Design methods for gear drive devices are well described in texbooks and handbooks, since the type of such design problems are recognized as routine design. Though the design methodology for individual machine elements has been formalized, it is still difficult and time-consuming to arrange total systems, such as gear drive devices, which are composed of a set of machine elements. In this paper, we develop a computer-aided design system for the above design problems. In the system, design knowledge for the individual machine elements is modularized as "functional modules" in the manner of object orientation, and the operation and control knowledge for design are integrated with it based on the machine structure defined by the designer. The system can deal with several classes of gear drive devices and automate the various design calculation in such design processes. Finally, we show examples of its application to design problems of gear reducers.

View full abstract

Recent advances of micromachine technology are attracting much attention in all sectors of the society. This suggests that people anticipate that micromachines will play a role in shifting the design paradigm of artifacts. Today, engineering is highly specialized in many fields where ever faster, bigger, and stronger artifacts are developed. The division into fields, however, does not seem to answer the problems our society confronts, such as space, resource, and market limitations. Because of the potential breakthrough in these problems, investigation of micromachines is drawing attention in various fields. The micromachine is an interdisciplinary technology involving mechanical, electrical, and material engineering. Moreover, the anticipation of a breakthrough attracts new ideas and applications to this developing field, which accelerate fusion among specialized domains. This paper addresses interdisciplinary knowledge-sharing via the computer.

View full abstract

The present paper describes a new design of a fluid driven microactuator which comprises a planar piston, a four-way control valve, and a mechanical feedback system that controls the output in accordance with an input displacement. Fabrication processes of the microactuator include bulk micromachining of two sheets of silicon substrate and their assembly by electrostatic bonding to glass plates. The output characteristics were measured supplying the compressed air. Generated output force was partially linear to supplied pressure and to valve displacement. However, the magnitude was asymmetric in terms of driven directions, which results from geometrical configuration essentially introduced by an anisotropic etching of the silicon. The position of the output shaft was sufficiently controlled with a designed magnification factor multiplied to the input displacement in both static and dynamic senses.

View full abstract

We present a framework for building a knowledge-based system that can consistently and automatically utilize graphs, tables and formulae as a knowledge source in mechanical CAD. In this paper, we describe both design object and design process using constraint networks which consist of constraint assemblies, constraint objects and constraint modules. The design object is described by a hierarchical structure of the constraint networks, and the design process is represented by constraint propagation between them. We introduce the concept of functional constraint in the constraint network and show that the database search for design variables can be carried out automatically. The usefulness of this system is illustrated by the application to the design of a gear system.

View full abstract

In order to assist a designer from the viewpoints of qualities (e.g., reliability), the authors have proposed a concept of design aid based on the concurrent processing of design tasks. This paper describes a system of design aid based on the proposed concept. This system performs fault tree analysis concurrently with the designer's modeling process to assist the designer from the viewpoint of reliability throughout the modeling. In this system, the fault tree analysis is decomposed into subproblems which are concurrently solved as the modeling proceeds. With the concurrent processing, the designer is informed of the information (e.g., malfunction of design object) in terms of the reliability early in the modeling. An example is given to demonstrate the effectiveness of the system.

View full abstract

To realize concurrent design, it is very important to prioritize customers' requirements since they are quickly diversifying to a broad extent. Otherwise, engineers will find it very difficult ot set definite goals for their products and will not be able to cooperate effectively toward product development. In this study, the AHP (analytical hierachical process) technique is introduced to prioritize customers' requirements. A prototype was developed in Smalltalk-80, utilizing its object-oriented programming environment, and the effectiveness of the approach is demonstrated with simple illustrative examples of the selection of joining method.

