All issues
Volume 18, Issue 11
Displaying 1-13 of 13 articles from this issue
- |<
- <
- 1
- >
- >|
-
Yoshinori KAWASAKI, Masanori GOTOH, Zenta IWAI1982 Volume 18 Issue 11 Pages 1037-1044
Published: November 30, 1982
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn this paper the design problem of decoupling control system is studied for the linear time-invariant systems which contain timedelays in the control variables. The new state feedback law which can be realized practically has the term consisting of the past informations about control variables in addition to the conventional feedback law. By introducing this feedback contool law, the decoupling of the system can be attained for the broad class of multivariable systems with time-delay elements.
Further, we have introduced the concept of input transformation for relaxing the constraints on the state prediction problem of the decoupling systems, so that the actual design conditions are improved considerably compared with the conditions proposed by Rekasius et al..
Similar problem stated here was treated by Nakanishi et al. in the general form. However, their conditions are seemed to be complicated, so we gave a similar results as Corollary 1 of Theorem 2 in legible form.View full abstractDownload PDF (880K) -
Masahiro KUROSAKI1982 Volume 18 Issue 11 Pages 1045-1051
Published: November 30, 1982
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSAn airplane carries gradiometers which measure all components of the gravity-gradient tensor. The airplane is flying at constant speed and altitude in a constant direction, and good estimates are desired of the gravity deflection along and perpendicular to the flight direction. The gravity variations are described by the top three layers of a 5-layear statistical model due to Heller. The estimation is analyzed by two separate methods: (i) a low-order state-space model is constructed to approximate the Heller model and to apply Kalman filter: (ii) Wiener filter theory is used to obtain asymptotic results for high gradiometer accuracy, corroborating the results obtained in (i).View full abstractDownload PDF (806K) -
Katsunobu KONISHI, Zhong-gui YOU, Toshio YOSHIMURA, Takashi SOEDA1982 Volume 18 Issue 11 Pages 1052-1058
Published: November 30, 1982
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper is concerned with the convergence problem of a recursive identifier for single-input single-output linear discrete-time stochastic systems with unknown constant parameters. The identifier was designed by introducing a compensator which satisfies a strictly positive real condition, and it includes the well-known recursive extended least squares method (the RELS method) as a special case of no compensator. Firstly, it is proved that the parameter estimates converge to its true values with probability one. Secondly, it is shown that the convergence analysis method given by Solo for the RELS method is not always applicable to the identifier because of the existence of the compensator, and that the evaluations given here on the rate of convergence and the identifiability condition become more precise than that given by the above analysis method.View full abstractDownload PDF (810K) -
Ichiro YAMADA, Mitsuo NAKAGAWA1982 Volume 18 Issue 11 Pages 1059-1066
Published: November 30, 1982
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSPositioning control is a fundamental technique common to many mechanisms such as data communication terminals. industrial robots, etc. To implement faster and more accurate positioning control, it is necessary to consider the oscillatory characteristics of the positioning mechanism. From a practical aspect, it is also important to ssure good positioning when parameter variations of the mechanism occur.
First, the positioning control, is evaluated by the amplitude of residual vibrations which are caused by natural frequency variations in the mechanism. The residual vibration characteristics for three basic control loop structures are examined in detail at several design points. These show the necessity for the insensitive design method particularly in the fast region.
Next, for positioning control which is insensitive to parameter variations, an design method. using the multi-design-point system is proposed, and its validity for comparatively large parameter variations is confirmed both numerically and experimentally.
