All issues
Volume 24, Issue 5
Displaying 1-18 of 18 articles from this issue
- |<
- <
- 1
- >
- >|
-
Kamekichi SHIBA, Koichiro DOI1988 Volume 24 Issue 5 Pages 433-437
Published: May 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSDifferential pressures type mass flowmeters, here reported, consist of a constant flow apparatus and 3 or 2 orifice meters. These components compose a kind of hydraulic calculating circuit.
In this paper, measuring principles of them are derived by the use of Bernoulli's theorem. Equations expressing these measuring principles are as follows:
ρQ=k3(ΔpA-ΔpB+ΔpC), ρQ=k2√ΔpAΔpC, where ρQ is mass flow rate, ΔpA, ΔpB, ΔpC are differential pressures produced by orifice A, B, C and k3, k2 are meter constants. Mass flow rates can be obtained by measuring 3 or 2 differential pressures.
The two kinds of differential pressures type mass flowmeters are produced by way of trial. Relationships given by the above equations are confirmed by experiments carried out with the trial mass flowmeters.
Constants k3 and k2 must be determined experimentally as they include discharge coefficients.View full abstractDownload PDF (525K) -
Norihide YAMADA, Sadao FUJIMURA1988 Volume 24 Issue 5 Pages 438-444
Published: May 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSAn optical model for the propagation of light in a leaf is constructed by assuming the multiple-reflection of light in the leaf and by taking the actual structure of the leaf into account. It gives spectral reflectance and transmittance of a leaf as functions of chlorophyll content of the leaf. We measured spectral reflectance, transmittance and chlorophyll content of leaves of hydrangea and chinquapin in order to confirm the validity of the model. Spectral reflectance and transmittance were calculated from the chlorophyll content by the model and compared with the measured values. Good agreement between them shows the validity of the model.View full abstractDownload PDF (880K) -
Kazuhiro YOSHIDA, Shigeo HIROSE1988 Volume 24 Issue 5 Pages 445-451
Published: May 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThe authors have developed a new type laser range finder, which enables a mobile robot to measure the distance to a natural object located one to five meters ahead with an error less than two percents. The system uses laser as weak as 1[mW], a safe level for human eyes, and is workable under bright daytime sunlight of about 100, 000[lx]. This laser range finder can quickly process data while compensating the motional error induced by the robot on which it is mounted and makes real time measurements of objects located ahead.
This laser range finder is featured by plural redundant range measuring systems and a newly introduced real time signal processing technique of the redundant signals. The laser range finder shows strong noise reduction characteristic by acting together with two conventional methods, namely, conventional optical filtering which preliminarily eliminates noise and the DDD signal processing proposed by the authors for separating reflected pulseform signals from ambient light on video signals.
The effectiveness of the proposed laser range finder system was demonstrated under direct sunlight by an experimental model mounted with two CCD cameras and a semiconductor laser projector of 1[mW] output power with 830[nm] wave length.View full abstractDownload PDF (2284K) -
Hideo TAI, Kazushige YAMAMOTO, Takeshi ABE, Takashi UEKI, Hiroaki TANA ...1988 Volume 24 Issue 5 Pages 452-458
Published: May 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIt is known that the 3.392-μm He-Ne laser emission (Ne 3s2-3p4) is strongly absorbed by methane while the 3.391-μm emission (Ne 3s2-3p2) is absorbed only weakly. This property of methane has been utilized in differential absorption laser radars (DIAL) for methane sensing. In conventional DIAL systems, a pair of lasers, one emitting the 3.392-μm line and the other 3.391μm, have been used as a light source. These systems need, however, not only complex optics but also its precise adjustment to align the two laser beams on the same optical path. These problems are solved by using a dual-wavelength He-Ne laser developed recently. This laser emits the two lines coaxially whose intensities are modulated with equal amplitudes and opposite phases to each other. In the present paper two simple and sensitive methods of quantitative remote sensing of methane using a dual-wavelength HeNe laser are demonstrated. In the first method, two detectors are used, one monitoring both wavelength components, and the other only the 3.391-μm component in the reflected radiation. In the second method, the double modulation technique makes it possible to use only one detector. Laboratory simulation shows that both methods are capable of measuring quantitatively density-pathlength products down to about 40ppm·m at a signal-averaging time of 3s with reflecting targets at a 2.5-m distance. When the reflectivity is high, a 100ppm·m leakage can be detected with a target as distant as 30m.View full abstractDownload PDF (1859K) -
Yasuhiko MUTOH, Lin TSUI, Kunihiko ICHIKAWA1988 Volume 24 Issue 5 Pages 459-466
Published: May 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn this paper, the structure of the generalized interactor matrix of the linear multivariable systems is concerned using the properties of the di-column improperness of a polynomial matrix. The gneralized interactor matrix of a transfer matrix T(s) is defined by the polynomial matrix L(s) which satisfies lims→∞L(s)T(s)=K: K is a nonsingular constant matrix where L(s) is not necessarily a lower left triangular matrix.
