This paper presents a new information processing model for engineering using a vibrating potential field. In the field of engineering, the conventional information processing models are based on central control processing and parallel distributed processing. Such conventional models are programmed by a sequence of procedures through internal connections between specified processing units, and for this reason it is difficult for all the units to express unity of purpose, enlarge of the system and observe the status of internal connections. However, information exchange between all depends on using field information and information processing of living units is a self-organizing process that can be achieved using field information. Many examples are found in nature where high-performance information processing can be attained without supervision. An analogous model of this self-organized information processing can give a useful control mechanism for PDP. This paper proposes a new information processing model using the vibrating potential field and discusses the results of computer simulations using this model.

View full abstract

Swing of a ropeway carrier is easily caused by wind and it is very difficult to reduce. In this paper, several types of dynamic vibration absorbers for a ropeway carrier are analyzed theoretically, and a general theory is formulated. A method of optimum tuning is obtained and the index (equivalent mass ratio) which indicates the effect of the absorber is defined. When the dynamic vibration absorber is located at the center of oscillation, swing is not reduced at all. The center of oscillation is the point that is located at the distance of equivalent length of the pendulum from the center of suspension (fulcrum). The absorber should be located as far as possible from the center of oscillation.

View full abstract

Linear programming is used to suppress the vibration of a flexible structure in motion. When the structure is well modeled, this method is effective, but errors are easily introduced to the models ; therefore robustness becomes an important problem. Moreover, when the vibration under motion is controlled by a feedback system, the effect is not as good as expected, because of a restriction of state feedback gain due to the instability problem. In this paper, the feedback gain is obtained through optical-regulation theory on the premise that the feedback system is used with linear programming. Using this feedback system, an vibration control optimum trajectory method of a flexible structure based on linear programming is proposed. This method is more robust than that using only the optimal trajectory derived from linear programming. The validity of this method is shown through simulation and experiment.

View full abstract

In this paper, we first describe the characteristics of engine damping that are necessary for the calculation of torsional vibration. Secondly, we describe an experiment in which two crankshaft pulleys with a torsional rubber damper are fitted to a 6-cylinder, high-speed diesel engine. The torsional waveforms of the damper inertia ring and the pulley are measured by means of phase-shift torsiograph equipment. The measured waveforms are harmonically analyzed and the dynamic characteristics of the stiffness and the damping are investigated from an experimental viewpoint. In addition, torsional waveforms are calculated by means of a method of simulating torsional vibration, in which a transition matrix method is adopted. As a result of comparisons with experimental data, certain dynamic characteristics of pulleys with a torsional rubber damper have been clarified.

View full abstract

We have already proposed a method for analyzing the nonlinear vibration response properties of rubber in relation to frequency and amplitude of displacement (Ref. (3)). That method determined the predominant frequency of rubber itself by using the FFT (Fast Fourier Transform). In this paper, we describe a new method for analyzing nonlinear vibration response. This method determines the predominant frequency of relative displacement of rubber itself by using a period on each one of the hysteresis loops. We were able to verify the appropriateness and accuracy of this method by directly comparing the analytical results with the experimental results for the dynamic response of a system. This new method compares favorably with the method of Ref. (3) in terms of the accuracy of calculation. We consider that our method may be applicable to a multiple-degree-of-freedom system.

View full abstract

A theoretical formulation of a new fast learning method based on back propagation is presented in this paper. In contrast to the existing back propagation algorithm which is based solely on the modification of connecting weights in between units (i.e., neurons) of different layers of the neural network, the present method calculates the optimum slope of the sigmoid function for each unit together with the variation of the connecting weights. The effectiveness and versatility of the present method is verified by the system identification of (a) linear and (b) nonlinear (Duffing and fluid-type) single degree of freedom mass-spring dynamic models. In all of the three cases, the present method excels in speed and accuracy compared to that of the existing method using a fixed slope sigmoid function.