View full abstract

With the increasing diversity of demands based on personal preferences for a variety of products, the conventional design procedures using series of consecutive operations are becoming inadequate to supply designs which satisfy personal preferences, and to produce what consumers are really demanding. As one solution, concurrent engineering proposes to make a design under simultaneous collaboration with what the design is going to produce to consider personal preferences. Our approach to this problem is to build a design support system with a graphical user interface (GUI) having a browsing function. Since a user observes a design object from a layered viewpoint, we propose a perspective model that supports the view. To implement the perspective model on GUI and enable browsing for a design, a new idea of Hypergraph is proposed. This paper describes the perspective model, the Hypergraph model and the prototype system HYDRA.

View full abstract

This paper describes a novel method of searching a multidimensional "design window" using a multilayer neural network. Here, "design window" is defined as an area of satisfactory solutions in a permissible space. The present method consists of three phases : (a) preparation of learning and unlearning data, which are produced through a number of finite element calculations with various design parameter sets, (b) training of the neural network and (c) searching of "design window" by using the trained neural network as an analyzer. To demonstrate the performance of the present method, it is utilized to obtain the design windows of the fusion first wall of ITER(International Thermonuclear Experimental Reactor).

View full abstract

A new curve generation method is proposed which permits a designer to draw a curve on a computer as if he or she is drawing on a piece of paper. This method is intended for the early stage of design and relieves a designer from the burden of generating a curve using control points. The conventional approaches require mathematical knowledge about control points to generate the desired curve, and what is worse, it is very difficult to modify it once it is drawn. Our new method is unique in that it pays attention to the trend of the data a designer generates and fits an exact curve by extracting feature points. Thus, we can modify the curve globally or locally as desired. Furthermore, it imposes very little burden on the computer.

View full abstract

The layout problem, in which a region should be partitioned into plural subregions of layout components so as to satisfy neighboring relationships between them and the size conditions for the whole region and respective subregions, occurs in several layout designs. It is difficult and computationally hard to find optimal solutions for them. In this paper, we propose an optimal space partitioning method based on rectangular duals of planer graphs, which is used for dealing with the topological structure of the arrangement of subregions. In the method, such a structure is optimized through a simulated annealing algorithm, and the layout structure represented with a rectangular dual graph is transformed into an embodiment layout by using the generalized reduced gradient method, one of the numerical optimization techniques for constrained nonlinear optimization problems. Finally, we show an example of an access control room layout in a power plant design for checking the effectiveness and validity of the proposed method.

View full abstract

In the design process of the products, various aspects should be dealt with concurrently and simultaneously, since they include a large number of functional disciplines. In this paper, we discuss the multi-disciplinary design problem of link mechanisms, especially from the viewpoint of multi-disciplinary modeling and optimization technique for linkage shape synthesis and structural strength design. Based on it, we show a formulation of such multi-disciplinary design problem, and develop an automated design optimization system by using object-oriented modeling technique, symbolic programming technique and algebraic programming language for automated generation of optimization model concerning the above several functional disciplines, and by using the generalized reduced gradient method as an optimization algorithm. Finally we apply it to the design of a hydraulic shovel in order to show its effectiveness.

View full abstract

The workload leveling technique in the scheduling problem is a non-integer planning problem and one kind of solutions to the problem of allotment of manpower resources. But we have to determine the sequencing of activities at the same time. As workload leveling technique does not simply fill the concave part with the convex part, the constraints are tangled in a very complex manner. Thus, it is not easy to solve the problem by the conventional approach. Further, it often happens these days that we have to change the project schedule because of the situation of the ongoing activities or due to the unexpected disturbances. In this paper, the workload leveling technique is formulated in the framework of a neural network approach using the interconnected network and its effectiveness was examined by simulation.

View full abstract

The importance of project scheduling problems is growing rapidly with the increasing diversity of products. The most important requirements are flexibility and simplicity. However, most of the methods proposed up to now have difficulty in coping with such problems involving varying workload activities. Therefore, a new technique based on figure manipulation is proposed to meet these requirements. Our new technique utilizes a shape model called CBR (Coded Boundary Representation) proposed by S. Fukuda, which expresses a shape in the form of lists using directional codes. Figures can be easily composed or discomposed and their features can be extracted without any difficulty. Scheduling is solved as a packing problem of the rectangles representing activities. Our approach greatly simplifies the overall procedures, since they can be reduced to simple list operations, and can cope with the varying work load problem with considerable ease.

