IEEJ Transactions on Industry Applications
Online ISSN : 1348-8163
Print ISSN : 0913-6339
ISSN-L : 0913-6339
Volume 111 , Issue 1
Showing 1-14 articles out of 14 articles from the selected issue
  • [in Japanese], [in Japanese], [in Japanese]
    1991 Volume 111 Issue 1 Pages 1
    Published: January 20, 1991
    Released: December 19, 2008
    JOURNALS FREE ACCESS
    Download PDF (104K)
  • Yoshinori Uesaka
    1991 Volume 111 Issue 1 Pages 2-9
    Published: January 20, 1991
    Released: December 19, 2008
    JOURNALS FREE ACCESS
    Download PDF (1128K)
  • Chihiro Fukui, Junzo Kawakami
    1991 Volume 111 Issue 1 Pages 10-19
    Published: January 20, 1991
    Released: December 19, 2008
    JOURNALS FREE ACCESS
    This paper presents a new application method of Hopfield-type neural networks for combinatorial optimization problems under inequality constraints.
    Since Hopfield showed that the neural network technique could solve combinatorial problems such as the Traveling Salesman Problem, many researchers have attemped to apply the neurals networks to various problems. However, the constraints of almost all of those problems are limited to equality constraints.
    To deal with the inequality constraints, we developed a new method in which we introduce special neurons that converge to intermediate values between 0 and 1. This method is applied to the switch pattern planning problem of power distribution systems. The stability analysis and simulation results show that the method is quite promising.
    Download PDF (1617K)
  • Ichiro Enbutsu, Kenji Baba, Harumi Matsuzaki, Mikio Yoda
    1991 Volume 111 Issue 1 Pages 20-28
    Published: January 20, 1991
    Released: December 19, 2008
    JOURNALS FREE ACCESS
    In this paper, an enhanced method for rule extraction from a multilayered neural network is proposed. Conventionally a learned network is treated as a black-box model, so internal distributed knowledge in it can not be indicated in explicit form. The authors analyzed the mathematical properties of a neural network, and based on the properties, defined an index to evaluate a weights matrix. By using this index, the causal relationships between each input and output are interpreted as production rules. Furthermore, a modified error backpropagating algorithm which can control excessive redunduncy in a network is proposed to enhance learning ability. This algorithm was applied to the operation support of coagulant injection in a water purification plant. The computer simulation showed advanced reliability regarding extracted rules and association.
    Download PDF (1313K)
  • Kimiaki Nakano, Kazuhiko Fukutani, Satoshi Seno, Yoshiyuki Shirakawa, ...
    1991 Volume 111 Issue 1 Pages 29-35
    Published: January 20, 1991
    Released: December 19, 2008
    JOURNALS FREE ACCESS
    The steel industry is very interested in the high accurate inspection technique of cold rolling strip because of requirement of high quality products and high speed production.
    Although a laser reflecting inspection system is generally used, the accuracy of inspection is not sufficient yet.
    We have developed new approach for a inspection method used advanced neural network technology joined with a conventional laser inspection equipment.
    In this paper, we show our proposed method, the performance through experiment with sample defects data and the comparison between proposed method and two conventional methods.
    Download PDF (970K)
  • Hiroshi Takenaga, Shigeo Abe, Masao Takatoo, Masahiro Kayama, Tadaaki ...
    1991 Volume 111 Issue 1 Pages 36-44
    Published: January 20, 1991
    Released: December 19, 2008
    JOURNALS FREE ACCESS
    A major problem in applying a multi-layered neural network to image processing is how to select input components from among a number of features or pixels extracted from a recognized object. This is especially important for achieving both high speed processing and a high recognition rate.
    This paper proposes a method of optimizing the number of input components, i. e. that of input neurons without lowering a recognition rate. In general, neural networks learn a recognition algorithm from input patterns and their desired output patterns. But if some input components are redundant, namely they are expressed by other input components or they do not contribute to recognition, their effect on the recognition algorithm, or their sensitivity to outputs is considered to be very small. Therefore, by analyzing input-output sensitivity of neural networks, redundant input components can be deleted. Repeatedly applying this method, the useful input components for recognition can be selected. For an experimental neural network for recognition of numerals with 12 feature input components, the input components could be reduced to eight without lowering a recognition rate.
