Restriction type flow meters are generally used for flow rate measurement of various fluids. Especially, a restricting mechanism which is formed with an orifice plate inserted in the middle of pipe is widely used. Orifice meters used for gas measurement have the following problem. When a difference arises between the temperature of gas and the design temperature of orifice, an error occurs in the measured flow rate value under the effect of the temperature difference. For this reason, it is stipulated that a thermometer for temperature compensation be inserted on the upstream or downstream side of an orifice where it exerts no effect on differential pressure measurement. However, it will be difficult to measure the actual flow rate in spite of employing the temperature and pressure compensation because of temperature difference between those two locations where the gas temperature measured at the thermometer location and that measured at the orifice location as well as those conditions of the piping system and heat radiation from the pipe well. Owing to recent technological progress, it has become possible to manufacture vary small sheath type thermometers that hardly disturb the fluid flow in a pipe. Such a thermometer can be installed near an orifice. Paying attention to this trend, we carried out various tests using currently available orifice meters to study the best temperature measurement location and the effect of a thermometer on differential pressure measurement. The results are reported here.

View full abstract

This paper concerns with active vibration control of non-uniform damped mass-spring systems by the wave control. Especially, this study considers clarifying a class of a damped mass-spring system that can be analyzed by the wave-based analysis. Because the wave properties are determined by the propagation constants, three conditions for the propagation constants are considered. Necessary and sufficient condition for the physical parameters to hold the three conditions is obtained. Moreover, for this class of damped mass-spring systems, properties of the propagation constants and the characteristic impedances, which achieve the impedance matching, are studied. Numerical examples are shown to prove efficiency of the impedance matching controller.

View full abstract

It is known that some of the mechanical and/or electrical dynamics can be written in the form called ‘Matrix 2nd-order system’ or ‘2nd-order system’ in short. On the other hand, pole-assignment is a commonly accepted technique in designing a performance guaranteeing regulator for the ordinal state space first order system. Recently, control design problems combining these two, that is, synthesis of a state feedback controller for the 2nd-order systems achieving closed loop pole-assignment while preserving its special form of dynamics, have been investigated. The proposed regulator is ‘numerically robust, ’ in the sense that a little perturbation on system matrices would not cause a large shift on the closed loop poles. But this does not suffice as a solution of the robust controller design problem such that matrices contained in the 2nd-order systems are the interval matrices with their lower and upper bounds given. In this paper, robust pole-assignment of the 2nd-order system with interval matrix uncertainties to a circular region in the LHP is discussed. Based on an evaluation of the perturbation in the coefficient space, we have applied the D-stable set expression in the coefficient space that we have obtained recently to determine the feedback gains. It will be shown that the design will finally be reduced to the problem of finding a solution of at most 16 simultaneous inequalities. Also, a numerical example will be given to illustrate the routine of the proposed design procedure.

View full abstract

This paper studies the l₁ controller design with frequency domain constraints using the duality theorem. For this class of problems it was shown in the literature that the standard finite dimensional approximation of the dual problem in the space of finite support sequences may not converge to the optimal value. In this paper, a condition assuring no duality gap is studied, and a way to approximate the dual problem with arbitrary tolerance by finitely many variables is proposed. A key observation is the property of so-called rank interpolation constraints.

View full abstract

This paper describes a design scheme of Generalized Minimum Variance Control (GMVC) for time-varying systems (TVS). Two controllers are considered: one having a time-varying parameter multiplication in consideration of a time-shift-operator (Diophantine type GMVC), the other being without the time-varying parameter multiplication (Model predictive type GMVC). Both algorithms are discrete in a noise characteristic and well suited for the TVS. The aim of this paper is to introduce an advantage for modifying the time-varying parameter multiplication.

View full abstract

PID controllers have been employed for most chemical processes. However, it is quite difficult to choose suitable PID parameters for such systems, because chemical processes are generally complex multivariable systems with strong interactions. On the other hand, model predictive control schemes, e.g., Dynamic Matrix Control (DMC), have been often applied for such systems, and relatively good control performances have been obtained. However, they require lots of cost and the long term for development. This paper presents a new tuning scheme of PID parameters, which is called a one-spot tuning scheme. Moreover, the newly proposed control scheme is evaluated on an atmospheric distillation process.

View full abstract

This paper studies on-orbit system identification experiment using ETS-VIII, which will be launched in 2003 by H-II A rocket. In the proposed method herein, the state-space model of the closed loop system is firstly realized by sigular value decomposion of Hankel matrices. The effect of controller is then eliminated by iterative least square curve fitting process using time response signal. Numerinal simulation of ETS-VIII shows the validity of the algorithm.

