This paper describes a technique to suppress vibrations of a structure under excitation by attaching a hollow rotor partially filled with liquid, whose shaft axis is parallel to the direction of the exciting force. Assuming the fluid motion is axially symmetric, we analyze the motion of the fluid and structure. The influence of the shape of a cavity of the rotor, the amount of liquid and the ratio of rotational speed to exciting frequency on suppressing vibrations of the structure is investigated. The condition in which the rotor partially filled with liquid acts as an dynamic absorber is clarified. The effectiveness of the proposed dynamic absorber is demonstrated experimentally. Theoretical results agree with experimental ones.

抄録全体を表示

A small rectangular test tank, installed with two vertical pipes at both sides and filled with water flowing due to water suction and ejection through holes in the pipes, was sinusoidally excited to generate sloshing oscillation in the flowing water. The frequency response characteristics of the sloshing amplitude were experimentally obtained for five flow patterns of water in the tank, and the natural frequencies and modal dampings of the 1st, 2nd, and 3rd modes were estimated using a conventional modal analysis technique. The results showed that sloshing natural frequencies decreased due to the centrifugal force generated by water surface flow velocity, acting on the sloshing liquid, and the dampings increased due to kinetic energy dissipation caused by a steady flow at the water surface. Thus the sloshing oscillation was markedly suppressed at a critical flow velocity on the water surface.

抄録全体を表示

In the previous papers, a survey was reported on the vibrations of electrical concrete breakers and two kinds of vibration reduction systems were adapted for them. This report presents the theoretical analysis and the digital simulation for them, based on the previous vibration and displacement measurement. Such an analysis has not been performed yet. Acceleration and displacement wave forms could be made graphically by a personal computer. Therfore, it was confirmed that the vibrations at the handle, resulting from the torque of the connecting rod, were almost negligible in comparison with those from the reaction force of the compressed air.

抄録全体を表示

The nonlinear oscillations of a levitated body, which is moved in a horizontal plane and is guided by magnetic forces on both sides, are examined theoretically and experimentally. The magnetic forces acting on the body are calculated exactly by applying Coulomb's law over and infinitely small area of magnets, and by integrating about four magnetic poles of two magnetic plates. The parametrically excited oscillations of the body, due to the relative vertical motion between the body and the magnetic guides, are studied theoretically by using the above magnetic forces. Also the experiments are preformed with a simple experimental apparatus. As the main results, the oscillation with a fractional-harmonic pair as predicted by the theory, is shown experimentally in the special case when two natural frequencies of the body are nearly commensurable.

抄録全体を表示

For attaining rotatory high speed in actual rotating machinery, it is necessary to control many critical speeds, that is, many resonances of the machinery, using an active control bearing system. This paper investigates a design method of the active control bearing system consists of a magnetic bearing and its controller. The essence of this method is to design the controller based on the transfer function of whole structure which include flexible shaft and casing obtained by using both methods of substructure synthesis and modal analysis. In order to get suitable transfer function for controlling resonances, therefore, it becomes easy to determine the desired structure through modification of a substructure optionally. In this paper, the effectiveness of this design method indicated by theoretical approach, will be demonstrated through experiments.

抄録全体を表示

An attempt is made to develop a new three-dimensional tympanometer, and the dynamical characteristics of the middle ear of normal subjects are measured. At the same time, by applying the combination of the impedence theory of the tube, and the kinetic and strain energy theory, the equation of the middle ear is induced, and the effects of each part of the middle ear upon its dynamical characteristics are clarified by means of numerical calculations. In this paper, emphasis is put upon the theoretical analysis, and the mechanism of the tympanometer and the results of the measurement will be discussed in detail on another occasion.