View full abstract

A prototype failure diagnosis system has been developed using neural network technology for the actuators of air-operated valves. Because actual failure data were not easily available, the data of 30 failure patterns were experimentally obtained using more than 10 sensors. The time series data of sensor signals are Fourier transformed. The data of magnitude spectrum, phase difference and other quantities are used as the characteristic parameters in our failure diagnosis. From the data, appropriate information for use in failure diagnosis was extracted. Furthermore, similarities among failure characteristics were found by fuzzy clustering and statistical analysis. The new system that we developed consists of many sub-networks and one main network. Each sub-network is related to one specific sensor signal, and deals with the magnitude spectra from the sensor signal. The main network makes the final decision according to the outputs from the sub-networks and other data. In our system, the number of network connections can be reduced by approximately 40% without degradation of the recognition capability in comparison with the conventional system that uses only one neural network.

View full abstract

In this paper, an attempt to control a real system by changing LQ control laws corresponding to its operation regions is reported. The system to be considered has nonlinear characteristics and is described by an inaccurate mathematical model. To develop the control scheme, the method of multi-sectional control is presented and discussions are given on the robust stability of the original real system, which is described by a first-order differential equation. The proposed method has been applied to the control of a simple hovercraft model which has nonlinear characteristics ; then the effectiveness of the method is demonstrated through several elevation control experiments.

View full abstract

In this paper, we discuss minimum energy control in the z-domain for a single-input linear time-invariant discrete-time system. The dynamic controller with constant parameters is induced from the closed-loop expression or the open-loop expression with time-varying parameters in the t-domain, by using the property that the state settles to zero within a finite time. Since the constant parameters do not depend on the initial state of the system, it becomes more useful to analyze, synthesize and implement the feedback system. Furthermore, the stability and the finite-time settling property of the feedback system are guaranteed.

View full abstract

A new approach is proposed for the design of digital model reference control systems which overcomes the problem of nonminimum phase sampling zeros. The design is based on the plant-input mapping method which was proposed recently as a digital redesign method and which guarantees the internal stability of the system for almost any sampling frequency. This method is applied, in the present work, to the direct digital design of a pole placement control system and, in particular, a model reference control system. The key to eliminating the effect of sampling zeros on the performance of a model reference control system is the use of a matched-pole-zero model of the plant in combination with the invariant model, rather than the truncated step invariant model which is used conventionally. Simulation results are presented to show that the proposed design provides superior results to the conventional model reference controller design.

View full abstract

In this paper, the design of the robust dynamic output feedback sliding-mode controller (SMC) for singular perturbation systems is investigated. All the gain matrices of the dynamic sliding-mode controller can be determined via the static output feedback control for the alternative model of the reduced sliding dynamics. The results obtained differ from those of Refs. (6), (7) and (13) with respect to dynamic output feedback and also from those of Ref. (31) with respect to sliding-mode control. An active suspension system for a quarter-car is exploited to illustrate the derived schemes.

View full abstract

There is a strong demand in practical fields to obtain better control performance by means of a more effective utilization of future information. One useful method for improving the dynamic performance of a control system is to utilize the future desired signal if possible. Recently, a synthesis method for unified predictive control, called UPC, has been developed by Soeterboek and coworkers. It can be said that by this method the unification of several well-known predictive control methods has been accomplished. The purpose of this study is to find out whether the UPC controller is applicable to an electro-hydraulic servo-system. A hydraulic system with the UPC controller was designed and the performance of the system was investigated by experiments and computer simulation. It was verified that the performance of the system was improved considerably by employing this synthesis method.

View full abstract

This paper discusses an algorithm for computing fingertip positions of a multifingered hand in three dimensions. The problem of computing fingertip positions which satisfy equilibrium equations becomes a nonlinear programming problem. The nonlinear condition of the problem is bilinear. Hence, the problem is reduced to a linear programming problem by using the optimum solution obtained from the relaxed problem of the nonlinear programming problem.

View full abstract

This paper describes a monolithically constructed, flexure-hinged translation mechanism suitable for mechanical scanning stages of very high precision. Two piezoelectric stacks are employed to drive the mechanism ; one piezoelectric stack pushes whilst the other pulls the moving platform. This push-pull drive system is effective for raising the natural frequency of the stage. The mechanism described in this paper functions as both a guide for rectilinear motion and a mechanical amplifier to amplify a limited output displacement of the piezoelectric stack. Static and dynamic characteristics of the mechanism are considered in detail, and design equations are presented. The validity of the design equations is confirmed by experiment. A brief description of the accuracy of motion of the moving platform is included. It is demonstrated that a scan range of 60 μm is achievable with a natural frequency as high as 1175 Hz in the scanning direction.

