This paper combines a verified interval optimization method with the FEM for designing structures, which is denominated as the Hybrid Interval Genetic Algorithm (HIGA). This algorithm can neglect formulated equations and interval analysis, and while determining the optimum interval parameters. Furthermore, it can also maximize the design scope. In this paper, this algorithm is implemented for both a truss and frame structure. The interval optimizations include the static and dynamic responses of these structures. The results show that the algorithm which combines the IGA with the FEM can determine the feasible interval design parameters of structures with allowable objective errors.

View full abstract

Bearingless motors combine contactless levitation and rotation in a preferably compact system design, because bearing as well as motor windings are located on the same lamination stack. The bearingless slice motor features comparatively low complexity for a fully magnetically levitated drive system, because it allows the passive stabilization of three degrees of freedom by reluctance forces. By the use of a proper control scheme and the superposition of different current components, bearing forces and motor torque can be generated simultaneously by applying concentrated windings. This leads to a further simplification of the mechanical configuration. The bearingless segment motor features such concentrated coils on separated stator elements, which reduce the stator iron and therefore weight and cost, especially for constructions with large diameter. However, so far all bearingless slice motors are designed with surface mounted permanent magnets on the rotor, neglecting the advantages of buried permanent magnets. In this paper a novel bearingless segment motor featuring a rotor with buried permanent magnets is investigated. The motor specific mathematical model of force and torque generation is presented, a proper control scheme is introduced and the optimization of the prototype motor is outlined. Motor specific considerations concerning the angular sensors are given. Finally, the performance of the bearingless segment motor with buried permanent magnets is shown by the comparison of simulation results with measurement data of the manufactured prototype.

View full abstract

This paper presents a method for controlling the velocity of a hydraulic cylinder using a predictive multirate output feedback sliding mode controller. The nominal plant model is second order with time-delayed input and disturbance. Only velocity is measured, so the scheme uses fast-sampled output to estimate the states. A discrete predictor is used to treat the time delay, and the controller is designed for the future predicted states. Simulation and experiment results verify the proposed controller's performance despite the nonlinearity and uncertainty inherent in the hydraulic actuator system.

View full abstract

One of the authors had previously proposed an adaptive nonstationary control (ANSC) method and described its applications for the positioning control of vibration systems. In that control method, the time-varying feedback gains were obtained by online computations of a time-varying Riccati equation taking the parameter variations of the controlled object into account. However, the previous paper discussed the state feedback control method. The improvement of this method to enable output feedback control is one of important issues for actual applications. Generally, the output feedback controller of the optimal control includes a Kalman filter as the observer. However, the use of the Kalman filter for the ANSC method implies a heavy computational load in actual implementations because the Kalman filter is designed by another time-varying Riccati equation. Therefore, this paper proposes a practical output feedback control method for ANSC that does not include Kalman filter. The proposed method corresponds to a two-degree-of-freedom control system; the system consists of an online simulator and a sliding mode controller. This structure enables us to reduce the computational load. The effectiveness of the method is verified theoretically as well as experimentally.

View full abstract

This paper deals with active mode localization control of a periodic structure comprising a series of coupled identical components. Once mode localization occurs, vibration energy normally distributed throughout a whole system is intensively concentrated in a particular component of a periodic system, and hence needs to be eschewed because of localized stress. This paper begins by elucidating the causes of the mode localization phenomenon which is likely to happen in a periodic structure under some conditions. Numerical analysis is then performed, investigating the mode localization phenomenon from a viewpoint of dynamical compliances as well as mode behavior of a total system comprising coupled flexible beams with springs. It is shown that energy confinement in terms of disturbance forces takes place when mode localization takes place. In consideration of the energy confinement, active mode localization control is presented, enabling the vibration isolation from the disturbance energy. Finally, an experiment is carried out, demonstrating the validity of the proposed method for the purpose of actively isolating power flow form one subsystem to another.

View full abstract

Algorithms of motion control for a 9-DOF finger-arm robot were proposed by using the finger's manipulability in a previous study. However, only the methods for completing an unconstrained motion of the finger-arm robot were discussed in the previous study. In this paper, the authors propose a novel method for a finger-arm robot to complete an impedance control by regulating finger's manipulability in a constrained task. Methods of combining the previous heuristic method to the passive impedance control and the active impedance control are firstly developed. Then, an impedance control combined with the steepest ascent method is also proposed. The features of the proposed methods are demonstrated through several experiments. The proposed impedance control method combing with the steepest ascent method shows a strong performance even when a dynamic external force was applied to the finger. By using the proposed method, the arm actively moves along a direction to effectively maintain the moving potential of the finger. As a result, the finger shows a robust performance to its singularity in both unconstrained and constrained tasks.

View full abstract

Many kinds of actuator-based (active type) haptic device have developed and utilized as rehabilitation robots. These systems have great advantages for rehabilitative activities, for example assistive forces and so on. However, from the view point of safety, we have room to consider utilizing brake-based (passive type) haptic devices as rehabilitation-tools. The effects and roles of active / passive force feedback for rehabilitative trainings have not been clarified yet. In this study, we have developed an active / passive switchable rehabilitation system for upper limbs (Hybrid-PLEMO) to address these questions. In this paper, we describe the force-feedback mechanism of the Hybrid-PLEMO.