抄録全体を表示

A device using seal rings was designed to suppress leakage from both ends of a fluid coupled restraint with an annular region in this study. Then an analysis was preformed to evaluate vibration characteristics of the restraint, such as added mass and added damping. As a result, it was proved that added mass increases as the gap thickness formed by the seal rings becomes smaller, and that added damping becomes maximum at the optimum gap thickness. Also, the validity of the analytical method was verified, comparing the analytical results with experimental results.

抄録全体を表示

To clarify the vibration characteristics of an elastic rotating shaft system, we should take into consideration both the elastic asymmetry of the shaft and the asymmetric elasticity of the supports. Analyses were made using the asymptotic method, and then were numerically calculated by the Newton-Raphson's method. In this report, we discuss mainly the 2nd higher harmonic response of the system. As a result of the analyses, it was confirmed that some obtained phenomena were different from those of the main resonances.

抄録全体を表示

This study deals with the coupled lateral-torsional vibration of a two-stage gear train system. First the equations of motion are derived, which have periodically varying parameter, and the solutions of equations are obtained analytically and characteristics of the vibration of the system are analyzed. The response to the forced vibration under static load is analyzed and some new results are obtained. Some resonances of dynamic tooth load appeared when the meshing frequency equals to fractional natural frequency. In some cases the phase angle between two meshing stiffneses influence the amplitude of dynamic tooth load. Tooth contact ratio also influences the amplitude of dynamic tooth load. By considering these results it is known that some resonances of the system can be avoid by adopting an adequate phase or contact ratio.

抄録全体を表示

This paper is concerned with a new technique for complex mode synthesis for general damping systems. The complex mode synthesis for general damping systems with fixed interface leads to difficulty in constructing rigid body modes for substructures possessing rigid body motion. Therefore, the equations of motion for rigid bodies are added to the complex mode synthesis, and substructures with a free interface complex mode are synthesized. Firstly, the technique is applied to a spring-mass-damper model with 4 degrees of freedom and the result is discussed. Secondly, the technique is applied to a 3 phase compact motor, in which the motor is divided into four substructures. The transfer function for each substructure is measured by impact testing, and the modal parameters, such as eigen values, damping factors and complex modes, are extracted by the curve fitting procedure. The transfer functions of the total structure of the motor are analyzed by the proposed method. These calculated results are compared with the experimental ones, and the effectiveness of the method is verified.

抄録全体を表示

In previous papers the authors presented a new technique of identifying nonlinear multi-degree-of-freedom systems, and showed that the technique can be used for identifying systems with smooth nonlinearity. To see whether this technique can be generalized and used for identifying systems with non-smooth nonlinearity, some systems subjected to Coulomb's damping forces are taken up. First a generalization for identifying such systems is presented. Then, using data generated numerically from the equations of motion of the systems, the applicability of the technique is checked. It is shown that the technique can in fact be used for identifying the systems. It is also shown that under noisy measurements a good accuracy can be obtained if the excitation frequencies are chosen near resonant points and if the first or second order terms are retained in the Fourier series expansion necessary in the process of identification.

抄録全体を表示

A method to estimate the dynamic reliability of hysteretic structures under combined time varying loads is investigated. The response statistical moments of structures, with hysteretic characteristics approximated by Wen's mathematical model, are obtained by using the Fokker-Planck equation method with a cumulant truncation technique. Furthermore, first passage probability considering stochastic load combinations is derived by the load coincidence method under the assumption of a Poisson process for threshold crossings. The uncertainties of power spectral densities pertinent to loads as stochastic processes, are also considered through the derivation of first passage probability. This paper, as a 1st report, describes the analysis of a single-degree-of-freedom system, and demonstrates that resulting variances and covariances of the response agree well with Monte Carlo simulation.

