Transactions of the Society of Instrument and Control Engineers
Online ISSN : 1883-8189
Print ISSN : 0453-4654
ISSN-L : 0453-4654
Volume 18, Issue 9
Displaying 1-14 of 14 articles from this issue
  • Yoshifumi SUNAHARA, Shin-ichi AIHARA, Masaaki ISHIKAWA
    1982 Volume 18 Issue 9 Pages 871-878
    Published: September 30, 1982
    Released on J-STAGE: March 27, 2009
    JOURNAL FREE ACCESS
    This paper is concerned with the state and free boundary estimates for two phases stochastic Stefan systems. The principal line of attack is to introduce the stochastic variational inequality as the system model, which plays a role to embed the free boundary condition into the basic system equation. Thus, the existence theorem of the solution is proved within the framework of Hilbert function space concept. Finally, both the state and free boundary estimator equations are derived by using the martingale and innovation approaches.
    Download PDF (914K)
  • Shinji HARA
    1982 Volume 18 Issue 9 Pages 879-885
    Published: September 30, 1982
    Released on J-STAGE: March 27, 2009
    JOURNAL FREE ACCESS
    Existence conditions of a compensator such that the compensated system is satisfied two servo cenditions, namely output regulation and internal stability, are considered for linear multivariable systems with portial parameter perturbations based on the concepts of genericity and well-posedness. Four solvability concepts, namely (i) solvability at θ* (nominal value of parameter θ), (ii) generic solvability, (iii) well-posed-solvability at θ*, and (iv) generic well-posed-solvability, are introduced. Their necessary and sufficient conditions with matrix operations are derived and the relationships between these conditions are clarified. A design procedure of desired servo compensators is also presented.
    Download PDF (848K)
  • Keiji WATANABE, Masami ITO
    1982 Volume 18 Issue 9 Pages 886-892
    Published: September 30, 1982
    Released on J-STAGE: March 27, 2009
    JOURNAL FREE ACCESS
    This paper is concerned with finite pole assignment problem for linear systems with delay in state variables. It is shown that spectrum controllability implies finite pole assignability. A systematic design procedure of the feedback law is presented.
    Download PDF (760K)
  • Hayao MIYAGI
    1982 Volume 18 Issue 9 Pages 893-897
    Published: September 30, 1982
    Released on J-STAGE: March 27, 2009
    JOURNAL FREE ACCESS
    This paper presents a new type Lyapunov function of the synchronous generator with the effects of flux decay and voltage regulator. Lagrange-Charpit method is applied to the construction of the Lyapunov function. Although flux decay causes a product-type nonlinearity in the formulation of the electrical power output, a well-defined form of this nonlinearity is considered and is taken into account in the construction process. It is shown that the positive difiniteness of the obtained Lyapunov function and the negative semidefiniteness of its time derivative can be easily verified by using the nature of this nonlinearity.
    The Lyapunov function may then be used to estimate the critical switching time for power-system transient stability. The critical switching times given by this Lyapunov function are compared with those obtained by conventional Lyapunov function, showing the superiority of the estimation.
    Download PDF (617K)
  • Jiro SHIMONISHI, Takao HINAMOTO, Sadao MAEKAWA
    1982 Volume 18 Issue 9 Pages 898-904
    Published: September 30, 1982
    Released on J-STAGE: March 27, 2009
    JOURNAL FREE ACCESS
    This paper treats the problem of constructing inverse two-dimensional (2-D) causal systems with inherent delay (M, N) for Attasi-type 2-D systems.
    First, the notion of an MN-delay inverse is introduced for 2-D causal systems. This is considered to be a generalization of an L-delay inverse for 1-D systems given by Sain and Massey. A necessary and sufficient condition for the existence of an MN-delay inverse is then shown for Attasi-type 2-D systems.
    Finally, a method is presented for constructing an MN-delay inverse for these systems.
