Transactions of the Society of Instrument and Control Engineers
Online ISSN : 1883-8189
Print ISSN : 0453-4654
ISSN-L : 0453-4654
Volume 4, Issue 1
Displaying 1-14 of 14 articles from this issue
  • Kunihiko ICHIKAWA, Katsutoshi TAMURA, Katsuhiko HIDA
    1968 Volume 4 Issue 1 Pages 1-7
    Published: March 30, 1968
    Released on J-STAGE: March 27, 2009
    JOURNAL FREE ACCESS
    For a bounded state variable problem, the optimal condition is shown by Pontryagin, but it seems to be pretty troublesome to solve the actual problems through its direct application.
    It is shown in this paper that the bounded state variable problem can be completely solved through the conventional maximum principle after divided into 3 sub-problems A, B and C. In problem B which is to be solved under the restricted condition that phase point runs on the boundary, the dimension of problem B decreases by one, so that we can apply the conventional maximum principle to the degenerated problem.
    Each problem A, B and C is not solved independently, but some continuity condition must be satisfied at the continuity point. The continuity conditions between problem A and B and between B and C were found by the authors. Pursuit of optimal backward time trajectory must be continued until the continuity condition is not satisfied. When the pursuit process is completed, we can obtain optimal control law.
    This method is not only simple but also has clear physical meaning and does not need any other principle. Two illustrative examples are explained in the last part of this paper.
    Download PDF (895K)
  • Norimasa NOMURA
    1968 Volume 4 Issue 1 Pages 8-14
    Published: March 30, 1968
    Released on J-STAGE: March 27, 2009
    JOURNAL FREE ACCESS
    In optimal control problems, it is known that not all the states of a system can be transfered to the desired final state by an admissible control. This paper presents a new method for calculating a controllable region which is defined as a set of the states that can be transfered to the desired state in a finite time by an admissible control. For this purpose, a certain function, which was originally used for computing time-optimal controls, is introduced, and the boundary of the controllable region can be obtained by evaluating the limit of this function as time t tends to infinity. As an example, a second order divergent system with complex eigenvalues is treated.
    Download PDF (965K)
  • Suteo TUTUMI
    1968 Volume 4 Issue 1 Pages 15-21
    Published: March 30, 1968
    Released on J-STAGE: March 27, 2009
    JOURNAL FREE ACCESS
    This paper is devoted to the analysis of a new optical tracking system with two mirrors rotating reversely to each other and an amplitude-modulated reticle, which are used for separation of azimuth and elevation error signals as well as to improve the transmitted-signal-to-back-ground-noise ratio. The tracking system presented here has no need of a reference wave used to produce the error signals, has no dead zones in the field of view of its optical system and has a narrow bandwidth. Experimental results are shown.
    Download PDF (2297K)
  • Akira NAGATA, Tsukasa UEDA, Takuya NISHIMURA
    1968 Volume 4 Issue 1 Pages 22-26
    Published: March 30, 1968
    Released on J-STAGE: March 27, 2009
    JOURNAL FREE ACCESS
    This paper is concerned with fundamental problems arising in the control of a linear sampled-data system with time-delay. An important and unsolved problem of such a system is the problem of finite time settling; namely, the problem of transferring the state of the system from an arbitrary initial state to an equilibrium state in a finite number of steps.
    In this paper, the problem is solved by transforming the given system into a higher order sampled-data system without time-delay. The state transition matrix of the derived sampled-data system may be nonsingular or singular, depending on the structure of the original system.
    In both cases, the problems of controllability and minimum time control are considered in detail. Finally, a numerical example is given.
    Download PDF (721K)
  • Minoru OKABAYASHI, Takuso SATO
    1968 Volume 4 Issue 1 Pages 27-31
    Published: March 30, 1968
    Released on J-STAGE: March 27, 2009
    JOURNAL FREE ACCESS
    In the analysis of control systems, it is often required to know the statistical properties of the response of the systems to random input signals. Conventional methods depend upon the utilization of the convolution integrals and mean value and auto-correlation function of the input signals. The results gotten by these methods are time averages of whole responses. But when the statistical properties at specified time, for instance, at the time of changes of the input signal are required, these methods are not effective.
    In this paper the method of stochastic matrix is introduced and a method of analysis for linear systems subjected to partially independent input signals is presented. This method is useful for the analysis of relay systems and other systems with partially independent inputs.
    Firstly the general form of the method is presented and as an example the responses for semi-Markovian inputs are analysed theoretically and experimentally. The results verify the usefulness of this method.
    Download PDF (487K)
  • Minoru OKABAYASHI, Takahisa HASHIMOTO, Takuso SATO
    1968 Volume 4 Issue 1 Pages 32-37
    Published: March 30, 1968
    Released on J-STAGE: March 27, 2009
    JOURNAL FREE ACCESS
    To solve stochastic problems which can be reduced to semi-Markov or compound Markov processes a simple electronic simulator is constructed. This device consists of a magnetic drum memory, two random number generators and logical circuits.
