gravity center control algorithm can be applied to various types of manipulator/vehicle systems.

抄録全体を表示

A general-purpose planning system is developed for assessing optimal operational strategies of energy supply plants. This system has three main functions : ( a ) data registration, ( b ) graphic editing, and ( c ) automatic programming and optimization calculation. Data on performance characteristics of equipment, tariff of input energy, energy demands, and ambient conditions are registered to form a data base that can be cited from a graphic editor. The graphic editor enables flexible and easy change of plant configurations for planners to investigate many alternative configurations of energy supply plants. Data on the plant configuration composed by the graphic editor are arranged automatically to carry out an optimization calculation, in which the operational strategy is determined so as to minimize the hourly operational cost subject to the energy demand requirement. Through an example, it is ascertained that the system is a useful tool for operational planning of energy supply plants from the viewpoints of rationality and time or labor reduction.

抄録全体を表示

The authors and others have previously clarified the mechanisms of the impact phenomena of a club head and a golf ball using a finite-degrees-of-freedom model. In this paper, the mechanisms of the impact phenomena in tennis are clarified for purposes of the optimum design of the tennis racket, in terms of the natural frequency calculated by the frequency domain analysis. A simple beam model is used for the frame, and mass-spring models are adopted for the gut and ball. The influences of the frame and gut, which have different mass and stiffness, on the repulsive velocity of the ball are investigated.

抄録全体を表示

Experimental and numerical studies were conducted to clarify why and how surface cracking occurs in an alumina ceramic ring of a mechanical seal. By using a Au layer vacuum-deposited onto the alumina surface except the rubbing part with a carbon ring as a part of a D. C. circuit, the moment of crack initiation was identified. Temperatures and thermal stresses produced in the seal by participation of the frictional heating were calculated using finite-element techniques. It has been found that surface-cracking is caused by high thermal tensile stresses applied in the circumferential direction. It has also been pointed out that in order to prevent cracking, it is desirable to reduce the coefficient of friction between the sealing faces and to choose a seal material with low modulus of elasticity, low coefficient of thermal expansion, high thermal conductivity and high tensile strength.

抄録全体を表示

The secondary couple components in the plane containing both axes of a constant-velocity joint (CVJ) are T tan (θ/ 2) ( T ; torque, θ ; operating angle) and the components perpendiculer to the plane are zero, if the friction in a CVJ is not taken into account. The secondary couples of a real CVJ are thought to be considerably affected by the frictional forces acting in it. The authors constructed a system by which to measure the secondary couple transmitted to the bearing of a CVJ shaft and initiated the study of the secondary couples of an actual CVJ. It is shown that the effects of the frictional forces on the former secondary couple components are related to the torque loss of a CVJ and that the torque loss can be calculated from the measured values of the component using a simple equation. It is also shown experimentally that the latter components exist but are small compared with the former components.

抄録全体を表示

The authors have investigated the sealing characteristics between the surface of an annular wedge-shaped plastic gasket and a smooth seal face of a rigid ball in static contact. The following results were obtained from gas leak tests, calculation of the deformation of a wedge-shaped gasket and a compression test. ( 1 ) In order to seal tightly, a contacting width of a wedge resulting from the plastic deformation of a gasket was required because of the work hardening of the surface irregularities on the gasket. ( 2 ) The contacting width of a wedge increased with a decrease in the conical half-angle of the seal face, θ. Therefore, the linear sealing load, P_c/l, decreased with a decrease in the θ-values. For example, the value of P_c/l at θ=15°was about 0.5 times that at θ=90°. ( 3 ) The recommended values of θ were in the range from 10°to 35°, and it was possible to forecast the P_c/l-values corresponding to that range.

抄録全体を表示

For the purpose of developing a gas bearing with the highest possible damping, we clarify the damping characteristics of a one-dimensional gas bearing model in relation to the squeeze number, feeding port area, restriction type and land width. The bearing has static stiffness K_st at zero squeeze number and dynamic stiffness K_dy at infinite squeeze number. Between zero and infinite squeeze numbers, we find that the damping becomes negative if K_st is larger than K_dy, and that the maximum damping is proportional to K_dy-K_st. We also find that K_dy is proportional to the static pressure. Thus, in order to increase damping, it is most effective to increase the feeding port area for both increasing the bearing static pressure, i. e., K_dy, and reducing K_st. A simple approximated formula to predict the maximum damping and associated optimal squeeze number is proposed. The limit of the potential damping ratio of gas bearings is also discussed.