Moreover, a design method using the multidesign-point system is shown to contain the conventional design method using parameter sensitivity functions, and therefore to be a more general design method.View full abstractDownload PDF (953K) -
Tsuneo YOSHIKAWA, Toshiharu SUGIE1982 Volume 18 Issue 11 Pages 1067-1073
Published: November 30, 1982
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn this paper, servo problems including sensor dynamics are considered in the transfer matrix setting. Firstly, we propose a new definition of internal stability which is suitable for transfer matrix representation and is fundamental in servo systems, and derive a necessary and sufficient condition for the internal stability. And a few characteristics of the stability are derived. Secondly, necessary and sufficient conditions for output tracking and output regulation of the stable system with sensor dynamics are obtained using above results. And in order to verify the validity of usual servo system synthesis methods which utilize the internal models and to have foresight in servo synthesis, the relation between the internal model and sensor dynamics are considered in detail.View full abstractDownload PDF (1032K) -
Takashi SUZUKI, Shinji SHINNAKA, Kanya TANAKA1982 Volume 18 Issue 11 Pages 1074-1079
Published: November 30, 1982
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSA generalized adaptive law for model reference adaptive systems was recently proposed by Landau et al., which could easily yield socalled a constant gain algorithm, a decreasing gain algorithm and a variable gain algorithm by forcing some constraints to it. This law was derived based on the hyperstability theory and the concept of positive dynamic system. Unfortunately the proof was very complicated and difficult to understand. This paper shows an elegant proof of the law using the direct method of Liapunov and the positive real lemma. It may be useful to deeply understand the law and its properties.
Next we examine an attractive algorithm obtained from the generalized adaptive law as a special case, which has a gain matrix keeping its trace constant. Attention has been called to it with experimental results showing its superiority to some well-known algorithms. No analysis of its properties, however, has been tried yet. The analysis is given here to make clear some interesting properties, which are demonstrated by simple numerical examples.View full abstractDownload PDF (704K) -
Yoshiteru KURODA, Akira MAKINO, Takafumi ISHIBASHI1982 Volume 18 Issue 11 Pages 1080-1085
Published: November 30, 1982
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThe purpose of this paper is to study some optimal boundary control problem with boundary observation in second order parabolic systems, where control and observation are performed in a pointwise way, respectively.
Pointwise controls are defined using a scalar multiple of the Dirac measure under the Neumann boundary condition. Pointwise observations are introduced using characteristic functions at observation points.
Results obtained are as follows: (1) In pointwise boundary control and -observation systems, both the control and observation can not be defined at the same time by Dirac measure; (2) If pointwise boundary control is defined by Dirac measure, the adjoint state will be reduced to a complicated system involving the Laplace-Beltrami operator.View full abstractDownload PDF (673K) -
Ario OHSATO, Takashi SEKIGUCHI1982 Volume 18 Issue 11 Pages 1086-1092
Published: November 30, 1982
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn this paper we consider the transformation of the polynomial form of generalized composite fuzzy relational equation into the equivalent monomial form in order to loosen substantially the severe difficulty in solving the polynomial form of generalized composite fuzzy relational equation for the unknown fuzzy relation.
First, we assume the generalized binary operation * on the range space of the membership function, _??__??_{u|0≤u≤1}.
Second, the generalized composition of two fuzzy relations is defined by applying an arbitrary ordered pair of two generalized binary operations on _??_.
Third, we show the necessary and sufficient condition for the transformability of the polynomial form into the equivalent monomial form. Furthermore, it is concluded that the transformability is guaranteed, if an algebraic structure of the range space _??_ is a commutative semi-ring.
Finally, we present some examples of the algebraic structure which guarantees the transformability of the polynomial form into the equivalent monomial form.View full abstractDownload PDF (916K) -
Yoshiaki ARAKAWA, Akio SHINOHARA1982 Volume 18 Issue 11 Pages 1093-1098
Published: November 30, 1982
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSA heat flow meter is to be attached on a surface to measure density of heat flow rate (W/m2) when heat flows from the surface of an object to surrounding gas. Theoretical analysis using a one-dimensional heat transfer model was performed to clarify the thermal disturbance that occurs under the influence of the meter.
The results obtained show that the reciprocal of sensitivity of the meter, i.e. the ratio of the density of heat flow rate without disturbance to the output voltage of the meter contains a correction term for the thermal disturbance and that its value is related to the total thermal conductance of the meter, thermal conductivity of the material the surface consists of and the differential of density of heat flow rate with respect to surface temperature. The part related to the thermal conductivity of the material of the correction term is given by the dimensionless quantity (1-β), where β=Δt/ΔT, Δt is the temperature variation on the surface caused by attachment of the meter, and ΔT the total temperature drop across the meter.