    The probability of each state of the process can be given with the accuracy of 1/64. The maximum numbers of the states of the semi-Markov process and compound Markov process are 32 and 8 respectively and the transitions of the state are done at the interval of 1/166 sec.
    As the output of the device both analogue and digital voltages are available.
    As examples of applications of the device, problems of clinical trials and maintenance of machines are shown.
    The procedures of the model construction and simulation of these problems are presented. The results of the simulation show the effectiveness of the device.
    Download PDF (1265K)
  • Shigeyuki HOSOE, Masami ITO
    1968 Volume 4 Issue 1 Pages 38-45
    Published: March 30, 1968
    Released on J-STAGE: March 27, 2009
    JOURNAL FREE ACCESS
    In a D.D.C. system, the amplitude quantization of the input, the internal, and the output signal of the digital processor is done. Usually the relatively gross quantization for the output signal is done, because it is greatly important to take account of the economical significance for the output devices, and because the quantization for the output signal does not affect the performance of the system so much as the others.
    In this paper we have discussed the method for obtaining the discrete describing function of the quantizer and analysing the self-sustained oscilations caused by the gross quantization at the output device of the digital processor. In the reference (2) the method has been discussed for analysing the self-sustained oscilations in the system shown in Fig. 1 by representing [sampler+quantizer+hold] in conventional continuous describing function.
    But with his method, it is very difficult to analyse the systems with digital compensation. Using the method obtained in our paper, it is easy to analyse the systems with digital compensation, and the systems with more than one syncronized samplers.
    Download PDF (865K)
  • Yoshifumi OKUYAMA
    1968 Volume 4 Issue 1 Pages 46-53
    Published: March 30, 1968
    Released on J-STAGE: March 27, 2009
    JOURNAL FREE ACCESS
    In the theory of sensitivity analysis in control systems, sensitivity measure is usually defined as a type of differential coefficient. However, such a measure represents only a first-order approximation of the quantity under interest corresponding to “small” parameter (or system) variation. The extent of the variations within which the measure is valid is not clearly defined: in addition, it is not much effective for the case of time-varying parameter.
    In this paper each variable of generalized linear systems is investigated in “normed space” and sensitivity measure is defined as norm of linear operator (sensitivity operator). Therefore, the influence of parameter variations is quantitatively evaluated and the sensitivity analysis of linear systems including the case of time-varying parameter is discussed at large. That is also investigated in some definite spaces. When the sensitivity measure is bounded, the perturbed system is stable. Hence, several conditions for the stability of perturbed linear systems are derived. Finally the case of frozen parameter (or slowly varying parameter) is discussed in detail.
    Download PDF (917K)
  • Kimio YAMAGUCHI
    1968 Volume 4 Issue 1 Pages 54-60
    Published: March 30, 1968
    Released on J-STAGE: March 27, 2009
    JOURNAL FREE ACCESS
    A time optimal control function cannot be obtained analytically when the order of the system becomes equal to or higer than four. In the recent years various iterative procedures have been developed for computing time optimal controls.
    This paper presents an iterative procedure for computing suboptimal controls. This procedure is an extension of a pulse perturbation method developed by the author and may be applied in analysing an nth order system described by the vector defferential equations x=Ax+Bu, where A is the n×n constant matrix with complex eigenvalues, B the nth order vector and u(t), a scalar, is a control function with constraint |u(t)|≤1.
    The principle of this method is as follows. Suppose that a control u(t) takes only the values of +1 or -1 and that the control time is ts, that is u(t)≡0, for t>ts. Pulse perturbations are added to u(t) at every switching time in such a way that C(ts)(=-∫ts0Φ-1(τ)Bu(τ)dτ) arrives at x0, where Φ(t) is the transition matrix for the system, x0 the initial state of x(t) and the final state is zero. These pulses are then approximated by rectangular waves sb that the perturbed control takes the form of a Bang-Bang control. The suboptimal control can be obtained if the iteration is continued until C(ts) becomes equal to x0.
    Download PDF (745K)
  • Tohru IDOGAWA
    1968 Volume 4 Issue 1 Pages 61-68
    Published: March 30, 1968
    Released on J-STAGE: March 27, 2009
    JOURNAL FREE ACCESS
    A new method, similar to the stroboscopic observation, for performing real-time auto-and cross-correlation of periodic signals is presented. Waveforms of auto-and cross-correlation functions are given in a magnified scale of time, by beat waveforms of two pertinent periodic signals. The method can be applied to signals expressed by periodic functions which are described as Fourier series. As an example, impulse responses of number of known electronic filters were recorded by a pen writing oscillograph using signals of m-sequences as the periodic test-signals in the stroboscopic correlation method. From these results, errors were estimated which resulted from the shape of the autocorrelation functions of signals of m-sequences. The method was proved to be useful for determining impulse responses of linear systems.