View full abstract

When all input links except one are fixed in the mechanism of an in-parallel actuated manipulator with N degrees of freedom (dof), N kinds of mechanisms with a single dof can be obtained, and pressure angles corresponding to each mechanism can be defined. In order to transmit power efficiently from input links to the output link, all of the cosines of the pressure angles must be as large as possible. The present paper proposes a new transmission index (TI) for in-parallel actuated manipulators based on the power transmitted from input links to the output link, which is defined as the minimum value of cosines of the pressure angles. For planar and spatial in-parallel actuated manipulators, the effectiveness of TI is investigated considering static characteristics of mechanisms. Working spaces of in-parallel actuated manipulators obtained with consideration of motion transmissibility are also shown.

View full abstract

This paper is concerned with the design of a slide mode control system from the view point of practical application. The problem is treated using the asymptotic reaching law method for assuring global stabilization, and the chattering phenomenon is alleviated by the saturated function of the boundary layer. The design algorithm is simple and applicable to a nonlinear system including disturbances and uncertainties with high robustness. Its effectiveness is proven by simulation as well as a bench experiment of a brushless motor vehicle.

View full abstract

In this paper, the static characteristics of infinitely long slider bearings lubricated with non-Newtonian fluids are examined theoretically by considering the fluid inertia effects. In the derivation of the modified Reynolds equation, the fluid inertia terms in the momentum equation for the lubricant film are averaged over the film thickness, and the Rabinowitsch empirical model is used as a constitutive equation of non-Newtonian fluids. Applying the modified Reynolds equation to the infinitely long slider bearings, the film pressure, load carrying capacity, friction force and inlet flow rate are obtained under various values of the dimensionless nonlinear factor and film thickness ratio, and the effects of fluid inertia on these static characteristics are discussed.

View full abstract

A continuously variable power transmission, CVT, has the potential to provide comfortable cruising, powerful acceleration, fuel economy and improvement of exhaust emission as an automotive propulsion system. The authors have been engaged in research and development of a half-toroidal traction drive CVT since 1980. In practical use, reliability and compactness are important issues, which motivate ceaseless improvement of traction drive materials, synthetic traction fluids and transmission efficiency. This paper focuses on the fatigue of traction rolling elements, endurance of high-temperature traction fluids and efficiencies of single- and dual-cavity CVTs for 1.5 L and 3 L passenger cars.

View full abstract

Four-wheel drive offers cruising stability and high road gripping force at the expence of the power weight ratio due to implementation of an additional mechanism of a center differential gear box with a torque split clutch. We developed a dual-cavity half-toroidal traction-drive continuously variable power transmission (CVT) for 3 L passenger cars in 1990, which can be easily remodelled into a CVT with two output shafts since it has two independent CVT units. Since the CVT has a hydraulic-control equal-force-transmitting mechanism for speed ratio control, it is easy to synchronize the force transmission of front and rear axles without using a center differential gear box. In this paper we describe the mechanism, efficiency and control principle for the equivalent function of a center differential gear box of the four-wheel-drive CVT with nominal input torque capacity of 250 N·m, rotational speed of 5500 rpm and maximum speed ratio range of 1 : 8.7.

View full abstract

In order to clarify the behavior of oil film between a piston ring and cylinder liner, pressure measurement and visualization of oil film are carried out simultaneously using a simulation rig. In this rig, the piston ring is simulated by a straight glass bar test piece and the cylinder by a reciprocating flat plate. The following observations were obtained. The oil film pressure of the full film region before oil film rupture is negative and the maximum magnitude of negative pressure, i.e., tension, reaches to 0. 35 MPa. Cavities are formed at 20°-30° crank angle from dead center in the full film region, and grow with the increase of the crank angle. Then a new series of cavities is formed at the rear edge of the test piece. The two series of cavities then coalesce and the flow becomes perturbed momentarily. Negative pressure is detected at a 20°-30° crank angle after dead center and is not released until this coalescence of cavities has occurred.