抄録全体を表示

This paper describes a general formulation for the in-plane flexural free vibration analysis of a multiple layered structure by the transfer influence coefficient method. The structure is modeled as a discrete system with lumped mass, lumped inertia moment, and massless linear and rotational springs. The present method does not require changing the fundamental algorithm, but necessitates only substituting appropriately large values into the corresponding spring constants when intermediate elastic supports are many and very stiff. Boundary conditions are also controlled by the spring constants. Amounts of arithmetic calculation are compared between the transfer influence coefficient method and the transfer matrix method. The occurrence mechanism and the simplified solution of method and the transfer matrix method . The occurrence mechanism and the simplified solution of false roots involved in the frequency equation are discussed in regard to the use of the bisection method.

抄録全体を表示

The authors apply the general formulation for the in-plane flexural free vibration analysis obtained in the previous report to a structure with two layers regarded as a discrete system. They describe the influence coefficient matrix at the left hand end of a vibrational system, and the rules of field and point transfers explicitly, and state the effective method of numerical computations suitable for the remarkably stiff intermediate elastic supports between layers in detail. The results of the simple numerical computational examples demonstrate the validity of the algorithm obtained in the previous repot, the is, the numerical high accuracy, the high speed vibrational analysis and the flexibility for programming of the transfer influence coefficient method, compared with the transfer matrix method on the personal computer.

抄録全体を表示

Acoustic characteristics of ducts or silencers are usually analyzed by the transfer matrix method in the frequency region where the plane wave theory is efficient. In this method, the precise estimation of the transfer matrix is important. The experimental acoustic duct analyzing method is also required, which can precisely predict the acoustic characteristics of a total duct system, due to those data measure on individual duct element. For the purpose of measuring the propagation constant, the acoustic impedance and the transfer matrix of any duct element, a new four-microphones method has been developed in this report. Based on the experimental results, this method has been proved to be sufficiently reasonable especially for analyzing acoustic characteristics of complicated duct systems.

抄録全体を表示

This paper presents a procedure for obtaining the source impedance of fan-duct systems. A new parallel circuit model and the two-load method were introduced for measuring the source impedance. This model could be used to establish the conditions necessary to achieve the precise estimation of the source impedance. Having established the basic relationships of acoustic characteristics, the measurements were employed to check the applicability of the parallel circuit model to source impedance specification. It was recognized that the postulated method provided useful comfirmation of the accuracy to be expected. In particular, sound insertion losses of a silencers based on measure source impedance were shown to coincide with the measure values.

抄録全体を表示

In this paper, a new method for structural dynamic modification and a way of combining the new method and the substructure synthesis method are proposed. In most cases of structural dynamic optimization problems, the number of natural frequencies to be modified is generally less than that of the structural modification parameter, for example the finite element thickness, in the structure. Therefore, in the past, these kinds of problems could not be solved practically, in general, or the methods were so complicated that it took too much CPU computer time for the calculations. The new method in this paper can solve the problems using much less calculation time. Furthermore, the efficiency of the analysis can be improved by combining the method's algorithm and the component mode synthesis method, which was proposed by the authors as one of the substructure synthesis methods. After the explanation about the proposed methods, three applications are shown, a cantilever plate, a structure in a character H shape and an actual fly-wheel for automobile engines. The results of these applications show the practical usefulness of the proposed methods.

抄録全体を表示

The transient response of the truck crane structure due to winch operation when the loading boom is operated in the direction of the crane frame axis has been treated by modal analysis. The boom and crane frame are constructed of several box beams with different cross-sectional areas and weight distribution, and the crane frame is supported at both ends with the outriggers. The transient vibration stress at the lower boom is observed in the practical track crane and its numerical simulations carried out. Then, relationships between the winch operation and transient vibration stress level are presented for the wind speed patterns of the winch.

抄録全体を表示

This paper describes a vibration modeling procedure for an arbitrarily shaped elastic structure using FEM. Based on this model, a method of controlling the system in which a sensor is not collocated with an actuator is also investigated. In this procedure, the controlled modes can be selected by taking the transient response amplitudes of each mode into account. A motion and an observation equation are derived employing these controlled modes as state variables, and the control law is determined using an optimum control theory. The influence of residual modes such as spillover on the control system is illustrated by numerical studies on the feedback control of an elastic rectangular plate. Based on these results, modeling criteria are obtained for reducing the effect of spillover.

