All issues
Volume 21, Issue 3
Displaying 1-16 of 16 articles from this issue
- |<
- <
- 1
- >
- >|
-
Setsuo SAGARA, Takami MATSUO, Kazushi NAKANO1985 Volume 21 Issue 3 Pages 213-219
Published: March 30, 1985
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThe aim of this paper is to construct an observer for linear multivariable systems with unknown inputs by using the following result: a full order observer converges to a reduced order observer as some poles approach infinity in the left half complex plane. We show that this observer is equivalent to the perfect observer, and investigate a condition to construct it without the divergence of the initial estimation errors. This construction method allows us to clarify the degree of freedom in assigning the eigenvalues and eigenvectors of the unknown input observer.View full abstractDownload PDF (860K) -
Mitsuo OHTA, Eiji UCHINO, Kazutatsu HATAKEYAMA1985 Volume 21 Issue 3 Pages 220-225
Published: March 30, 1985
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn this paper, a new trial on the parameter estimation for stochastic environmental systems has been proposed. The measure of statistical independency between an input signal and an intruded observation noise is newly adopted as the error evaluation measure for this parameter estimation. This measure can evaluate not only the neighbourhood of mean value (such as, related to the well-known least-square-error criterion), but also the end of level distribution form of random phenomenon (such as, used very often as one of actual environmental standards in the field of environmental noise evaluation). Concretely, the joint probability density function, P(x, v), for an input signal, x, and an observation noise, v, is positively expressed in some series expansion form to grasp the statistical independency hierarchically. Then, after naturally defining the deviation from independency by ε(x, v)_??_P(x, v)-P(x)P(v), several types of new estimation methods have been derived according to how to evaluate and then decrease this deviation. The validity and effectiveness of these proposed methods have been experimentally confirmed by applying them to the actual estimation problem for the sound insulation parameters in an acoustic field.View full abstractDownload PDF (904K) -
Takashi MORIYASU, Yoji MASAGO, Koichi INOUE1985 Volume 21 Issue 3 Pages 226-232
Published: March 30, 1985
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper presents an evaluation method based on pair comparisons with preferential strength which is obtained as categorical judgment by a decision maker. Considering judgmental consistency of pair comparisons with preferential strength, we discuss to find consistent structures which are the nearest to judgment by the decision maker. The problem is formulated as minimizing a criterion which measures difference between two matrices. The problem is solved by an integer programming. We present an application of this method to evaluate some journals of personal computers.View full abstractDownload PDF (980K) -
Makoto TSUKIYAMA1985 Volume 21 Issue 3 Pages 233-240
Published: March 30, 1985
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIt is important to locate valves properly in a water distribution network in order to control the water pressure. This problem was hard and needed a large amount of time and a large amount of experienced designer's knowledge to be solved.
In this paper, a method for locating valves in designing the water distribution optimal pressure control systems is presented, by which the good location of valves is obtained systematically and simply. This method saves the time required in designing the control systems.
First, the valve location problem and the difficulty of solving it are discribed. Second, we discuss the relation between pressure controllability and valve location, considering the network connection and the water distribution network equations. We also study this problem based on the numerical experiment for a simple water distribution network. And we propose a method of valve location. Finally, this method is verified to be effective by experiments applying to the real water distribution network.View full abstractDownload PDF (1170K) -
Kunihiko ICHIKAWA, Akihiro ITO1985 Volume 21 Issue 3 Pages 241-247
Published: March 30, 1985
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn this paper, a design method of multivariable discrete time adaptive control is presented. In the first, decoupling with simultaneous exact model matching for multivariable system is developed and then is applied to the construction of adaptive control system. The design method developed in this paper is based on the clear recognition that the model reference adaptive control is an extention of exact model matching under the condition of that the plant parameters are unknown. By choosing a reference model as a diagonal transfer matrix where each element represents a desired transfer function, decoupling and exact model matching are achieved simultaneously. Computer simulations show good results.View full abstractDownload PDF (905K) -
Hitoshi YANO, Masatoshi SAKAWA1985 Volume 21 Issue 3 Pages 248-255
Published: March 30, 1985