View full abstract

Most of the studies on the vibrations of cracked shafts investigated the major critical speed and the secondary critical speed. However, it is known that other kinds of vibrations appear in cracked shafts. In this paper, resonance phenomena of the 3/2-order super-subharonic and 1/2-order subharmonic oscillations are investigated. They occur when the natural frequency is equal to 3/2 or 1/2 times the rotating speed. In the theoretical analysis, a two-degree-of-freedom model whose spring characteristic is represented approximately by a power series is adopted. The following are clarified. (a) These two kinds of oscillations have similar characteristics. (b) They occur only in horizontal shafts. (c) Only the steady-state oscillations appear. (d) The amplitude of the steady-state oscillation decreases as the unbalance increases. (e) These theoretical results are verified by experiments.

View full abstract

In this paper, a theoretical analysis is presented for the free vibration behavior of pretwisted thick cylindrical cantilever panels, using the Rayleigh-Ritz approach. The method is based on the three-dimensional elasticity theory with consideration of rotary inertia and shear deformation. Numerical results for pretwisted cylindrical cantilever panels having typical geometries are compared with those available in the literature. The effects of pretwisting and aspect ratio of panels and central angles of the cylinder on the results are also discussed.

View full abstract

As a practical example of a self-excited actuator theoretically presented in the previous paper, we developed an extended theory of the two-degree-of-freedom swing wing vibratory mechanism and showed its experimental prototype as a model of insect wings. In order to realize a self-excited system, we proposed not only a positive velocity feedback with nonlinear damping similar to the Van der Pole system, but also a more efficient Coulomb friction-type positive feedback system. Since the first-mode swing motion tends to be drawn into the second-mode limit cycle, we proposed two methods to excite only the first mode : one of them is the suppression of the second mode by means of a low-pass filter and the other is a cross feedback capable of exciting either mode.

View full abstract

As a practical example of a self-excited actuator theoretically presented in the 1st report, we developed a biped walking mechanism driven by a self-excited actuator. Referring to a biped walking toy which walks on a slope with a self-excited mechanism using potential energy, we proposed a mechanism which walks on flat ground with two single-link legs. Biped walking motion is realized by the swing roll motion of the upper body driven by a two-degree-of-freedom self-excited vibration system with asymmetric stiffness matrices. The fundamental characteristics of the self-excited actuating system are analyzed theoretically, and its efficiency and robustness are discussed by comparing the equivalent forced excitation system. The theoretical results are verified from the experimental study of the prototype. It is also found that the self-excited system has a better robustness to disturbance and variation of system parameters caused by walking motion, and can realize stabler walking motion, compared to the forced excitation system.

View full abstract

In the previous paper, the general theory of exact solutions in a family of Duffing oscillators including hard, soft and snap-through spring systems was studied. In particular the paper discussed the hard spring system in detail, and a listing of the numeration program was included. As a continued report, the soft spring system is dealt with here in detail. All the free vibrations in a family of Duffing oscillators are solved exactly and formally in terms of a family of Jacobian elliptic functions ; however, their precise numeration is a very important task. The present study is devoted to discussing excellent algorithms/programs. A FORTRAN program with the precision of REAL * 16 is presented, which ensures an accuracy with relative error of less than 10^-30. The listing of the program and numerical results of the dynamics of the Duffing oscillator with the soft spring system are demonstrated.