抄録全体を表示

A dynamic analysis concerning the automatic paper feeder of a facsimile is described here. Including the step-motor system, the equations of motion for the each mechanical element system are introduced. The contact friction forces of the friction clutch and one-way clutch, the friction forces between a roller and a paper or a paper and paper, and furthermore the elastic forces of the ring spring and its clearance when built in, or the belt forces are described considering nonlinearity. The dynamic characteristics concerning the variable kinds of function of the auto paper feed of a facsimile which are obtained as a result of the numerical analysis agree well with the practical dynamic motions, and the propriety of the motion description is obtained.

抄録全体を表示

In this paper, a new design method for a preview control system is proposed. This method is based on a partial model matching technique and the whole controller design is to be accomplished in two steps: the first step is for the feedback controller and the second for the preview controller given the feedback control system. Any technique may be used for determining the feedback controller. The preview feedforward gains for processes with predetermined feedback controllers are determined so that the total system is almost equivalent to the given system. Simulation examples are presented to show the effectiveness of this method. Lastly, this method is implemented to a controller of a single joint flexible manipulator. The result of this experiment confirms the computer simulation results.

抄録全体を表示

This paper discusses a general form of a type-1 deadbeat serve system which nullifies the effect of the step disturbance and transfers the output to the step command signal within a finite time for a discrete-time linear time-invariant system. The type-1 deadbeat servo system is formulated by using the augmented system and its deviation system. Then, a general form of the type-1 deadbeat servo system is obtained. The controllability indices of the system are (n₁, …, n_m), where n₁≥…≥n_m. When the order r of the output is equal to the order m of the input, the controllability indices of the deviation system are (n₁+a, …, n_m+1) and the minimum settling time is n₁+1. When r<m, the controllability indices of the deviation system are not determined only by (n₁, …, n_m), and the minimum settling time is either n₁ or n₁+1.

抄録全体を表示

In this paper the design of a crane system which carries an unknown load weight beyond an obstacle in an optimal way is described. At the first stage of the control a load of unknown weight is wound up and an adaptive identifier estimates the unknown value from the input and output data of the winding mechanism. The identifier is designed to have only one unknown parameter so that quick convergence is obtained, which makes it applicable to practical systems. Then, the optimal regulator is used as a controller of the crane, where the optimal feedback gain is a function of the load weight and the estimated value is used. Using this controller, performances of the crane are achieved such that i) and unknown load is carried beyond and obstacle, ii) the crane is stopped at the destination with good accuracy and with no swing of the load, iii) the travel time is shortened. Thus, automatic and optimal operation of the crane system can be preformed without pre-information about the value of load. Details of the design method and the experimental results are given.

抄録全体を表示

For moving in indoor space, it is generally recognized the biped locomotion is suitable. This paper proposes a hierarchical control strategy for the lower level where the position control or the force control at each joint is implemented. In the upper level control, the robot motion is divided into a sagittal plane and a lateral plane. We applied the optimal control algorithm to the motion control in the lateral plane in order to improve the robustness of the control system. The effects of these control schemes are shown by the experiments using the new walking robot BLR-G 1 and the parallel calculation system. BLR-G 1 has 9 degrees of freedom and equips the foot-pressure-sensors and a rate gyroscope. Complete dynamic walking is realized, in which the cycle for each step is about 1.0 second.

抄録全体を表示

This paper deals with the coordination force distribution at finger joints for an articulated hand, under the assumption that the articulated hand with multiple degrees of freedom and a multiple articulated joint structure grasps some objects. In this study, four algorithms for this force distribution problem in grasping some object are derived from the static equilibrium conditions. Furthermore, a method for changing contact and non-contact fingers in grasping is derived and some simulations are shown, where changing fingers implies that some non-contacting fingers begin to make contact with the object, while the other contacting fingers do not contact it, keeping the stable grasping conditions by hand as a whole. Finally, based on the proposed five control algorithms, a control flow of stable grasping methods for the force distribution at each joint is made for an overall control of the finger system by determining grasping forces to changing grasping fingers.