View full abstract

We present a kinematic formulation, referred to as the joint displacement method. In the kinematic analysis, the closed-loop kinematic chain of interest must be transformed into several open-loop chains by imagining that some joints are relaxed (disconnected), and by replacing them as physical entities with a set of constraint equations. This formulation requires only a minimal number of generalized coordinates, hence it provides a highly efficient approach for real-time kinematic analysis of spatial mechanisms, heretofore unseen in the literature. For example, the Jacobian matrix for the RSCR mechanism is 3×3 and the CPU time required to perform Gauss elimination is 182 times shorter than that in the traditional link-coordinate method which requires a 17×17 Jacobian matrix. This method is applicable to both single loop and multiloop spatial mechanisms with revolute, cylindrical, translational, spherical and universal joints. A general-purpose computer program that implements the joint displacement method has been developed and tested on a variety of mechanisms.

View full abstract

In magnetic abrasive machining, the internal finishing of a nonferromagnetic tube is performed by means of the magnetic force generated between magnetic abrasive and the N-S magnetic poles. In this experiment, "mixed-type magnetic abrasive" are applied to finish the internal surface of a tube, in which large iron particles, generating a high finishing pressure, are mixed with magnetic abrasive resulting in finishing. By this new process using the mixed-type magnetic abrasive, the finishing efficiency is increased markedly. In this paper we describe the effects of both the size and the mixed weight percentage of iron particles on the surface roughness of a stainless steel tube. We also describe the application of the new process to the internal finishing of the bottom of a clean gas bomb for practical use. As a result, the surface roughness before finishing, 7 μm R_max, was reduced to 0.2 μm R_max over the entire inner surface of the bottom.

View full abstract

A system for verifying assemblability is required in CAD/CAM systems for mechanical products, by which the shapes of parts can be modified to enable easy machining and assembly on the factory floor. This paper introduces a part assemblability verification system. Substituting the assemblability with disassemblability, a system that generates paths for disassembly is constructed based on a solid modeling system for mechanical products. The developed system takes contact constraint and contact transition into consideration. Algorithms to derive possible motions of parts which maintain contact states, and to calculate the positions and orientations (configurations) where contact states change are presented. An assemblability verifying method is proposed that uses tree structural data whose arcs and nodes denote possible motions, configurations respectively. The efficiency of the developed system for design verification and assembly planning is demonstrated through an example.

View full abstract

The photoforming process is one of the optical forming methods used to produce plastic models. It has unique features such as the ability to form various objects, with few restrictions on shape, and to process them quite simply. In this paper, computer simulation considering photon behavior in liquid resin clarified that fine control of the exposure dose, that is, the light energy applied to a unit area of the surface, and soft focusing are the keys to achieve high precision. Some results of experimental forming demonstrate the possibility of producing fine structures with a forming resolution within 10μm. Finally, the results of this study suggest that this process can be applied to the construction of three-dimensional fine structures and precise mechanisms.

View full abstract

For the operation of automated guided vehicles (AGVs) in factories, the rescheduling of AGV routing is necessary to recover operational delay in AGV routing plans and bridge the gap between the schedule and the practical operation. This study deals with such a rescheduling problem. First, the problem is formulated as a problem of a network with time constraints which are referred to as time windows. Secondly, an algorithm is proposed to obtain two kinds of scheduling information, the earliest arrival time to each node and the latest departure time from each node, to assure the earliest possible arrival of the AGV at a goal node. This algorithm is a dynamic programming method which uses recursion. Lastly, a rescheduling procedure using the information is proposed to recover the practical delay in route scheduling.

View full abstract

Surface roughening of metals with plastic deformation is governed not only by the strain but also by the crystal grain size of the materials. Therefore surface roughening may be expected to decrease when a material is produced with a smaller surface grain size. In the present investigation, a thin layer with small crystal grain size is coated on the surface of an aluminum specimen with coarse grains and the effects of the crystal grain size, as well as the thickness of the surface layer, on surface roughening are evaluated in an upsetting test. It is shown that surface roughening is dominated only by the crystal grain size at the surface layer, and thus surface roughening can be suppressed by coating the surface of the material with a thin layer of small crystal grain size.

View full abstract