    Download PDF (661K)
  • Koichi INOUE, Takashi MORIYASU, Tadashi KIMURA
    1982 Volume 18 Issue 9 Pages 905-911
    Published: September 30, 1982
    Released on J-STAGE: March 27, 2009
    JOURNAL FREE ACCESS
    It is a methodology developed to construct a preferential structure of a decision making group, by aggregating their preference upon majority rule on each pair-comparison. In so doing, those involved gain a deep understanding of the problem and critical discussion is stimulated. It is well known that majority rule does not satisfy transitivity. Extending the method by Bowman et al., the problem is formulated to find the nearest transitive structures from the result of majority rule. The problem is equivalent to a (0-1) programming problem to minimize the value of majority decision function. Using the Extended Lawler and Bell's method, it is solved through a computer. The solution is shown as hierarchical directed graphs.
    An example is presented, which is a structuring problem of research fields in environmental science in view of urgency. This methodology constructs transitive sturctures from opinions of nine experts. Through the example, the effectiveness of this method is explained.
    Download PDF (880K)
  • Ichiro KIMURA, Yoshio NAKAMURA
    1982 Volume 18 Issue 9 Pages 912-917
    Published: September 30, 1982
    Released on J-STAGE: March 27, 2009
    JOURNAL FREE ACCESS
    A new temperature-sensing actuator to operate at a fixed temperature has been developed.
    The device utilizes the magnetic transition at Curie temperature of ferrite. When the temperature of the ferrite is lower than the Curie point, the main magnetic flux from a permanent magnet passes through the ferrite. When the temperature comes up to the Curie point, the main flux passes through air gaps and a moving iron piece because of higher magnetic resistance of the ferrite. Thus the magnetic flux passage is switched to either of the two closed circuits in accordance with temperature.
    This device does not contain open air circuits in its main flux passage so that the magnetic disturbance is little even if ferromagnetic material exists close to the device.
    The actuating force upon the moving piece is magnetic so that the device needs no electric source nor wiring.
    In general, it is necessary to calculate permiance coefficients in order to design permanent magnet circuits because of nonlinearity of demagnetizing characteristic of the magnet.
    A rare-earth Co magnet which has linear demagnetizing characteristic is used to simplify the design formulas and at the same time to minimize the size of the device.
    The actuating force of a test device weighing about 25g was about 7N, and responce time was about 80∼4s for 1∼30°C over or under the set temperature.
    The device is expected to have high stability and reproducibility because of the stability of Curie temperature and seems to be useful as a wiring-free simple temperaturecontroller for energy-saving especially at extensive outdoor plants or at the place where explosive gas exists.
    Download PDF (718K)
  • Kamekichi SHIBA, Koichiro DOI
    1982 Volume 18 Issue 9 Pages 918-922
    Published: September 30, 1982
    Released on J-STAGE: March 27, 2009
    JOURNAL FREE ACCESS
    A two differential pressures type parallel mass flowmeter is composed of compound pipes in parallel with two branch pipe lines, a constant flow apparatus and two differential manometers. Both of the branch pipe lines have two straight portions.
    The liquid to be measured flows in the pipe line and then is divided into two branches without control. A flow of constant volume flow rate is sent from a position between two straight portions of one branch pipe line to the corresponding position of the other. The differences of pressures at the straight portions of both branch pipe lines, are measured by the differential manometers.
    Applying Bernoulli's theorem to the stream lines through both branch pipe lines, it is concluded that the difference of the two differential pressures is proportional to the mass flow rate. The error due to viscosity may be corrected by introduction of a discharge coefficient.
    The results of experiments with two trial mass flowmeters show that the mass flow rate can be obtained by a parallel mass flowmeter of two differential pressures type with a constant discharge coefficient.