抄録全体を表示

A power collector is presented, which has a control mechanism of contact force between a contactor of the power collector and a contact wire. By applying the optimal control theory, the optimal control laws with a feedforward link of wire unevenness are derived for two cases of sliding on a rigid contact wire and an elastic contact wire which is simply modeled as an infinite beam in this study. The feasibility of the power collector is considered by performing numerical calculations and an experiment. As a result, it was made clear that the optimal state feedback control of the power collector is useful for the case of a rigid wire, but not useful for the case of an elastic wire. In other words, in the case of the elastic wire, the feedforward control of disturbance is indispensable and the detection or the estimation of wire unevenness is an important technical subject.

抄録全体を表示

In this paper, we discuss the control of the flow out quantity of a known amount of granule from a hopper. First, for the control algorithm, the model reference adaptive control algorithm extended to a non-minimum phase system by compensating zeros of the system was used. Next, adaptive algorithm for a process with unknown deterministic disturbances was investigated. As for the results, it is shown that the proposed algorithm is very useful for the process with a deterministic disturbances and with time variant parameters.

抄録全体を表示

To measure the flow stress of rod or wire under forming conditions, i. e., at large strains, high strain rate and elevated temperature, a method based on the side compression test is proposed. In this method, plane-strain deformation is supposed and rod and wire are compressed between very rough tools on which grooves are made to the axial direction. The load and reduction in height measured in the test are converted into average flow stress and average equivalent strain by using the result of the rigid-plastic finite element method. This method is simple to carry out and is free form the effect of variation of frictional condition between the rod (wire) and the tools. Application of the method to continuous measurement of flow stress is investigated by compressing a part of wire. It is confirmed that the present method is suitable for determing flow stress-strain curve of rod and wire efficiently.

抄録全体を表示

A dynamically scaled model, which simulates quantitatively the property at twice the testing speed of 160 km/h, is used to make clear how the performance of the existing system deteriorates with an increase in speed, concentrated mass or a decrease in tension of the contact wire and how the system can be improved. Contact loss and stress of contact wire become so large that it is considered impossible to realize satisfactory current collection at a speed of 300 km/h with no improvement. It is also found that these factors depend mainly on the wave propagating velocity of contact wire and can be eliminated by introducing a higher tensioned or perhaps a lighter contact wire.

抄録全体を表示

Many experiments have been made to clear the true state of adhesion coefficient μ which significantly affects the performance of rolling stock. However, such experiments have been made mainly in laboratory rooms, and the experiments made on main lines have been simply a medley of data only at the spot of wheel slipping. The author employed a "Slipping adhesion test truck" built on a new idea of realizing the measurement of μ continually while running, and observed the true state of μ along the mainlines. From the results thus obtained, the author derived the statistically processed μ, and analyzed the relation between μ and the rate of wheel slipping. In addition, the author referred to the various factors that reduced the value of μ.

抄録全体を表示

In this report, the modified averaging approach to the solution for the dynamic performance of an externally pressurized gas-lubricated circular thrust bearing in a laminar flow regime is pressented under the assumption of a small harmonic vibration. This approach can evaluate the inertia effects under highly unsteady conditions. Compared with this, the accuracy of the approximate solutions presented in the previous report is evaluated under such conditions and the following conclusions are obtained: (1) The conventional averaging approach in which all the inertia terms are averaged out across the film thickness gives a fairly good approximation. (2) The first-order solution of the perturbation method becomes remarkably inaccurate as the unsteadiness becomes high.