View full abstract

Nonlinear sloshing motion of the liquid in a partially filled circular cylindrical tank is investigated. The tank is subjected to horizontal, vertical and pitching excitations. The nonlinear ordinary differential equations governing the liquid surface oscillation are derived by applying Galerkin's method. It is confirmed that the vertical excitation causes the parametric excitation. In addition, it is noted that the pitching excitation also causes the parametric excitation when the pitching axis does not intersect the symmetrical axis of the circular cylindrical tank. The time histories of the liquid surface displacement are calculated to the harmonic pitching excitations. An experiment is carried out using a model tank. A fairly good agreement was found between the theoretical and experimental results. It is shown that, even if the tank is only subjected to the pitching excitation, the amplitude of liquid surface oscillation grows owing to parametric resonance.

View full abstract

This paper presents experimental results on chaotic vibrations of a post-buckled beam carrying a concentrated mass. The concentrated mass is attached at the point apart from the center of the beam. The beam is clamped at both ends to a base frame. The beam is deformed into a post-buckled configuration by an initial axial displacement. The displacement is controlled by the thermal elongation of the beam to the base frame. The chaotic vibrations of the post-buckled beam appear within a restricted range of exciting frequency. The chaotic responses are generated in the sub-harmonic resonance regions of both 1/2 and 1/3 orders. The maximum Lyapunov exponent is 1.4. The chaos attractor of the dynamic response is focused clearly onto the Poincare section. The correlation dimension saturates to d=3.6 within the embedded dimension up to 10 or more. Other chaotic responses appear at the regions of a superharmonic resonance of the second order and the ultra-subharmonic resonance of fourth to fifth order corresponding to the lowest mode of vibration.

View full abstract

Analytical results are presented on chaotic vibrations of a post-buckled beam carrying a concentrated mass. The mass is located at the point apart from the center of the beam. Both the beam and the mass are excited by periodic acceleration. Applying the Galerkin procedure to the basic equation, the nonlinear differential equation of a multiple-degree-of-freedom system is reduced. The time progress is calculated by the direct integration technique. The results point to the following conclusion. By adopting more than three fundamental modes of vibration, the chaotic attractor and the correlation dimensions agree consistently with those of the preceding experiment. The asymmetric deformation causes the more complicated projection to the chaotic attractor of the beam.

View full abstract

The vibration of a rotating elastic disk subjected to base excitation is theoretically analyzed. Formulation of the steady-state response to harmonic excitation is derived as an eigenfunction series where the eigenfunctions are given in the form of rotating waves. It is presented that a parallel motion to the axis excites only the axisymmetric modes. It is also shown that a tilting motion around an arbitrary diameter excites the forward and backward rotating modes whose circumferential wavenumbers are one. The Doppler's effect and the coupling effect between the inertia force and the Coriolis force are discussed. The response curves and the vibratory modes of the rotating disk are studied in detail. It is shown that the standing waves fixed in space can be formed. The relation between the excitation frequency and the configuration of the vibratory modes is also presented.

View full abstract

When normal active mass dampers using servomotors continue to work beyond the capacities of the servomotor systems, the servomotor drivers stop the motors to protect the motors and/or the drivers themselves against overheat due to overload. However, the hybrid mass damper developed can release the servomotors from the operation before their protection systems start to work, and can provide them with time to cool, while the mass damper continues to work as a passive mass damper ; then the servomotors can work again by mode switching from passive to active at the mass damper controllers' request. Shaking table tests were carried out for a large-scale experimental model of the hybrid mass damper with a 4 t moving mass supported by an XY-motion mechanism, and the driving mechanism in each direction consists of an AC servomotor of 5.5 kW capacity. The tests showed that the analytical models were obtained to express dynamic behavior of the mass damper in the active mode, the passive one and the mode switching. Simulations were carried out to predict performance against wind and earthquakes, where application of the model to a 7-story building was supposed. The results showed that the hybrid mass damper had good performance throughout a wide range of excitation intensity.