The purpose of this paper is to discuss the necessary and sufficient condition for the polynomial matrix L(s) to be the generalized interactor matrix of T(s). It turnes out that the property of the minimal di-Γ-matrix which is a constant matrix derived from the numerator polynomial matrix of T(s) decides the structure of the generalized interactor matrix L(s).View full abstractDownload PDF (917K) -
Avoiding the Problems on Unstable ZerosTsutomu MITA, Yuichi CHIDA1988 Volume 24 Issue 5 Pages 467-474
Published: May 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn the design of digital controller, the existence of unsatble zeros (for example, unstable limiting zeros) makes it difficult to construct many control systems, such as inverse systems, robust controller, model matching system and model following adaptive controller. This paper proposes a new control methods in which two outputs are used which are the output at the frame sampling point and auxiliary output at the time leading to the frame sampling point by mT(0<m<1), where T is frame sampling period. If such a system description is used, the resulting system is proved to have no invariant zeros, that is, the two set of outputs have no common invariant zeros. When such a system description is used, it is easy to construct stable inverse even though the continuous time plant is non-minimum phase, and unknown input observer is constructed.
The dual description of this system is used to assign invariant zeros and used to design model matching system. Such control strategy is completely new but simple, so many applications are expected.View full abstractDownload PDF (1051K) -
Katsunobu KONISHI, Toshio YOSHIMURA, Hiroaki OTSUKA1988 Volume 24 Issue 5 Pages 475-482
Published: May 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper presents a method to solve the Bezoutian equation for the indirect pole-placement control of single-input single-output linear non-minimum phase systems with unknown constant parameters. The method consists of three steps; (i) the elementary transformation of the Sylvester resultant matrix into the upper triangular form by pivot searches, (ii) the exclusion of the common factors from each rows of the matrix obtained in the step (i) when a pivot becomes too small or zero, and (iii) the calculation of controller parameters from the linear simultaneous equations of which coefficient matrix is the transpose of the matrix obtained by the step (ii). Since the polynominal composed of the common factors is obtained directly from the row elements above the small pivot, the factrizations of the system polynomials are not required. These solusion procedures are illustrated by the numerical examples, and the validity of the proposed method is proved by using the concept of the linear transformation of state variables.View full abstractDownload PDF (1086K) -
Yasuyuki FUNAHASHI, Hisao KATO1988 Volume 24 Issue 5 Pages 483-489
Published: May 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn this paper, we propose a design method for the minimal-time deadbeat digital controller with optimal robustness using two-parameter compensation scheme. The whole system is guaranteed to be internally stable, because the disired controller is determined among the class of all stabilizing controllers.
In the case of one-parameter compensation scheme, the robustness optimization is in conflict with the deadbeat steps. So robustness can only be improved at the cost of the increase of deadbeat steps.