View full abstract

A power assist system has lately attracted considerable attention to lifting-up an object without low back pain. We have been developing power assist systems with pneumatic actuators for the elbow and shoulder to farming support of lifting-up a bag of rice weighing 30kg. This paper describes the mechanism and control method of this power assist system. The pneumatic rotary actuator supports shoulder motion, and the air cylinder supports elbow motion. In this control method, the surface electromyogram(EMG) signals are used as input information of the controller. The joint support torques of human are calculated based on the antigravity term of necessary joint torques, which are estimated on the dynamics of a human approximated link model. The experimental results show the effectiveness of the proposed mechanism and control method of the power assist system.

View full abstract

The paper presents a new algorithm for system identification in the time domain using impulse response. The algorithm is aimed at the identification of single-input multipleoutput (SIMO) systems with non-zero initial states; that is, where the initial state vector contains one or more non-zero elements. The Eigensystem Realization Algorithm (ERA) realizes an SIMO system with a zero initial state vector. We modify this method and first estimate the system matrix and initial states using the system's free decay response. The system matrix is then obtained by singular value decomposition in the same manner as ERA, and the initial state values are refined by recursive least squares iteration. Next, deducing the effect of the initial states from measured impulse response data, we reconstruct the ideal impulse response, then identify the input and output matrices using another recursive least squares iteration in the time domain. Numerical simulations are performed to verify the new algorithm, and the results of example cases show the validity of the new identification scheme. Finally, we apply the new algorithm to the actual on-orbit data of the Engineering Test Satellite (ETS-VIII) in orbital control mode (OCM) for its system identification.

View full abstract

This paper describes the modeling of a semi-active shock absorber that uses magnetorheological (MR) fluid, which is a material that responds to an applied magnetic field and changes its apparent viscosity. We propose a dynamic system in which the friction of a falling body is taken into consideration, and a model in which the shearing resisting force depends on the velocity of flow. The validity of this model and a Bingham model is verified by comparison of the experiment with a simulation. This study clarifies that resisting forces obtained by the simulation corresponded to those obtained by the experiment, and that the current model is effective in predicting the behavior of a MR shock absorber.

View full abstract

This paper deals with a countermeasure against pattern formation phenomena generated on thin sheet winder using dynamic absorbers. The thin sheet is widely used as paper, aluminum, etc. Pattern formation phenomena of the wound sheet sometimes occur in operation because of the viscoelastic characteristics of the sheet. The pattern formation phenomena lead to a fatal damage on the thin sheet. This problem is similar to the other pattern formation phenomena, for instance, polygonal deformations of the smoother roll of paper making machines and pattern formation of the elastic yarn of textile winding machines. The authors conducted experiment to generate the pattern formation using experimental apparatus, and theoretically analyzed to clarify the mechanism of the pattern formation phenomena considering the elastic deformation of the rolls. They also theoretically examined the countermeasure to suppress the phenomena by attaching some dynamic absorbers. The effectiveness of the tunable dynamic absorbers was confirmed by the theoretical analysis, and an example of the tunable dynamic absorber was introduced.

View full abstract

A pressure wave propagating in a tube often changes to a shock wave because of the nonlinear effect of fluid. Analyzing this phenomenon by the finite difference method requires high computational cost. To lessen the computational cost, a concentrated mass model is proposed. This model consists of masses, connecting nonlinear springs, connecting dampers, and base support dampers. The characteristic of a connecting nonlinear spring is derived from the adiabatic change of fluid, and the equivalent mass and equivalent damping coefficient of the base support damper are derived from the equation of motion of fluid in a cylindrical tube. Pressure waves generated in a hydraulic oil tube, a sound tube and a plane-wave tube are analyzed numerically by the proposed model to confirm the validity of the model. All numerical computational results agree very well with the experimental results carried out by Okamura, Saenger and Kamakura. Especially, the numerical analysis reproduces the phenomena that a pressure wave with large amplitude propagating in a sound tube or in a plane tube changes to a shock wave. Therefore, it is concluded that the proposed model is valid for the numerical analysis of nonlinear pressure wave problem.

View full abstract

Dynamic performance of multiwound foil bearings with the effects of foil local deflection is investigated. The foils, separated and supported by projections on the ir surface are treated as thin plates. Deflections of the foils are solved with a finite element model. The air pressure is calculated with the Reynolds' equation by treating the lubricant as an isothermal idea gas. The effects of foils are simulated with the deflection of top foil added to the film thickness. A finite difference computer program is developed to solve the Reynolds equation and the elastic deflection equation, simultaneously. Perturbation method is used to determine the dynamic coefficients. The effects of foil deflection is discussed by comparing the dynamic coefficients of a foil bearing and a rigid bearing. Experimental data from a test rig supported by two multiwound foil bearings are used to validate this numerical solution.

View full abstract

The tube bank is used in boilers, heat exchangers in power plants and steam generators in nuclear plants. These tubes sometimes vibrate violently and come to the fatigue failure due to the flow induced vibration which is caused by the cross flow. This phenomenon is that the large vibrations arise at the critical flow velocity and it is called fluid elastic instability. However the relation between the onset velocity of fluid elastic instability and the tube array's geometry has not been clarified sufficiently. There is a few reference related to the relation between the pitch to diameter ratio and the onset velocity even in the lattice arrays. In this paper, the influence of tube arrays on fluid elastic instability is examined by experiments. As a result, it is clarified that the tube vibrations become large as T/D increases and L/D decreases, and the tube vibrations strongly depend on the dynamic characteristics of tubes such as the natural frequency and the damping ability.

View full abstract