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSDuring the last few years, numerous methods related to multiobjective optimization and decision-making have been proposed. Among them interactive techniques assume that the decision maker (DM) is able to assess his preference on a local level to a particular solution. In order to help the DM express his degree of preference, the trade-off rates information between two objectives is very useful. On the other hand, one of the most powerful methods of generating Pareto optimal solutions is the weighted Tchebycheff norm method. It can be applicable to multiobjective optimization problems involving nonconvexities, but the trade-off rates between the objectives have not been given explicitly. In this paper, we develop the meaningful results which relate the trade-off rates within the Pareto optimal solution set to the Lagrange multipliers and the weighting vectors of the weighted Tchebycheff norm problem. Furthermore, various aspects of the results obtained are illustrated by the numerical example.View full abstractDownload PDF (956K) -
Stanislaw K. DUDA, Ryozaburo TAGAWA1985 Volume 21 Issue 3 Pages 256-261
Published: March 30, 1985
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSA general theory concerning a least possible order d of a linear dynamic feedback for stabilization of linear SIMO systems is presented (d≤v0-1, where v0 is the observability index of a given system). The method applies to controllable and observable systems of an arbitrary order. A computationally attractive design procedure which provides the output feedback in a form suitable for direct realization and an illustrative example are given.View full abstractDownload PDF (708K) -
Shizuo MAWATARI1985 Volume 21 Issue 3 Pages 262-269
Published: March 30, 1985
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSWhen we analyse various dynamic behaviours of a real system under consideration by the method of system dynamics (SD), we must begin with construction of an SD model which serves modeler's purposes. The most essential step in the modeling process is to include some major structures necessary to represent the relevant dynamic behaviours of the real systems. At this point, clarifying the characteristic of the included structure is very important. Usual methods for describing the charcteristic are intuitive devices applying causal-loop diagrams, flow diagrams, structural equations and verbal statements. But only these are not always sufficient for such description. Therefore, we must go over the step which merely explain structural equations and causal relations among system elements and then it is necessary to establish the theoretical method which exactly describe and analyse all characteristics of the model's structure.
In the present paper, we shall try to provide the means of solving the need. Accordingly, after establishing the general SD model by an axiomatic method, we construct geometrical representations of the SD model and the total flow diagram by the application of the combinatorial topology. On the basis of the representations, we can make clear the relation between integral cycles of the complexes and feedback loops in the SD model, thereby determine the structures of the SD model and the feedback loops.View full abstractDownload PDF (1205K) -
Toshihiko YAMAWAKI1985 Volume 21 Issue 3 Pages 270-276
Published: March 30, 1985
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn sampled-data systems, informations are transmitted in continuous time as well as at discrete instants. Therefore, a strict and unified representation of the systems needs a new pair of state and output equations.
Initially, the variable space of the systems is formulated to the sum set of real functions and Dirac's delta function series. And a sampling operation which satisfies the causality is defined in this space, in order to realize approximately sampled-data systems with or without holding elements. Then, a new system equation is set up for the systems obtained by introducing the above sampling to proper and linear continuous time systems.
Finally, some basic properties of sampled-data systems are shown by analizing generally the state transition of linear and first-order sampled-data systems, and the new system equation is suggested to be a unified expression of continuous time systems, discrete time ones and sampled-data ones all of which are proper and linear.View full abstractDownload PDF (943K) -
Kouji TAKAHASHI, Kensuke HASEGAWA, Zbigniew BANASZAK1985 Volume 21 Issue 3 Pages 277-283
Published: March 30, 1985
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThe paper describes an algorithm to synthesize a Petri net which can generate a given firing sequence.
The Petri net is one of the effective tools to represent event-condition systems which act asynchronously and concurrently, such as production systems in which work pieces, data or informations are processed in order by the system components. As we realize such systems including system control, how to realize the processing order given and how to arbitrate conflicts are fundamental problems. Taking them into consideration, the authors propose the synthesis algorithm of a Petri net which is able to generate a given firing sequence.
In a Petri net, the activation of a system component may correspond to either a firing of a transition or a token in a place. In the paper, however, the authors decided that the activation corresponds to a firing of a transition, then the firing sequence of the transitions is regarded as the order of the process. The firing sequence applicable to the proposed synthsis algorithm is allowed to have repetitive appearances of the same transition and partial parallel firing sequences which do not contain the same transition in common, in other word, are independent each other.