View full abstract

The purpose of this study is to develop an analytical method which can simulate vibration phenomena of a flexible rotor controlled by active suspensions. Feedback gains were defined by solving the matrix Riccati equation as the optimal regulator problem. We also used a finite-element method (FEM) in the formation of a rotating shaft in order to assume an arbitrary shape and higher vibration modes. We show the calculated results of vibration modes of a model rotor, whirling eigen-frequencies vs. spin speeds of the rotor, time history responses in the case that the rotor was controlled by the active suspensions, and frequency responses. The capability for development of a control device of a flexible rotor system was demonstrated.

View full abstract

This paper proposes a method to reduce sound noise radiated through walls using an active vibration control technique. An active dynamic absorber is used for controlling vibration modes of a steel plate structure related to sound noise. In order to control the plate structure with distributed parameter using the modern control theory, it is necessary to make a reduced-order model with lumped parameters. In this paper, a three-degree-of-freedom system constructed by a new method we proposed is introduced. It is demonstrated that a control system is able to control well the first and fifth vibration modes which occurs effective sound radiation when the model and LQ control theory are used.

View full abstract

We proposed a new vibration control mechanism and control method for cabin horizontal vibration of super-high-speed elevators. An AC servo motor and ball screw on the car frame drive the control force to the cabin where the horizontal vibration is to be suppressed. The vibration of the cabin, detected by an acceleration sensor, is integrated to determine the velocity and the position. The control force is determined based on the above position and velocity, and on the velocity of a motor encoder taking into account the car dynamics. The derived dynamics of the elevator car with a control mechanism and the effect of vibration control was verified by numerical simulations. The results indicate good performance of the vibration control system, which can suppress the 1st- and the 2nd-mode vibrations of the car. Control force and actuator stroke remained within the practical extent of designed values.

View full abstract

We have developed a super-high-speed elevator considered to be the fastest in the world. However, the vibration of elevator cars increases in proportion to the traveling speed ; It is therefor necessary to develop new vibration suppression methods for comfort in riding. First, we developed oil damper for roller guides which can change damping coefficient easily. Second, we developed the vibration control mechanism proposed in our former report. Lastly, the car vibration test equipment system of the real size was developed for verifying these suppression methods. The results of experiments indicate that the oil damper has a high damping effect. It can suppress car vibration effectively. Furthermore, the car vibration control mechanism could suppress vibration to half of the uncontrolled case. We also investigated problems which occurred when we applied this mechanism to super-high-speed elevators.

View full abstract

This paper presents the flutter suppression of a cantilevered pipe conveying fluid by means of the output moment of servomotor. The cantilevered pipe is connected directly to a servomotor at the supported end. For the present system with the servomotor, it is difficult to obtain a robust controller using the H_∞ control theory. Hence the evaluation method for a structured singulr value is used to obtain a robust controller. Numerical simulation and experimental results show that the flutter is effectively suppressed by the present method.

View full abstract

Components and piping systems in current nuclear power plants and chemical plants are designed to employ many supports to maintain safety and reliability against earthquakes. However, these supports are rigid and have a slight energy-dissipating effect. It is well known that applying high damping supports to the piping system is very effective for reducing the seismic response. In this study, we investigate the design method of the elastoplastic damper [energy absorber (EAB)] and the seismic design method for the piping system supported by the EAB. Our final goal is to develop the technology to apply the EAB to the piping system of the actual plant. In this paper, the component test results are presented. From the test results, it is confirmed that the EAB has a large energy-dissipating effect and is effective for increasing the piping system damping. The EAB design method using beam theory is also developed.

View full abstract

The scanning-type suction cups (SSC) developed at the Mechanical Engineering Laboratory is analyzed and some of its features are presented. First, the linear graph theory is extended to be applicable to the simulation of the pneumatic components. Next, the prototype SSC is simulated and the simulation results of the SSC are compared with the experimental ones. Finally, the characteristics of the SSC are simulated with changing parameters such as scanning speed, time for connection between the manifold and each suction cup, ratio of the tube resistance between the vacuum pump and the manifold to that between the manifold and each suction cup, and sealing ability of the material. The simulation results clarify the effects of parameter changes and show the optimal parameter values which improve the features of the SSC.