    Download PDF (1708K)
  • Atsusi Iga, Akira Takechi, Mitsugu Aihara, Yasutaka Fujiwara, Jiro Kaw ...
    1991 Volume 111 Issue 1 Pages 45-54
    Published: January 20, 1991
    Released: December 19, 2008
    JOURNALS FREE ACCESS
    As a diagnostic apparatus of deterioration of 6kV class XLPE cable on the hot line, a practical apparatus has been developed, which detects the voltage signal from a potential divider contacted with a high voltage live part and the charging current from a grounding conductor current and displays tan δ and electrostatic capacity. The authors have investigated the various methods, which are adopted to the state of cable line, in order to find the most important input procedure of the voltage signal for the application of this apparatus in the work site operation. We have proposed a method of using the secondary voltage of the GPT (ground potential transformer), which are installed especially at the many places such as substations and the like, and have developed and proved the method. We can expect that it can be applied not only for the 6kV class XLPE cable but also in the broad areas such as the continuous monitoring system for the cable and electrical equipments because it has outstanding properties especially in safety, operability and so on.
    The following items are discussed in this paper:
    (1) development of the new method/program, which can calculate the phase difference (phase angle) between the primary and secondary voltage signals of GPT, (2) method/equipment for measuring the burden of GPT, which is required for calculation of the phase angle, and the measuring results, and (3) the proven results in the field.
    Download PDF (2564K)
  • Eikichi Taira, Hayao Miyagi, Katsumi Yamashita
    1991 Volume 111 Issue 1 Pages 55-62
    Published: January 20, 1991
    Released: December 19, 2008
    JOURNALS FREE ACCESS
    The power system stability problem has been one of the major subjects concerning power system engineering, and is becoming much more significant today in accordance with increasing size and complexity of interconnected power systems. In this problem, it is well known that the direct method of Lyapunov is the most suitable method of on-line transient stability estimation for power systems, and several different techniques have been proposed for the construction of Lyapunov functions up to this time. However, the effectiveness of these Lyapunov functions for power systems have been demonstrated using only digital computer simulations.
    The purpose of this paper is to construct microprocessor-based on-line transient stability estimation system using Lyapunov function constructed by the Lagrange-Charpit method and estimate the transient stability of a microalternator connected to a large system. The experimental results show that this system can act satisfactorily for estimating the transient stability of the power system.
    Download PDF (2040K)
  • Satoshi Ogasawara, Tomoyuki Joichi, Hirofumi Akagi
    1991 Volume 111 Issue 1 Pages 63-69
    Published: January 20, 1991
    Released: December 19, 2008
    JOURNALS FREE ACCESS
    In recent years, attention has been paid to direct-drive for industrial robots which are essential to factory automation. Direct-drive robots, which are directly actuated with no gear reducer, have no backlash, low friction and good mechanical stiffness. This results in the excellent repeatability of positioning, that is, an order of magnitude better than that of gear-drive robots. It is, however, more prominent in the design of control system for direct-drive robots that the arm inertia variation depending on the arm configuration and the Coriolis and centrifugal torques are directly reflected to the motor axes due to the direct coupling.
    In this paper, a direct-drive robot with a two degree-of-freedom is developed for planer assembly tasks. The motion of the manipulator arm is constrained within a horizontal plane, so that this type of robot is referred to as a SCARA (Selective Compliance Assembly Robot Arm)-type robot. A balanced manipulator arm, the purpose of which is to cancel the inertia variation and nonlinear effects, is introduced to the direct-drive robot, thus leading to improved control performance as well as simple manipulator dynamics.
    In addition, the combination of feedforward compensation and a new approach to positioning control is proposed to perform stable and fast positioning without overshooting.