The synthesis algorithm consists of the following steps. (1) Connect the transitions to each other by places and arcs in the order of the given firing sequence. (2) Find out conflicts in the path, and if exist, prescribe the firing order among the conflicting transitions depending on the given specification. (3) For each conflict, make a path for the prescribed firing order by the same way as (1), and add it to the former graph. (4) Find out conflicts in the paths added at (3), if exist, prescribe the firing order and back to (3) again. By repeating them until there exists no conflict, a Petri net which generates a given firing sequence is completed.
The algorithm is executed by a computer, and we confirmed by simulation that a well-behaved Petri net is synthesized.View full abstractDownload PDF (659K) -
Yasuki NAKAYAMA, Katsumi AOKI, Hiroaki OHTA1985 Volume 21 Issue 3 Pages 284-291
Published: March 30, 1985
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSAs to laser Doppler velocimeter, the relation between the visibility and the diameter of a seeding particle was calculated by the simple method assuming that the intensity of scattered light from the particle is proportional to the integrated intensity of interference fringes on particle.
The analytical results using this method accurately agreed with the experimental results that were obtained by using newly developed wire disc and water channel disc respectively, which can correctly determine particle diameter, and it was clarified that fine wire diameter or particle diameter can be measured from the visibility.
Next, systematic measurements were carried out on the beam intersecting angle, the intensity ratio of two beams, the density of particles, the dimensionless aperture radius, the backward and forward scattering and the presence of plural particles in the measuring volume, and the relation between seeding particle size and the output characteristics of a laser Doppler velocimeter was clarified.View full abstractDownload PDF (3080K) -
Hiroshi HARADA, Hiroshi KASHIWAGI1985 Volume 21 Issue 3 Pages 292-297
Published: March 30, 1985
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIt is known that we can obtain primitive pentanomials over GF (2) from primitive trinomials f(x)=xn+xp+1 by sampling every s digit of the M-sequence generated by f(x). In this paper, the authors first describe the relations among n, p and s where we can obtain a primitive pentanomial g(x) by this method. Then we show that the remainder polynomial r(x) modulo g(x) at one of the i/s phases (1≤i≤s-1) of the sampled M-sequence consists of a few terms in some cases, and that r(x) becomes a rational polynomial r(x)=P(x)/Q(x) in other cases, where P(x) and Q(x) are both polynomials consisting of two or three terms.
From these results, the authors show that the partial properties of the sampled M-sequence at i/s phases resemble each other. Therefore, when we use an M-sequence generated by g(x) for generating pseudo-random numbers, we must avoid, as an initial n-tuple, not only an unsuitable n-tuple and some of its cyclotomic phases, but also its i/s phases.View full abstractDownload PDF (662K) -
Yoshiyuki SAKAWA, Akira NAKAZUMI1985 Volume 21 Issue 3 Pages 298-304
Published: March 30, 1985
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn this paper, we first derive a dynamical model of a rotary crane whose boom angle is controlled by a hydraulic cylinder. On the basis of the dynamical model, we propose an open-loop plus feedback control scheme. The crane makes three kinds of motion (rotation, load hoisting, and boom hoisting) simultaneously. The goal is to transfer a load to a desired place in such a way that at the end of transfer the swing of the load decays as quickly as possible. We first apply an open-loop control input to the system such that the state of the system can be transferred to a neighborhood of the equilibrium state. Then we apply a feedback control signal so that the state of the system approaches the equilibrium state as quickly as possible. The results of computer simulation prove that the open-loop plus feedback control scheme works well.View full abstractDownload PDF (767K) -
Hisato KOBAYASHI1985 Volume 21 Issue 3 Pages 305-307
Published: March 30, 1985
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSDownload PDF (338K) -
Hiroshi SHIBATA, Genmon MARUOKA, Kiyoshi HIRAI, Shiroh HATA1985 Volume 21 Issue 3 Pages 308-310
Published: March 30, 1985
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSDownload PDF (309K) -
Takehiro MORI1985 Volume 21 Issue 3 Pages 311-313
Published: March 30, 1985
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSDownload PDF (312K)
- |<
- <
- 1
- >
- >|