抄録全体を表示

Sheet valves used in a reciprocal compressor are subjected to a pulsation of impact force which k results in cracks forming at the tip during prolonged use, and sometimes chipping also occurs due to their connection with the cracks. In this study, the effect of geometrical improvement of the sheet valve on the impact fatigue life was investigated. The impact fatigue life was defined by the number of pulsations to crack initiation at the tip. Impact fatigue experiments were conducted using commercial steel sheet valves and three improved types of valves. A small upward bend in the valve tip increased the fatigue life considerably. An inclined cut at the valve tip did not result in a clear increase in fatigue life. On the other hand, scatterings in fatigue lives of improved valves were somewhat increased compared with that of commercial valves. Dynamic stress states are also investigated by FEM analysis.

抄録全体を表示

The limiting load for scoring and frictional loss of hypoid gears were investigated. Four kinds of hypoid gears with various offsets were chosen for our research. Further various kinds of lubricating oils were used for our experiments, that is, turbine oil, gear oil with additive of Zn-P type, gear oil with additive of S-P type and hypoid gear oil of GL 5. The results of these hypoid gears were compared with those of the spur gears. Some new tendencies of load carrying characteristics of hypoid gears were discovered. Limiting load for scoring expressed with Hertzian pressure decreases with increasing offset. Further limiting load for scoring expressed with Hertzian pressure is dependent upon specific sliding along the line of contact on the tooth surface. However the frictional loss grade is not dependent upon those factors but dependent upon value of sliding velocity or revolutional speed. These new results are summarized by using empirical formulas.

抄録全体を表示

In order to obtain higher bending fatigue strength of gear teeth required to cope with severe driving conditions of gears, shot peening is widely used. It is a very effective method to improve the bending fatigue strength as is well known, but it also roughers the tooth surface, resulting in lowered tooth surface strength. To overcome this weak point, fillet rolling at the root of the tooth is proposed. The surfaces at the roots of S45C gears and austempered ductile iron (ADI) gears are markedly work hardened by this method. Bending fatigue strength of fillet-rolled gears is investigated compared with that of unrolled gears. Bending fatigue strength of S45C gears is improved by up to 33%, but that of ADI gears is lowered by about 9% due to microcracks connecting neighboring graphite grains just below the surface of the tooth face, produced in fillet rolling. Bending fatigue strength of ADI gears, where a 3-mm-thick layer is removed from both side edges by turning, is improved by about 16%.

抄録全体を表示

This paper presents a study on acoustic emission (AE) in the bending fatigue process of spur gear teeth. Research was conducted on AE signals, AE cumulative event count, AE event count rate, frequency spectra of AE signals and crack length in the bending fatigue process of case-hardened and normalized spur gear teeth. The relationship between the AE parameters and the fatigue crack propagation, as well as the effects of the gear material and the case depth on the AE characteristics in the fatigue damaging process of the gear tooth, was determined. Basic data were obtained for the prediction of bending fatigue crack initiation of the gear tooth by means of the AE method.

抄録全体を表示

In order to investigate the effect of a tooth bearing on static and dynamic fillet stress, the fillet strain is measured under various transmitted tooth load in a power-circulating-type gear testing machine for a straight bevel gear, and the measured results are discussed. The results are as follows. The state of the tooth bearing influenced by increase in transmitted tooth load differs in the toe bearing and the heel bearing. At the toe bearing, the dynamic load has a marked effect on the distribution of fillet strain in the tooth trace direction, but it does not at the heel bearing. While the bevel gear runs, there is a change in the position of maximum fillet strain in the tooth trace direction at the toe bearing under various transmitted tooth load.

抄録全体を表示

A program to simulate the state of tooth engagement of strain wave gearing (commonly known as harmonic drive gearing) was developed with a personal computer for the purpose of determining quantitatively the number of teeth in active engagement of the gearing. This program is able to analyze the orthogonal and static state of tooth engagement in the cross section of the gearing. With the use of such a program, the states of tooth engagement were simulated for three kinds of strain wave gearings. From the results obtained, the following conclusions are drawn. The number of teeth in contact increased as the load torque increased. The percentage of teeth in contact was approximately 6%11% (at no load), and 10%17% (at rated load), and the new teeth which came into contact upon increased load torque appeared to be asymmetrical with respect to the major axis of the wave generator, and biased in the opposite direction of the load torque.

抄録全体を表示

The authors have conducted studies on the Oerlikon-type spiral bevel gears for several years. However, these studies are mainly about characteristics of drive gears, and those of driven gears are not yet clarified. In the case of spiral bevel gears, a pair of teeth of drive and driven gears mesh on a concavo surface and a convex one and consequently, it is the detenmined that meshing characteristics of driven gears differ from those of drive gears. In this paper, the authors examined stresses at the tooth root of the driven gears under various running conditions and compared the results with those in the case of drive gears. From this study, it was found that the experimental results of stresses at the tooth root of the driven gears agree with the computed results obtained by means of the W.Coleman equation, and that application of this equation for compution of tooth root stresses of the driven gear is effective.