    Download PDF (1612K)
  • Masahiko Kamatani, Iwao Shibata
    1991 Volume 111 Issue 1 Pages 70-76
    Published: January 20, 1991
    Released: December 19, 2008
    JOURNALS FREE ACCESS
    Characteristics analysis using the matrix method of stator-fed AC commutator motor are presented. Recently, many variable-speed AC motor drive systems with semiconductor switching devices have been developed. However, the power supply systems suffer from higher harmonic currents during switching. On the contrary, three-phase AC commutator motors have essentially no detrimental influence on the power supply systems. Consequently, the motors are being reconsidered for use in spite the fact that they need periodic maintenance for the replacement of commutators and brushes.
    So far, motor analysis was calculated using various vector methods: in this case complicated calculations are required, with accurate solutions difficult to be obtained. We derive the fundamental equations using the matrix method which can be applied to the transient phenomenon. Also, we obtained an agreement between calculations and experimental results of the steady-state characteristics.
    Download PDF (1096K)
  • Toshikatsu Sonoda, Ryuzo Ueda
    1991 Volume 111 Issue 1 Pages 77-84
    Published: January 20, 1991
    Released: December 19, 2008
    JOURNALS FREE ACCESS
    Vector control in induction motor drive system gives a theoretical foundation for realizing high accurate torque control. However, the control accuracy in practice depends on the performances of voltage and current sensors used for the purpose. Particularly, the current sensor is required to satisfy the following: (i) It is non-contacting type; (ii) The detectable frequency range covers from DC to the one beyond over the inverter chopping frequency; (iii) The current range over from mA to several 10 A has to be covered with the accuracy of 0.1% in FS (Full Scale); (iv) The same detection characteristics is easily obtainable for the application to each phase in three phases; (v) The effect of the environmental temperature variation on the detection characteristics is small; (vi) The sensor is little affected by the external magnetic noise and surge voltage.
    This paper presents a current sensor which has a possibility of satisfying almost all the above items by using two magnetic cores.
    Download PDF (3696K)
  • Tatsuo Horii, Genjiro Wakui
    1991 Volume 111 Issue 1 Pages 85-93
    Published: January 20, 1991
    Released: December 19, 2008
    JOURNALS FREE ACCESS
    We will introduce a basic analysis on the harmonic quantities of AC motor, flux density, m. m. f., e. m. f., and etc.. The quantities which degrees of space and time alternation are equal to those of fundamental currents on positive-phase-sequence and fundamental fields excited by its currents, are called fundamental harmonics. And the quantities to be not equal to degree of space alternation or time alternation are called higher harmonics.
    At first, the actual conductor arrangement is made a relation to sum of the density distribution of winding for harmonics including a fundamental. After deriving the relationship equation of the harmonic m. m. f. and the flux density on the gap region, and the equation of harmonic flux density and e. m. f. on the density distribution of winding, we have the general expression of equivalent impedance on any phase sequence for AC motors.
    Download PDF (2898K)
  • Tatsuo Horii, Genjiro Wakui
    1991 Volume 111 Issue 1 Pages 94-100
    Published: January 20, 1991
    Released: December 19, 2008
    JOURNALS FREE ACCESS
    In the latest paper, we have introduced the general expression of equivalent impedance on any phase sequence deducing from the basic analysis on the harmonic quantities for AC motors. And the equivalent impedance is divided into four parts, a gap impedance, a stator impedance, a rotor impedance, and an induced current impedance. In this paper, we show the equivalent circuits of AC motors, a wound-rotor or squirrel-cage rotor induction motor, a solid rotor induction motor, a salient-pole reluctance motor, an interior slit rotor reluctance motor, a salient-pole synchronous motor and a hysteresis motor.
    Download PDF (1400K)
  • Hiroshi Yamaguchi, Yukihiko Sato, Teruo Kataoka
    1991 Volume 111 Issue 1 Pages 101-102
    Published: January 20, 1991
    Released: December 19, 2008
    JOURNALS FREE ACCESS
    Download PDF (717K)
feedback
Top