抄録全体を表示

A computational method for analyzing the spacing between a tape and the video-head of a rotary head recording device, such as VTR, R-DAT etc., is described. This kind of study has been carried out, so far, without considering the direct contact of the tape and the head. We can, however, observe that the head wears as it moves with the tape. It follows that the contact of these materials should be take into account. We introduce a mathematical model of this contact which is obtained in consideration of the roughness of the material surfaces. It is of particular interest that this model is effective to analyze the spacing and the pressure when the spacing might become narrower than the material surface roughness. The calculation results show a good agreement with the experimental data, and it is concluded that the contact model is useful and accurate.

抄録全体を表示

In the previous report we examined the characteristics of composite plastic gears filled with glass fibers, and pointed out the existence of some defects in the gear in terms of abrasion resistance. In this study, we prepared composite plastic gears filled with carbon fibers for the same purpose as for the glass fibers, and made running tests to evaluate the gears in terms of abrasion of the teeth, temperature of the teeth, roughness of the tooth surface, change of the tooth profiles, and noise of the gears. Comparing with the test results of the other plastic gears we also studied the practicability of these gears.

抄録全体を表示

This paper presents a study on the deflection and root stress of WN (SymMarC) gear tooth. The tooth deflection and root stress of a SymMarC gear tooth due to a concentrated load were calculated by the three-dimensional finite element method (FEM) and compared with those measure. The validity of the gear model and constraint condition used in this FEM analysis was confirmed, and the critical section of the SymMarC gear tooth and the effects of the number of teeth and the face width on the tooth deflection and root stress were clarified. Furthermore, on the basis of these results, the approximate equations for the deflection and bending moment due to a concentrated load on a SymMarC gear tooth were derived.

抄録全体を表示

The heat transfer coefficient on tooth facings must be known when calculating temperature rise on tooth faces. Though analytical cling-off cooling of gear teeth and experimental results about the adhesive amount of oil have been already reported. There have been few reports about heat transfer coefficients on tooth facings based on experiments and calculated from actual conditions. In the present paper, The temperature of tooth faces and of oil was measured, so the heat transfer coefficient was evaluated based on the temperature difference between the tooth face and oil. And the heat transfer coefficient concerning two conditions was evaluated analytically.

抄録全体を表示

The amount of upset during friction welding is a quite important criterion of optimum friction welding conditions because it closely concerns the mechanical properties of a welded specimen. However, it is very difficult to estimate the amount of upset before friction welding, because the friction welding process is extremely complicated. If the amount of upset during friction welding can be estimated beforehand, it is expected to extend the usefulness and application limits of friction welding. In this study, first of all, a nonuniform upsetting process firing friction welding under various friction welding conditions were correctly measure. Next, the formulation of experimental equations for the amount of upset in connection with friction welding conditions and kinds of steel (SC steel) was made. Two experimental equations were calculated. One is based on the amount of upset considering a nonuniform state during friction welding, and the other is based on the total amount of upset after friction welding. Both equations agreed well with experimental results.

抄録全体を表示

This paper presents a boundary element methodology for the numerical solution of bolted connections where a clamped plate is a T-flange and there is a flat unstressed surface. The boundary element technique using Melan's solution is particulary well suited to this type of problem and reduces the data preparation necessary to solve the problem with a free flat boundary. In this paper, Melan's solution was briefly examined and the necessary fundamental solutions derived. Then, discussions were made on the distribution of contact stresses and the force ratio (a ratio of an increment of axial tension of a bolt to an external load). The numerical results are in fairly good agreement with the analytical ones obtained by the theory of elasticities.

抄録全体を表示

The purpose of this paper is to report that a weighted upstream scheme for the gas-lubricated Reynolds equation which is stable and highly precise for a wide range of bearing number Λ. The following two methods of calculating the weight of the upstream have been presented. (1) The exponential method: The weight is given by analytical integrations of a one-dimensional Reynolds equation. (2) The stabilization method : The weight is given by the condition under which the one-dimensional discretized equation for node pressures is stable. The investigation results for infinitely long--stepped slider bearing and optimum weights are decided. These optimum weights have been certified to show that the optimum weight for a one-dimensional is the optimum weight for a two-dimensional. It has been clarified that the stabilization method is more precise and more economical than the exponential method.