Experiments and numerical analyses indicated that the value of β decreased as the thermal conductivity of the surface material increased, and, in the case of a surface in iron, as the thickness of iron increased. For instance, β=0.55 and 0.125, for iron plates 0.27mm and 4.5mm thick, respectively. The β value of 0.55 was obtained also for a glass plate with a thermal conductivity of about 1W/(m·K), which indicates thermal disturbance of the same order as that for a 0.27mm thick iron plate.View full abstractDownload PDF (758K) -
Sukita NAKAHARA1982 Volume 18 Issue 11 Pages 1099-1103
Published: November 30, 1982
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThe near-infrared absorption type apparatus which measures continuously the rate of perspiration or the amount of sweating in certain period on the body surface is described.
Dry air is passed at a measured rate through the measuring cell adhered to the skin surface tightly, then into the near-infrared hygrometer. In this process, dry air contains moisture given off from the skin surface. The out-put of this hygrometer is recorded and the rate of perspiration is calculated.
This apparatus is fast enough to detect perspiration changes without appreciable lag, and sensitive to perspiration levels as low as 0.01mg/10cm2·min. The accuracy of the apparatus better than 3% is obtained.View full abstractDownload PDF (665K) -
Theoretical Investigation by Mathematical ProgrammingKaname YOSHIDA, Yasushi SHIMIZU, Toshi TAKAMORI, Masatada YONEMOCHI1982 Volume 18 Issue 11 Pages 1104-1109
Published: November 30, 1982
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper presents the investigation of the optimal valve control which minimizes the consumption of air, or that of energy. At first, the simultaneous differential equations which describe the dynamics of a pneumatic cylinder are changed into a set of discrete equations, and the nonlinear optimization problem of several variables, which aims the minimization of the consumption of air, is set up. And then, we apply the nonlinear mathematical programming to this problem, and the optimal state variables at each time are obtained. Finally, we simulate the continuous model with the optimal valve control patterns which are obtained by solving the optimization problem, and compare the simulated responses with the solutions of the problem. The comparison between the two systems shows that the solutions of the problem agree well with the results of the simulation. And with this control, the consumption of air will be decreased to about 50∼60%, and it will be possible to position a piston precisely. We can conclude that the proposed method is very effective when we decide the position of the piston under condition of minimizing its energy consumption.View full abstractDownload PDF (680K) -
Hideo HANAFUSA, Tsuneo YOSHIKAWA, Yoshihiko NAKAMURA1982 Volume 18 Issue 11 Pages 1110-1116
Published: November 30, 1982
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSContouring control of a robot arm is a significant problem in many manufacturing processes, such as arc-welding, sealing, painting, shape measuring and various inspections. For this control play-back control has been applied so far. But it has following inherent problems.
i) Play-back control requires much time and labour in the teaching process.
ii) Play-back control system cannot adapt itself to setting errors and to individual differences of objects.
iii) Control performance depends directly upon the operator's teaching ability and upon how accurately objects are placed.
We proposed Autonomous Contouring Control of a robot arm in order to overcome those problems. In this control the robot system obtains the information about the position and the shape of the object using the sensors attached to the robot hand.
In this paper we analyzed the basic characteristic of Autonomous Contouring Control System in two dimensional case. Based upon the analysis we proposed a scheme for the control system design. And then applying the scheme to an articalated robot arm driven by hydraulic cylinders, we verified its effectiveness for contouring control of unknown curves by experiments.View full abstractDownload PDF (1739K) -
Akira MOHRI, Makoto IKI, Masaru TAKATA1982 Volume 18 Issue 11 Pages 1117-1122
Published: November 30, 1982
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn this paper, some considerations about improvement of transient characteristics of multi-input control systems are described. Those are done from the viewpoint of eigenvalues and eigenvectors assignments. In general, for multi-input systems, degrees of freedom remain in the feedback gain matrices after the eigenvalues have been assigned. Therefore, transient characteristics may be improved by making use of these degrees of freedom. In this paper, this improvement is realized.
First, the maximum or minimum peaks of the transient responses are approximately estimated by the use of the spectral decompositions of matrices. Next, the degrees of freedom which remain after the eigenvalues have been assigned are expressed as the variable parameters in the eigenvectors and the values of parameters which reduce the maximum or minimum peaks estimated above are determined. Finally, the effectiveness of the method mentioned above is shown by the simple examples.View full abstractDownload PDF (683K)
- |<
- <
- 1
- >
- >|