抄録全体を表示

This paper describes a new type of twist drill. The web of this drill is three times thicker than that of the conventional one. Thus, it has high flexural and torsional rigidity. The main flank and the rake of the central part are formed by a flat grinding method and the drill tip forms a point perfectly. The chip is a conical helix type at first, successively changes to segment type, and then to a long-pitch helix type. In spite of the narrow flutes they do not clog with chips ; all types of chips are exhausted out of the drilling hole very well. Therefore, it can drill a deep hole under continuous feed and with a high feed rate. Under conditions of wet drilling of a S50C strip, a 12 mm-diameter twist drill with a carbide (M20) tip achieved deep drilling of 22 times the drill diameter, which is the limit of its flute length, and a 10 mm-diameter HSS drill achieved a total drilling depth of 80 m (through hole drilling of 60 mm x 1 340 holes).

抄録全体を表示

In a machining-assembly automatic transfer line system which mass-produces machine elements such as bearings, a "matching method" must be considered to produce high-quality elements efficiently. This paper deals with modeling of material flow in the system where the assembly stage considers "matching", and analyzes the effects of the machining accuracy of the parts of elements produced in the machining stages and the buffer capacity on the line efficiency. From the numerical results, it is presented that the buffer storage for the matching part which is selected in the assembly stage is more useful for improving the line efficiency than the buffer storage for the part which selects the matching part, and that the difference in the accuracy distribution of the parts between machining stages has an important effect on line efficiency.

抄録全体を表示

In the previous report, it was shown that harmonic analysis results on the three-dimensional shape of a steel ball obtained by use of spherical functions gave good evaluations of characteristics of steel balls used in a bearing assembly. However, this type of analysis is inadequate for evaluating the shape of the ball geometrically. In this paper, first, the possibility of three-dimensional geometrical evaluation by spherical harmonic analysis is described. The 'reference direction' is defined as the index for directivity of each order of the spherical component and has no relationship with the coordinates for measurement. This 'reference direction' or the value of each spherical component for which this direction is obtained, is utilized to illustrate and evaluate the spherical shape of the steel ball. Second, the method for estimating the three-dimensional shape of the steel ball by two-dimensional measurement is explained (e. g., a roundness meter). Mean Fourier coefficients of cross-sectional forms of the steel ball are theoretically calculated, and the correlation between this value and the harmonic analysis result obtained through the use of the spherical functions is obtained.

抄録全体を表示

It is important for precision profile measurements in on-machine conditions to construct some accurate datum for scanning. A kind of inclination method using three or more displacement sensors was proposed recently for this purpose. However, its characteristics in relation with the spatial frequencies of the profile have not been considered precisely. The z-directional resolution power of the method has not been strictly estimated. In the present paper a new and general expression of the multiprobe methods, including the inclination method, is derived to estimate the errors in relation to spatial frequencies. In all these methods, a similar datum of the parabolic curve is introduced by data processing, which is called the software datum. The influence of accidental errors is theoretically clarified in relation to the spatial frequency of the probed profile. The meaning of zero adjustment of the sensors for software datum of the scanning method is also clarified. A procedure for achiering good zero adjustment is also presented.

抄録全体を表示

This paper attempts to build up a diagnostic expert system for ultrasonic testing (UT) which inspects the weld defect encountered in welding structures. The proposed system can handle knowledge and data which include various kinds of uncertaities and ambiguities by use of fuzzy set theory. The knowledge is represented in terms of fuzzy production rules involving certainty factors, which help accommodate reasoning with uncertainty. The ambiguities of knowledge and data are represented by linear membership functions. The computational time required for inference is reduced because the proposed system uses the blackboard model dividing the knowledge source and working memory. In addition, the system has the ability to explain the reasoning that led to solutions of diagnosis. The results obtained by the proposed expert system are compared with those by UT experts, and the usefulness of the system is denoted by high consistency in the comparison.

抄録全体を表示

A decision-making method has been proposed for distributed dynamic task scheduling controllers that involves automatic acquisition of scheduling rules through learning. The proposed method consists of stochastic rule-based decision making and its learning modification by stochastic matrix learning automata (SLA) theory : The action of determining a machine to be allocated is performed independently at each of the distributed agents, where the stochastic rule that consists of a condition and multiple actions is applied and one action is selected according to the probability vectors. SLA theory is applied to the learning modification of these probability vectors, which is proven to maximize a given objective function under a certain environment. To ensure the effectiveness of this learning for dynamic task scheduling, computer simulations are conducted, the results of which show that the method can obtain the same performance as that of a conventional dispatching rule, and that under a highly stochastic manufacturing environment, it can obtain hither performance than that using the dispatching rule.