    Download PDF (504K)
  • Kaoru NAKAGAWA, Ryoji OHBA, Kazumi MURATA
    1982 Volume 18 Issue 9 Pages 923-928
    Published: September 30, 1982
    Released on J-STAGE: March 27, 2009
    JOURNAL FREE ACCESS
    A new method is described for accurate and real time measurement of distance to an object. The method determines the distance by the triangulation, in which the parallax is obtained by peak position of the cross correlation between a pair of the object images. A normalized first order momentum of the correlation is used to estimate the position accurately. An experimental apparatus was constructed on the basis of the method with single lenses and photo-diode arrays. It was verified from experimental results that the apparatus could determine the angle of parallax with errors less than 10-4rad.
    Download PDF (807K)
  • Teruyuki IZUMI, Hiroshi KASHIWAGI
    1982 Volume 18 Issue 9 Pages 929-935
    Published: September 30, 1982
    Released on J-STAGE: March 27, 2009
    JOURNAL FREE ACCESS
    M-sequences generated by a feedback shift register are frequently used as a binary random number. When an N-th order M-sequence is generated repeatedly, it is necessary to initialize the state of an N-stage shift register so that the subsequence of the M-sequence has good random property for the length 2n(n<N).
    This paper first defines n-th order random sequences with the length 2n, and deduces the n-th order ideal random sequences whose subsequences for the length 2n, 2n-2, 2n-4, …, etc. become the random sequences of order n, (n-2), (n-4), …, respectively. Secondly, the initialization procedure of the high order M-sequence is investigated in order to approximate the M-sequence to an ideal random sequence as closely as possible. It is shown that [log2N]-th order ideal random sequences are suitable for the initial values of the N-th order M-sequence in order for the M-sequence to look like an ideal random sequence.
    Download PDF (885K)
  • Muneyuki YOSHIOKA, Kensuke KATAOKA, Yoshinobu MORIKAWA
    1982 Volume 18 Issue 9 Pages 936-941
    Published: September 30, 1982
    Released on J-STAGE: March 27, 2009
    JOURNAL FREE ACCESS
    In this paper, the stability of the annular jet proximity sensor composed of an annular nozzle, a transmission line and a load volume is theoretically investigated. The sensor system is formulated by using the transfer function of simple waves in the line and the reflection coefficients at the sending point and receiving end. On the basis of this formulation and Nyquist's stability criteria, the stability condition of the sensor system is obtained, and then the stability limits are calculated by using various immittance function models of system compornents. By comparing these stability limits with ready-published experimental results, it becomes clear that the agreement between the theoretical result based on Brown's line model and the experimental result is the best. Therefore, the stability map based on Brown's line model is shown by the use of non-dimentional system parameters.
    Download PDF (791K)
  • Hiroaki OZAKI, Akira MOHRI, Masaru TAKATA
    1982 Volume 18 Issue 9 Pages 942-949
    Published: September 30, 1982
    Released on J-STAGE: March 27, 2009
    JOURNAL FREE ACCESS
    This paper formalizes the problem of moving the hand of a manipulator to the goal position and orientation avoiding the collision with obstacles in the work space and gives an algorithm to solve the formalized problem. The algorithm is given as follows: First, an efficient method of detecting the interference among links of a manipulator and obstacles is presented. Second, movable conditions of joints are shown and the problem of planning of collision free movement is formalized using these conditions. Third, an algorithm to solve the problem by the use of linear programming is given.
    The obtained algorithm is applied to the planar movement of a manipulator with four links and four degrees of freedom and its effectiveness is shown.
    Download PDF (972K)
  • Tsutomu MITA, Masayuki MUKAIDA, Yoshinori SAKANAKA
    1982 Volume 18 Issue 9 Pages 950-951
    Published: September 30, 1982
    Released on J-STAGE: March 27, 2009
    JOURNAL FREE ACCESS
    Download PDF (255K)
  • 1982 Volume 18 Issue 9 Pages 953
    Published: 1982
    Released on J-STAGE: March 27, 2009
    JOURNAL FREE ACCESS
    Download PDF (65K)
feedback
Top