View full abstract

Elevator traveling cables are forced to vibrate due to various causes such as the vertical and horizontal movement of the car, the vibration of the building caused by earthquakes, strong winds, etc. This paper discusses the dynamic characteristics of the cables with respect to the above factors using simulation analysis. The model cable used in the simulation consists of multiple rigid links which are serially connected by rotational joints made of spiral spring. As a result, the following conclusions have been obtained. (1) The shape of the cable changes to a large extent due to the vertical movement of the car. The magnitude of the cable vibration is not affected by the car speed or the cable length, but is mainly dependent on the car acceleration. (2) With respect to the horizontal vibration caused by the vibration of the car and building, a characteristic period of 2.5 seconds was obtained, with the cable length of 5.4 meters, from the response simulation. On the contrary, the results of the eigenvalue analysis differ largely from the above simulation in both frequency and mode. Therefore eigenvalue analysis is not effective in the study of cable vibration.

View full abstract

When assembly tasks are performed by robot manipulators, it is necessary to identify interactions between grasped objects and external environments such as contact positions and conditions, which are DOFs at contact point and contact directions. This paper discusses the identification of unknown parameters such as those mentioned above, when grasped objects are exposed to external environments. The general formulation of contact conditions is presented. Applying this formulation, a scheme is presented for identification of contact conditions by using the active force sensing method.

View full abstract

The objective of this study is to describe aerodynamic effects on vehicle dynamics. This paper concentrates on the mathematical formulations and the experimental certification of the longitudinal single-degree-of-freedom motion. The lift and pitching moment coefficients of a scaled car model are measured with the deviation of the body attitude. Quasi-steady aerodynamic modelling reduces the formulation to a linearized system. The transient responses simulating bouncing motion under high-speed conditions are calculated and found to be in good agreement with the results obtained in the wind tunnel tests.

View full abstract

When a hydraulic converter is operated under the stall condition, the input power is fully absorbed into its converter oil. In such a case, the input power absorbed can be estimated from the revolution speed of the converter's input shaft or the heat quantity of the converter oil. Japanese railway companies are regulated to estimate periodically estimate the engine performance of hydraulic railcars. The authors proposed to apply the above mentioned principle to the estimation of the engine performance without dismounting the engine from the railcar. This paper describes about the process how the new method is useful to be of practical use.

View full abstract

Force-motor direct-operated servo valves are increasingly used to improve the thickness control accuracy of hydraulic screw-down systems in rolling mills. The authors have proposed to use expandable, compressible metal springs instead of conventional rubber springs to position the spool in a servo valve, and a unique method to damp it with eddy current. The experiment on commercially operated mills has proven that the metal-spring valve exhibits excellent system response compared to the rubber-spring valve, though the independent response of the former is lower than the latter. Furthermore, a simulation study of dynamic characteristics of servo valves to control the roll push-up cylinder under an intended load was conducted using mathematical models in quadratic form considering complex or viscous damping. The result generally agreed with that of experiment and clarified that damping characteristics of the valve and the spring constant of the spool-supporting spring have significant influence on frequency response of the roll push-up cylinder.

View full abstract

In cogeneration systems (CGS), the dynamic characteristics of the refrigerator strongly influence the effective utilization of waste heat for space cooling. In this study, a dynamic simulation model is developed for a CGS composed of a gas engine, waste heat refrigerator and other components using DYNAMO. The accuracy of this model is evaluated particularly focusing on the starting charateristics of the system by comparing measured data of an actual operating plant with data from the simulation and the validity of the model is confirmed.