View full abstract

This paper studies the problem of stabilization of extending space structures. We derive the equations of motion of the structure extending in orbit under the gravity gradient torque environment. Because the equation is a time-varying system, the existing condition of gravity gradient stabilization developed for the non-extending system is not valid for the case. Therefore we newly derive a stability condition of time-varying systems, based on which the asymptotically stabilizing control law is derived. The obtained result is confirmed by some simulations. It is also shown that the same stability condition is applicable when the structure has physical parameter errors.

View full abstract

In this paper we use the theory of dynamic splines to reconstruct from noisy directional data the boundaries of a room. We show that we can determine the location of the corners and walls of the room with high accuracy. The dynamic splines that are used are periodic and are obtained by optimizing over both control and initial data.

View full abstract

This paper proposes a co-evolution model toward the understanding of the emergence of social morals. The model has two layers, agents' layer and scores' layer. An agent means an individual in Iterated Prisoner's Dilemma (IPD) and a score means a payoff matrix in dilemma games. In this model, behaviors of IPD agents and each value in payoff matrix are co-evolved. Simulation results showed that scores of IPD games are gradually evolved toward the cooperative score which gives more payoff for cooperative strategies and less payoff for defective strategies under all parameters we have tested. However, under a certain limitation of the scores, we observed that strategies of IPD agents could be evolved toward defective, even if the scores are cooperative. Furthermore, we could observe a periodical changes of the scores and the strategies as follows under a certain limitation of the scores. First, all values of all scores move to the cooperative scores and most agents take cooperative strategies because defective strategies of agents earn less payoff. Under the cooperative circumstances, values of scores which have no influence on cooperative moves are mutated freely. Then, this mutation allows invasion by defective strategies created by mutations. Then the cooperative scores which allow the invasion die out, and the world is dominated by the cooperative scores again.

View full abstract

In this paper, we suppose an agent network system in a multi-agent environment for process control and propose a calculation method of agent credibility, which is computed by aggregating other agents' evaluations. Two models called “center-of-gravity model” and “reflection model” are presented as the calculation methods. A test simulation was performed using a simple tank system, where the tank level is to be kept at a designed level. When an agent sends wrong data to other agents, the recipients calculate the credibility of the sender to identify faulty agents. Diagnosis agents identify a faulty condition by evaluating credibility of other agents based on constraints related to the task process. A tank agent has been introduced to distinguish the tank failure from agent failures. The results showed agents identified faulty agents while the controller agent maintained the tank level without any problems. Agents also detected the tank failure through decline of the credibility of the tank agent.

View full abstract

Humans are able to generate various kinds of cooperative behaviors simultaneously with mutual adaptation. The purpose of this study is to propose new concept of the human-robot cooperation by regarding the process of ‘Co-emergence’ observed characteristically at the human-human cooperation as ideal. Especially, in this paper, we analyzed the dynamics of Co-emergence, which characterized that cooperation, by measuring the cooperative walk between human and walk-support robot ‘Walk-Mate’ equipped with Co-emergence model. As the results, it was clarified that the temporal development of the dynamics of human's walking was similar to of the robot's walking at the cooperation them. At the robot side, the emergent process as the process of mutual constraint between the body model and the internal model that were the components of Co-emergence model was observed. Correspondingly at the human side, such process was observed between the step of foots and the swing of arms. In addition, it was clarified that both these processes were cyclic and developed synchronically between them. Thus these results suggested that these cyclic processes were developing with adapting their own dynamics mutually, and the validity of Co-emergence model.

View full abstract

Various mathematical models of rotation are used for various applications of engineering. Rotational motion can be described by Euler's motion equation and the angular rate vector can be obtained by solving the equation. Various differential equations by which rotation can be obtained from angular rate vector are there corresponding to the models of rotation.
In some problems concerning the rotation of a spinning rocket or a satellite, the rotational axis fixed on the body may be considered. It is shown that rotation can be represented using of two parameters showing the orientation of the rotational axis fixed on rotational system and the rotation angle around the axis. Its deferential equation is derived in this paper.

View full abstract

This paper presents an optimal current control design for permanent magnet synchronous motor by ILQ design method. It is clarified that the ILQ design method leads to excellent current-control which is robust for variation of armature resistance.

View full abstract

Two-stage tabu search is proposed to solve parallel machine problems. At the first stage the candidate list is employed and the jobs assigned to each machine are ordered by NEH. At the second stage the best solution obtained at the first stage is improved by enlarging its neighborhood.

View full abstract

In time-series data prediction, periodic and long-term components are often separately handled in a pre-devided form. We have investigated a neural network model that can identify the two parts, simultaneously. The model consists of two parts of the network arranged in the answer-in-weights structure. In this article, we study its long-term component part.

View full abstract