In this paper, using two-parameter compensation scheme, the minimization of deadbeat steps and robustness optimization can be carried out separately. The method possesses a feature that the optimal robustness is attained with the minimal-time deadbeat controller.
We also design a robust deadbeat controller among the class of robust tracking controllers. The robust tracking system is the one whose steady-state error is zero whenever internal stability is maintained for the slightly perturbed plant. It is shown that optimal robustness and minimal deadbeat steps are also achived simultaneously for this system.View full abstractDownload PDF (920K) -
Toshiro NORITSUGU, Tsutomu WADA, Junichi TOMONO1988 Volume 24 Issue 5 Pages 490-497
Published: May 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper examines the availability of a modern control theory to the pneumatic servo system implemented with a pneumatic cylinder and electro-pneumatic proportional control valves. Due to the low stiffness from air compressibility, a pneumatic system is limited in a response speed and very sensitive to the variation of frictions or load forces. These difficulties may be overcome with the high gain control or the control law including an integral action. On the other hand, the available electro-pneumatic control valves involve a considerable operating delay time, which makes such controls more difficult.
In this study, to compensate the valve delay, which is formulated as an input time-delay in the discrete state equation, for which the 0-type and 1-type optimal servo systems are designed. Experiments show that the control system design considering the valve delay makes the 1-type servo system satisfactory with both stability and response speed, so that the advanced position control performance can be obtained independently of load forces.
The proposed servo systems require only a position sensor, and estimate other state variables by the observer. Moreover, the used computer has so sufficient computational ability that it can also control some servo systems simultaneously to lower the cost for one cylinder. Consequently; it proves that the application of a modern control theory can fairly improve the control performance of pneumatic servos without losing its simplicity and low cost advantages.View full abstractDownload PDF (1004K) -
Kouji HAMADA, Kozo MATSUSHIMA1988 Volume 24 Issue 5 Pages 498-505
Published: May 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSA new adaptive model on the human operator in a Man-Machine system is proposed.
The model is constructed on the basis of a Man-Machine control experiment result. In this study, as a controlled object was used the 2nd order servo mechanism whose feed back loop suddenly breaks during operation.
The experimental results show that whole control process can be separated into 3 parts. The second part of them is the adaptive process in which human operator has to adapt his/her control action to the new situation, and which is a kind of damped Bang-Bang operation.
As a result, the model has a hierachic structure with 2 sets of Meta rules and 3 sets of Fuzzy rules. In identifing the structure of Fuzzy rules, the clustering technique is applied. And the validity of the adaptive Fuzzy model is verified with simulation.View full abstractDownload PDF (1002K) -
Ichiro AKAHORI, Shinji HARA1988 Volume 24 Issue 5 Pages 506-513
Published: May 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSA new type CAD system for control system analysis and design has been developed on a workstation, which is written in Smalltalk-80, an object oriented language, combined with C and Lisp. The system has the following features: 1) The system can aid designer not only for numerical computation served in the conventional CAD systems but also for symbolic manipulation and knowledge processing. Since these all facilities are integrated in the uniform Smalltalk environment, the system can provide the advanced functions, and then the designer can conduct almost all operations required for analysis and design.2) The system is based on the concept of “virtual environment for analysing and designing control system”, which can provide an environment very close to real situation using real instruments. The object oriented programming paradigm is useful in making this environment. 3) The system has user friendly interface with multiwindow and mouse.View full abstractDownload PDF (2375K) -
Norito IWAO1988 Volume 24 Issue 5 Pages 514-521
Published: May 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn order to develop steady state models of blast furnase systematically, the auther proposed “Basic Model of Blast Furnace operation (BM-BF)” in his Paper (1).
According to this BM-BF, the equation of relation between reaction raw materials and reaction products (I/O-BE: Input/Output-Basic Equation) could be casily derived by making use of material balance and reaction products analysis data.
And the equation of relation between reaction rate, reaction products and reaction raw materials (N-BE: Node-Basic equation) could be also obtained at each node of BM-BF.