View full abstract

Wavelet analysis consists of a versatile collection of tools for analysis and manipulation of signals such as sound and images, as well as more general digital data sets. In this filed, wavelet theory seems to be an efficient alternative to Fourier analysis that suffers from a lack of localization of the analyzing function. In particular, this new method enables engineers to process nonstationary signals in an appropriate way. In this study we apply wavelet analysis to monitoring of cutting conditions in milling. We monitor dynamic cutting torque, surface roughness and intermittent sound of cutting, and analyze intermittent sound of cutting by wavelet analysis and the pattern matching technique. Finally we demonstrate that our proposed technique provides a reliable detection of tool breakage in milling.

View full abstract

A new complex-valued formulation using an integral equation and a scheme for eigenvalue determination by boundary element method are presented for the Helmholtz equation. With help of similarity of the present formulation to the one by the Multiple Reciprocity boundary element method, unknown eigenvalues appear explicitly as polynomials in coefficient matrices, which facilitates recomputation of the determinant of the coefficient matrices. Eigenvalues are determined as local minima of the determinant mentioned above by the Newton iteration method. The advantage of the present method is discussed and demonstrated by simple example computations.

View full abstract

It is frequently important in engineering to analyze complicated problems with a large number of small holes on an interface boundary. A muffler used for reduction of exhaust noise is a typical example of these problems. The boundary element method could also be applied to such a problem, but this analysis requires a huge number of boundary elements to obtain accurate results. To circumvent this difficulty, the present paper proposes a new boundary element method which can be applied effectively to 3-D acoustic problems with small holes on the interface boundary, governed by the Helmholtz equation. The usefulness of the proposed method is demonstrated through numerical computation for two sample problems.

View full abstract

A study was conducted on the structural sensitivity analysis of the thermal off-track in hard disk drives. Indexes were proposed to measure the influence of the deviated thermal expansion coefficient on the off-track. These were the influence coefficient for each component, the influence coefficient for each connecting part and the regulation coefficient for the deviation of the material for components. The analysis procedure was described, including a modeling step for computational simulation and an evaluation step using the coefficients. The method was applied to a head disk assembly with linear head positioner, and the calculated results were demonstrated in the case study.

View full abstract

Circular S-polarized light is proposed to reduce error in an optical surface roughness measurement system. Theoretical analysis of the focused circular S-polarized light shows that as the circular aperture ratio increases, 1) the spot size decreases, 2) the focal depth increases, and 3) the first-order defracted light increases. The measurement accuracy using the circular S-polarized light is experimentally evaluated. Circular S-polarized light is separated from randomly polarized light using a calcite crystal with birefringence. A nanometer-order step surface is then fabricated by sputtering carbon graphite on a glass plate. The step heights are measured mechanically with a stylus, and are used as standard measurements. The same step heights are also measured optically using randomly polarized light and circular S-polarized light. Comparison of the latter two measurements with the standard measurements shows that measurement error is reduced more with circular S-polarized light than with randomly polarized light.

View full abstract

In this paper, a new method has been devised for obtaining an arbitral pressure and driving the artificial muscle by controlling the electric field applied to the electro-rheological fluid. The characteristic of this method lies in that the necessary pressure is generated at the intermediate pressure tap owing to the shear stress, which is induced by using the double tubes as electrodes between which the ERF flows and by applying the electric field to them. The applications of the electric field of pulse-type (digital) variation as well as the linear (analog) one showed that both the methods generate the necessary pressure. It was experimentally confirmed that the artificial muscle, if it is being mounted at the middle port, can be driven smoothly even under the alternate voltage application. Further-more, was examined about the possibility whether or not the fingers were driven independently to each other even in the case that the fewer electrodes were used, and this method was proved to be the practical one.

View full abstract

In this paper, we proposed an adaptive stabilizing state feedback control method for MIMO (Multi-Input and Multi-Output) systems and its application in an inverted pendulum system. The adaptive control algorithm presented here is based on SAC (Simple Adaptive Control) basically and has the special features of simple adaptive controller structure and robustness in terms of disturbances and parasitics. Furthermore, the stability of the overall adaptive control system is proven. The adaptive stabilizing output feedback control method is also proposed in this paper. The effectiveness of these proposed adaptive stabilizing control methods was confirmed by experiments on an inverted pendulum system.

