Automatic Control
Online ISSN : 1883-8162
Print ISSN : 0447-7235
ISSN-L : 0447-7235
Volume 8, Issue 3
Displaying 1-9 of 9 articles from this issue
  • Yasundo TAKAHASHI
    1961 Volume 8 Issue 3 Pages 146-151
    Published: June 15, 1961
    Released on J-STAGE: October 29, 2010
    JOURNAL FREE ACCESS
    Continuous pröcess identification is becoming a crucial step in some of the future adaptive-and self-optimization controls. Use of an adjustable model or a set of models to identify partly unknown process dynamics is investigated in the paper. Any exact simulation of a physical process is generally not possible nor necessary. The closeness of a model to a process is evaluated by means of an error criterion function. The criterion function suggested in the paper is based upon the difference between the response of the process and a model under a normal operating signal. This not only simplifies the implementation, but also provides a direct method to find the best fit model for the purpose of self-optimization. For example, a best fit model determined for a white noise input may not be the desired best approximation of the process if the process is working in a colored noise pattern. Experimental results for open-and closed-loop arrangements demonstrates the importance of the noise pattern in selecting a best fit model. Time lags involved in process identification is another factor of importance in self-optimization control systems. The approach used in the experiments indicates a shorter time lag compared to a rigorous, statistical method suggested in other papers.
    Download PDF (422K)
  • Yo IKEBE, Kozo MATSUSHIMA, Yasuhiro HASEGAWA
    1961 Volume 8 Issue 3 Pages 152-157
    Published: June 15, 1961
    Released on J-STAGE: October 29, 2010
    JOURNAL FREE ACCESS
    Dr. Y. Ikebe devised a mechanical frequency response tester which makes use of the polar coor dinate. This was called the Rotary Disc Type. He and Mr. K. Nakano improved this original type and devised the second electro-mechanical tester, which was called the Fixed Disc type. In this type the null-method of measurement is employed.
    Now the authors developed the third tester called the Stroboscopic Type which is further advanced from the preceding two types. This paper describes this new type and presents a comparison with the former two types.
    The potentiometer multiplier of this new stroboscopic tester is greatly improved as compared with that of the fixed disc type tester. The maximum possible frequency bandwidth is now from 0 to 100c/s per second whereas it was from zero to several cycles with the fixed disc type.
    The frequency response testers using the polar coordinate have several advantages such as simple design, ease in maintenance and handling and portability. These advantages are all reserved in this new type.
    As an example of the use of the new type tester, the frequency response from 10c/s to 80c/s was taken wtih a simple R-C low-pass filter, and the result is reported.
    Download PDF (1809K)
  • Masami ITO, Tatsuki NORIMATSU
    1961 Volume 8 Issue 3 Pages 158-168
    Published: June 15, 1961
    Released on J-STAGE: October 29, 2010
    JOURNAL FREE ACCESS
    The authors have investigated, using a digital computer, into the influences of torque saturation, static and coulomb friction and feedback compensation network on a servo system as shown in the figure given below, and have obtained the results which can be used as the fundamental data for the design of the servomechanisms of this type.
    (1) The temporary stop phenomena occur for the step input, owing to the filter H (s) which is inserted to reduce the steady-state speed errors.(2) The occurrence regions of jerking motion which is experienced for the ramp input are expanded owing to the filter H (s).(3) There are no direct interaction between the torque saturation and the friction.(4) In servo system which have the large coulomb friction, the steady-state errors are large, and the maximum higher acceleration, the maximum acceleration and the maximum velocity decrease, and the temporary stop phenomena and the jerking motion are apt to occur.(5) It is not advisable to design a servo system in which the friction torque is fairly large compared with the maximum torque. It is considered to be impossible to specify quantitatively the permissible friction torque, because the spesification for the servo system differ according to circumstances.(6) In the case where the torque saturation and the nonlinear friction exist, it is necessary to use compensation parameters K0 and T0, whose values are larger than those in the case of linear system, in order to obtain the good dynamic performance of the system for both step and ramp inputs. The values of the compensation parameters which are used to obtain the best dynamic performance of this system for the step input, are larger than the values of the parameters for the ramp input.
    Download PDF (1103K)
  • [in Japanese]
    1961 Volume 8 Issue 3 Pages 169-175
    Published: June 15, 1961
    Released on J-STAGE: October 29, 2010
    JOURNAL FREE ACCESS
    Download PDF (504K)
  • [in Japanese]
    1961 Volume 8 Issue 3 Pages 176-184
    Published: June 15, 1961
    Released on J-STAGE: October 29, 2010
    JOURNAL FREE ACCESS
    Download PDF (2944K)
  • [in Japanese]
    1961 Volume 8 Issue 3 Pages 185-193
    Published: June 15, 1961
    Released on J-STAGE: October 29, 2010
    JOURNAL FREE ACCESS
    Download PDF (1570K)
  • [in Japanese], [in Japanese], [in Japanese]
    1961 Volume 8 Issue 3 Pages 194-197
    Published: June 15, 1961
    Released on J-STAGE: October 29, 2010
    JOURNAL FREE ACCESS
    Download PDF (1117K)
  • [in Japanese]
    1961 Volume 8 Issue 3 Pages 198-202
    Published: June 15, 1961
    Released on J-STAGE: October 29, 2010
    JOURNAL FREE ACCESS
    Download PDF (973K)
  • [in Japanese]
    1961 Volume 8 Issue 3 Pages 203-205
    Published: June 15, 1961
    Released on J-STAGE: October 29, 2010
    JOURNAL FREE ACCESS
    Download PDF (1472K)
feedback
Top