抄録全体を表示

An approach to the scheduling of a shop Floor which may involve assembly machines is proposed. The objective of this approach is firstly to maximize productivity by avoiding idle time on the bottleneck machine which has maximum workload, and secondly to minimize average net flow time by making consecutive operations as close as possible. Initially the operation sequence on the bottleneck machine is determined by applying a dispatching rule. On the basis of this sequence, remaining operations are assigned to the specified machines using forward and backward methods to minimize average net flow time. As the initial schedule does not consider the conflict, the human scheduler modified this schedule interactively until it becomes feasible. From the experiment the following four points are clarified : ( 1 ) schedulers can change the initial schedules to feasible ones steadily, ( 2 ) performance of the minimization of idle time and the average net flow time largely depends on the scheduler's scheduling ability, ( 3 ) schedulers having aptitude for scheduling often seem to produce near optimal schedules, and ( 4 ) the proposed approach is suitable for shop floors where the workload ratio of nonbottleneck to bottleneck machines is small.

抄録全体を表示

In this paper, we propose a new method of designing a digital controller. We consider the design problem for a two-degree-of-freedom control system with a feedforward controller. First, we design a cascade controller in order to obtain a desirable sensitivity and stability margin. Then, we design a feedforward controller so that the resulting control system may satisfy a specification with respect to the reference command. The proposed method is based on minimizing a frequency-weighted error so that the resulting control system approximates the specified reference model. A square error criterion is introduced as a measure of the approximation. By minimizing the criterion, a linear equation is obtained for parameters of a FIR (finite impulse response) controller. Using the method, we can design a control system by a low-order controller which realizes the desirable accuracy of the approximation without regard to the order of the plant or the reference model.

抄録全体を表示

The present paper refers to the studies on the frictional characteristics between piston rings and cylinder liners in reciprocating internal combustion engines. An improved floating liner method by means of hydrostatic bearings has been developed to measure accurately the frictional force as a function of crank angle under the firing operation. The major results are that : ( i ) the measured frictional force shows the boundary or mixed lubrication near the dead centers, but shows mainly the fluid lubrication during mid-strokes ; ( ii ) the total frictional loss is reduced, but the peak friction just after the dead centers increases with the decrease in engine speed or oil viscosity ; and (iii) the peak friction at the beginning of the expansion stroke increases remarkably due to the gas pressure in cylinder, and this increasing tendency is promoted according to the reduction in the number of piston rings.

抄録全体を表示

A surface flattening test by inside-ironing was carried out for hot-extruded aluminum pipes. The pipes have many scratches with a depth of about 1μm formed by extrusion on their outer surface. A simulation test was also conducted using pipes with V-shaped grooves of about 14 μm depth which were preformed by a diamond tracer. There exists a difference in flattening process between the scratches and the grooves due to the difference of their initial configuration. The cross-sectional curve of the scratches consists in low-frequency waviness and high-frequency roughness, and the waviness is almost flattened at an earlier stage, followed by the roughness. The scratches thus disappear. The effects of the relative slip between metal and tool surface on the flattening of the scratches should be further clarified.

抄録全体を表示

Two types of tool path generation algorithm are proposed for the machining of a rotating workpiece with the free-formed surface using a cylindrical tool. The first algorithm is applied to the free-formed surface formed on a hard model. In this case, the hard model is measured by a cylindrical probe contacting its side surface. The tool path is calculated from the locus of the cylindrical probe center. The second algorithm to generate the tool path is developed for the free-formed surface defined by a set of numerical data. On each calculating system, the cutter interference and the minimum radius of curvature of the free-formed surface's projection curves are calculated. The accuracy of calculation on the radius of curvature is also discussed. A golf club is used as an example of the free-formed surface.

抄録全体を表示

A method for distributing tools to each machine in a 2-machine flowshop-type FMS without duplicated tools is proposed when there are precedence constraints on job operations. All of the tools required for processing various kinds of jobs in the FMS are allocated to each machine according to the from-to chart which represents the technical precedence constraints on the tools. Some simulation results show that when the number of jobs is large and the frequency of utilization of each tool is uniformly distributed, the utilization of completely separated functional FMS (CSF-FMS) designed by this method is as high as that of the full functional FMS (FF-FMS) where each machine has all of the tools. It is also made clear that when the number of jobs is small and the number of job operations is exponentially distributed, the utilization of the CSF-FMS is far less than that of the FF-FMS.

抄録全体を表示

A method for distributing tools to each machine in a 2-machine flowshop-type FMS with duplicated tools is proposed when there are precedence constraints on job operations. The duplicated tools are selected according to the from-to chart which represents the technical precedence constraints on the tools. Some simulation results show that the proposed method is effective for increasing the utilization of the FMS, especially when the utilization of the completely separated functional FMS is low.

抄録全体を表示