View full abstract

Power match control of hydraulic power systems means to control continuously the output power of the hydraulic pump so as to agree with the power of the load actuator. As a fundamental study for attaining high performance in power match control systems, this study deals with a pressure match control system. One of the serious problems in conventional pressure match control systems is that the systems are usually very oscillative and may sometimes even become unstable. In this study, a nonoscillative and very stabilized system is proposed. It is verified by experiment and digital simulation that the system has good control performance.

View full abstract

The proposed control system is based on the super-real-time simulation in which a physical model is introduced to present the controlled system dynamics. The optimal actuating signals have been obtained by solving the physical model equation so that control variables are kept constant at every specified time interval with the aid of parallel microprocessors. It is shown by computer simulation that the method has been successfully applied to level control of a cascaded tank system. The following results have been obtained by the calculation of the SRTC transfer function. (1) Feedforward compensation by the target control variable is added to the actuating signal. (2) SRTC yields proportional feedback control. (3) SRTC compensates the lag of the plant transfer function. (4) SRTC also compensates disturbance to the plant.

View full abstract

Production of a soft tone on the piano is difficult even for pianists. To determine the input waveform for the stable swing stroke which produces the soft tone, this paper has studied the movement of the piano's hammer until it strikes a string. A simple dynamic model of the piano-action is proposed, and the movements in the model are analyzed. In addition, movement of the hammer when the key was struck by a developed actuator is measured. On the basis of our analysis, it can be said that the calculated behavior shows good agreement with the measured one. This indicates that the proposed model is useful, especially since can clearly show the behavior of the hammer. The desirable input waveform for the swing stroke leading to the soft tone is determined, and the stable swing stroke can then be achieved.

View full abstract

In this paper, a learning control algorithm using neural networks (NN's) for controlling the motion and force of a direct-drive (DD) robot with a variable table, which has six degrees of freedom, is proposed. The motion of an end effector attached on the robot arm and the internal force between the end effector and the table through an object can be obtained by the DD robot arm in cooperation with the variable table. The control of the motion and the internal force is achieved by hybrid controllers and a feedforward controller using NN's. After the completion of learning, the NN acquires a model of inverse dynamics for the DD robot arm and the DD table. Thus the outputs of the hybrid controllers are nearly equal to zero. The effectiveness of the proposed control algorithm is demonstrated by an experiment on the control of both the position of an end effector and the contact force on the constraint surface of the object.

View full abstract

An investigation was carried out to elucidate the vibration of a tapered roller bearing. In this report, the vibrations of three tapered roller bearings were studied. The tapered roller bearings were operated at a constant speed under an axial load. The radial and axial vibrations on the outer ring of each tapered roller bearing were monitored by an accelerometer. The vibrations were examined using a fast Fourier transform analyzer. From the experimental results, analyses and considerations, the frequency characteristics of vibrations on the outer ring of tapered roller bearings were clarified, and the causes of main peaks were specified.

View full abstract

Estimation of friction and wear properties was attempted using the AE count rate in wear tests for three kinds of metal pins rubbed on four kinds of ceramic disks under unlubricated and lubricated (oleic acid) conditions. The relationships between the AE count rates emanating from the end of the pins and the specific wear rate, the topographies of micro-roughness of the wear surfaces, and the mode of wear were established. As a result, we found that the sum of the wear rates for the combination of pin and disk increased with an increase in the AE count rate under both unlubricated and lubricated conditions ; thus it is possible to evaluate the specific wear rate from the AE count rate. The specific wear rate and the AE count rate also increased with the increase in the surface damage such as the equivalent cross-sectional area of microgrooves of the pin and disk.

View full abstract