    Download PDF (2027K)
  • Tohru IDOGAWA
    1968 Volume 4 Issue 1 Pages 69-76
    Published: March 30, 1968
    Released on J-STAGE: March 27, 2009
    JOURNAL FREE ACCESS
    Experiments for determining sound velocity, using a class of pseudorandom sequences, known as m-sequences, are described. The m-sequence signal generated by a shift register are applied to a sonic system formed by a speaker and a condencer microphone. On the surface of the microphone the phase of the sonic signal is controlled constant by varying the shift pulse frequency which gives the sound velocity. Detection of the phase was carried out on real-time by 1) ordinary input-output crosscorrelation and 2) stroboscopic correlation method. The first method affords a simplified measuring system, while the second method enables stable and accurate measurement compared with the first one, though it requires two m-sequence generators. Continuous record of pendulum motion, room temperature and wind velocity obtained by the second method are also presented. The accuracy of the room temperature determination was estimated to 0.005m/s in terms of sound velocity change.
    Download PDF (3110K)
  • Ichiro KIMURA, Jin-ichi SAKURAI, Hideo KONDO
    1968 Volume 4 Issue 1 Pages 77-82
    Published: March 30, 1968
    Released on J-STAGE: March 27, 2009
    JOURNAL FREE ACCESS
    In this paper, the characteristic features and analysis of the new type electronic differential pressure transmitter developed by the authors are described.
    Differential pressure transmitters are used for measurement and control of process variables such as flow rate, pressure or liquid level. The mechanical part of the transmitter described here consists solely of a diaphragm assembly and a cantilever on which strain gages are bonded in a four active arm bridge.
    It has no movable pressureproof seals such as sealing pipes or sealing diaphragms, and the influence of line pressure is 0.2% FS of the minimum span at 100kg/cm2.
    The transmitter has no other springs and levers, so it can be designed with such a high natural frequency as 130c/s.
    The dynamic characteristics of differential pressure transmitters have been designed by empirical methods. In this paper, a method of calculation of static and dynamic characteristics of the transmitter is described, and the experimental results show the usefulness of this designing method.
    The break point frequency of this transmitter is adjustable in as wide range as 1:100 turning the adjusting screw 8 turns by adoption of a variable damper of a novel design utilizing tapered spiral groove. The response setting is not too sensitive and can be made merely by turning the adjusting screw around a certain number of turns without testing the frequency characteristics each time.
    Download PDF (1497K)
  • Setsuso SAGARA, Tsuyoshi OKITA
    1968 Volume 4 Issue 1 Pages 83-88
    Published: March 30, 1968
    Released on J-STAGE: March 27, 2009
    JOURNAL FREE ACCESS
    In this paper, the parameters of transfer function are derived from time moments of its impulse response by use of the derivation in the s-plane and moments of impulse response is evaluated from input moments and output moments.
    In this method, the transfer function is not expanded in the s-power series but determined in the actual form and the parameters of the transfer function are estimated by using the higher order moments of the input and the output as same as the order of the transfer function from the lower order terms. Futhermore, the method can be easily extended to the random process by means of the auto-and cross-correlation functions of the process input and the output signal.
    It is not desirable to use the nonlinear equipments such as multipliers and t-power generators to measure the time moments, since the accuracy of this method depends upon how accurately they may be measured. The time moments, however, can be measured precisely by the simple operators such as the integrator, the adder and the comparator, since they are transformed to the multiintegration with fixed measuring time, by using the integration by part.
    It is found from the experimental results that this method is available not only for off-line identification, but also the on-line identification in the adaptive control system, because the transfer function can be estimated by the relatively simple operation with sufficient accuracy.
    Download PDF (813K)
  • Norihiko MORINAGA, Hisashi OSAWA, Toshihiko NAMEKAWA, Kenji AOYAGI
    1968 Volume 4 Issue 1 Pages 89-97
    Published: March 30, 1968
    Released on J-STAGE: March 27, 2009
    JOURNAL FREE ACCESS
    In terms of FM correlation system, we mean an analog correlation system which consists of two channels, containing a FM detector in each channel. This paper presents a detailed analysis of the output for FM correlation system when two input signals to be correlated have the arbitrary time difference and magnitude, and are superimposed by the statistically independent stationary Gaussian random noises. First, the expressions are derived for the output of FM correlation system when each channel input is a frequency-modulated signal with constant amplitude superimposed by random noise, and the features of FM correlation system are shown in comparison with the conventional correlation system. Second, we analize the output of FM correlation system for the following two cases: namely, the one in which each channel input consists of an amplitude-limited narrowband Gaussian random signal XL1(t) (XL2(t)) and random noise, and the other case in which each channel input consists of a no-limited one X1(t) (X2(t)) and random noise, and show the difference in the effect of input carrier-to-noise ratio for the output of FM correlation system for the two cases. In addition, it is shown by σ, the rms band-width of X(t), that we are able to regard the band-width as ±(3∼4) σ, for XL(t) and X(t) under the condition of equal mean power.
    Download PDF (1249K)
feedback
Top