抄録全体を表示

Wrinkling during the ironing of very thin wall aluminum cups has been investigated. Three types of wrinkly were observed on the finished cup wall, i. e., a circumferential single wrinkle, and short-and long-pitch wrinkles generated within the band areas of the wall. All are a kind of buckling at the die exit due to the axial compressive stress acting on the ironed wall metal. The enhancement of the stress is caused by the decrease in the die surface friction which is induced by a metal bulging just before the die entrance. The wrinkly type is thus subject to the bulging behaviour affected by the initial thickness of the cup and the ironing reduction ratio.

抄録全体を表示

A flexible forging method for the production of disk-shape products in which the billet is successively compressed by a pair of upper dies is proposed. The distance between the upper dies, the ram stroke and the angle of rotation of a lower die are controlled by a micro-computer. To examine the problems associated with this method, disk-shape billets of plasticine are compressed on a small-size flexible machine. It is observed that flows are caused over the compressed surface of the product without large flaws is obtained when disks are repeatedly compressed with a small reduction in height by dies having a round corner, whose radius is greater than the reduction in height. The non-uniform deformation on the circumferential surface is effectively reduced by employing a large rotation angle after each compression.

抄録全体を表示

A new method of tube bending is proposed and the features of the bending machine made on an experrimental basis are examined. In one of the traditional methods of tube bending, the workpiece is pressed onto a rotating form by a pressure die. Essentially, the new idea in the proposed method is that a drawing die is used instead of the pressure die such as in the method of press bending. As experimental parameters, the diameter, of the tube, inclined angle of the drawing die and guide angle of the workpiece are chosen. The following features of the machine are verified experimentally. (1) All of the workpieces are bent precisely so as to hold a prescribed radius of curvature. The roundness of the outer fiber in the plane of bending is superior to conventional methods and the springback is observed to be negligible. (2) Concerning the tube shape in a crosssection, the workpiece larger in diameter than the drawing die can be bent without flattening, while it is flattened as usual when the workpiece is small and free from the drawing die. (3) The thickness strain is positive at the inner wall for bending and negative at the outer one. The difference between the maximum and minimum thicknesses in an arbitrary crosssection increases when the tubular diameter is larger.

抄録全体を表示

A shape design problem is studied for a linkage mechanism used for power shovels based on a mathematical optimization method. With the object of investigating the dynamical relationship of the mechanism, the Lagrange equation of motion is derived by adopting the algebraic programming language REDUCE. The design problem to determine the optimal shape of the linkage mechanism is then formulated as a nonlinear optimization problem of the minimax type; i.e., the sum of the respective maximum values of three driving forces is considered as the objective function to be minimized under constraints related to the dynamic of the mechanism. An algorithm combined with simulation and optimization calculation is proposed to solve the problem numerically, and it is ascertained that the shape of the mechanism can be designed rationally by adopting the method proposed here.

抄録全体を表示

This paper deals with a three-dimensional deformation analysis of workpieces mounted in fixtures for the case where workpieces are finite solid cylinders. The analysis is carried out by a three-dimensional theory of elasticity regarding clamping forces caused by chucks as an axi-asymmetrical load. In numerical computations, the effects of a number of chucks, the diameter of workpieces and the grasping area on the roundness of workpieces are shown. For verification, experiments by three-dimensional photoelasticity are performed. In addition, the deformation of workpieces is measured by an electric micrometer. The analytical results are in fairly good agreement with the experimental ones.