In his next Paper (2), the auther proposed “Basic Model of Simplified Blast Furnace (BM-SBF)” in order to avoid mathematical complexity resulting from many I/O variables of real blast furnace. He showed that the equations to be solved outnumbered the unknowm variables to be obtained when calculation was made to get reaction products by I/O-BE.
Therefore, he indicated that the various combination of solution could be obtained depending on how to select the equations.
In the first part of this Paper, based on two Goodness Criteria (rmsD, rmsN), the various calculating methods of reaction products mentioned above were compared each other by adopting the process data (raw materials and reaction product analysis data) generated randomly.
In order to apply two Goodness Criteria quantitatively, it was necessary that the calculating methods of reaction products should be linearized with the process data.
Attempts were made to linearize the calculating methods and he found the way of linearization with quite good accuracy. The methods of linearization and its results were also described, in the second part of this paper.
Finally, the auther indicated that the calculating method employed traditionally could be derived from two Goodness Criteria proposed here.View full abstractDownload PDF (1024K) -
Hisao MIYANO1988 Volume 24 Issue 5 Pages 522-527
Published: May 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSA new regression algorithm for multidimensional scaling is given to find an optimal monotone function which approximates the given data with partial order restrictions under a least squares criterion. In multidimensional scaling, Kruskal's “Up-and-Down Blocks Algorithm” has been used extensively, which is only applicable to the simple order restrictions, and ordinal data which produces the partial order restrictions has been preprocessed in order to get simple order restrictions. Our algorithm makes it possible to analyze such data rather directly.View full abstractDownload PDF (742K) -
Yoichi UETAKE, Shigenori OKUBO1988 Volume 24 Issue 5 Pages 528-530
Published: May 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSWe study feedback pole placement for continuous descriptor (generalized state-space) systems depending on parameters, which have the formE(ω)x(t, ω)=A(ω)x(t, ω)+B(ω)u(t, ω).Here, x(t, ω) and u(t, ω) are a descriptor variable and an input, respectively. The entries of the above matrices are real- or complex- valued continuous functions defined on a subspace of a multi-dimensional euclidean space.
Such systems arise in the study of descriptor systems whose coefficients are continuous functions of one or more parameters, neutral delay-differential systems, and spatially-distributed systems in descriptor form.
In the paper, we first define the pointwise-controllability for continuous descriptor systems depending on parameters of the above form. Next, we derive a sufficient and necessary condition for pointwise control lability in terms of a pointwise rank criterion. Then, for pointwise controllable systems, we describe pole placement by the descriptor variable feedback u(t, ω)=f(ω)x(t, ω) using feedback gains that also depend on parameters.View full abstractDownload PDF (392K) -
Tomomichi HAGIWARA, Mituhiko ARAKI1988 Volume 24 Issue 5 Pages 531-533
Published: May 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSA method of pole assignment is proposed for discrete-time systems with delays at plant input. The procedure consists of the usual pole assignment for the system without delays and simple algebraic calculation involving tuning parameters.View full abstractDownload PDF (325K) -
Shinji SHINNAKA, Naoki KAWAMATA, Mikio SATO1988 Volume 24 Issue 5 Pages 534-536
Published: May 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper presents a well generalized parameter adjustment algorithm for adaptive systems. It enables unified analysis of properties and relations of a variety of algorithms reported individually.View full abstractDownload PDF (371K) -
Fuminori KOBAYASHI, Hidekazu MACHIDA1988 Volume 24 Issue 5 Pages 537-539
Published: May 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThe calculation of the Hilbert transform has conventionally been accomplished by using FFT, has some redundancy. The proposed algorithm attains speed by employing the fast Hartley transform and calculation-step reduction. This algorithm is 30∼50% effective than the conventional ones.View full abstractDownload PDF (345K) -
Shinobu SASAKI, Yoshikuni SHINOHARA1988 Volume 24 Issue 5 Pages 540-542
Published: May 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSDownload PDF (364K)
- |<
- <
- 1
- >
- >|