抄録全体を表示

The outside of a workpiece like a disk is turned and orthogonally cut with a cutting-off tool fed by the cross spindle of a lathe, and the workpiece is assumed to be a system with two degrees of freedom. Numerical calculations of the equations of motion for regenerative chatter and also cutting tests are carried out. As a results, it is shown that the regenerative chatter is initially caused by a small disturbance, it can be classified by a width of cut into disappearance, constant amplitude and growing amplitude. Further, the behavior of chatter is compared with the stability criterion that is given by a set of harmonic solutions of the equations for regenerative chatter, and it is verified whether the chatter disappears or grows in its amplitude is determined fairly by the stability criterion.

抄録全体を表示

The effects of the rake angle and the roundness of the cutting edge on the static and dynamic cutting forces are researched experimentally with a tool whose rake angle is negative. And the result is compared with the result which the authors have already reported for a tool whose rake angle is positive. The following findings were obtained: (1) the increasing rate of the horizontal cutting force to the depth of the cut changes at two points; the first change is caused by the roundness of the cutting edge and the second change is cause by the negative rake angle; (2) chatter vibrations occur at these two points; (2-1) at the first point, the same chatter vibration as already reported by the authors occurs; (2-2) at the second point the negative rake angle makes the chip flow unstable, which is one of the causes of primary chatter vibration.

抄録全体を表示

In the case that the cutting edge leaves a workpiece, the effects of the roundness of the cutting edge and the depth of the cut on the dynamic cutting forces are researched experimentally with two tools whose shapes are different from each other. The following findings were obtained: (1) with both tools the same characteristics of vibration and cutting forces appear in this primary chatter; (2) the non-linearity of cutting forces to the depth of the cut changes the figures of dynamic cutting forces under the condition of a gradually increasing depth of the cut; (3) the penetration effect in the rubbing region differs from that in the cutting region; (3-1) in the rubbing region the shear angle and the chip flow changes in the dynamic condition, which is the cause of this primary chatter: (3-2) in the cutting region, the effect of the interaction between the chip and the rake face is predominant and controls the chatter.

抄録全体を表示

This paper studies an approximate procedure to solve a production planning problem for a multi-stage production system producing many different parts and assembling them into products under a capacity limitation. A generalized model of the production planning problem is formulated as a mixed-integer programming problem, and is approximated as a linear programming problem by applying the Datzig-Wolfe decomposition principle. Then an approximate algorithm is developed in detail for a two-stage production system. Many numerical examples are solved to examined the validity and efficiency of is proposed algorithm.

抄録全体を表示

The performance of four typical aggregate production planning methods, i. e., Linear Programming Model (LP Model), HMMS Linear Decison Rules (HMMS LDR), Goal Programming Model (GP Model), and Production Switching Heuristics (PSH), in an uncertain production situation which has completer demand forecasting and some unexpected orders was investigated from an experimental study. The results showed that (1) HMMS LDR and PSH in a stable production situation and the GP Model in the situation with intensive fluctuations have good adaptability; (2) the LP Model leads to results with low total production cost but intensive workforce level fluctuation in all situations; and (3) the GP Model gives poor performance in the situation in which distinct interference and/or advantageous differences between the objectives are not recognized.

抄録全体を表示

This paper deals with an optimal control problem of dispatching parts into an automated manufacturing system. The system consists of the same kind of machines located along a loop type conveyor and a loading-unloading station with buffer space. The optimal control problem is formulated as a Markov decision process. The properties of optimal control and the system production rate under optimal control are considered with respect to the effect of the presence of the buffer space.

抄録全体を表示

This paper deals with an optimal part selection problem in an automated manufacturing system in which a flexible manufacturing cell produces different parts for several production lines. The optimal control problem is formulated as an undiscounted semi-Markov decision process with the cost structure consisting of shortage cost and holding cost in the lines. In order to minimize the expected cost, the cell must decide what type of part to fabricate next at each epoch of the cell completing a part. Numerical results are shown in order to study the sensitivity of the optimal solution on the shortage cost and the holding cost.

抄録全体を表示