All issues
Volume 13 (1977)
- Issue 6 Pages 533-
- Issue 5 Pages 425-
- Issue 4 Pages 313-
- Issue 3 Pages 209-
- Issue 2 Pages 105-
- Issue 1 Pages 1-
Volume 13, Issue 5
Displaying 1-16 of 16 articles from this issue
- |<
- <
- 1
- >
- >|
-
Stabilization by Moving Weight HorizontallySetsuya HAYASHI, Hideaki KANOH, Masami MASUBUCHI1977 Volume 13 Issue 5 Pages 425-432
Published: October 30, 1977
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper presents a stabilization control of inherently unstable systems in their nature. An inverted pendulum which is a typical, unstable mechanical system, is considered and stabilized by moving a weight horizontally through a feedback control. A stabilization problem in this system coresponds to that in the postural control of a man.
It is shown that, so far as the dry friction at the bearings etc. is negligible, this inverted pendulum is stabilizable when the linearized system is controlable and observable. At the experiment, the output, the weight position r and the pendulum angle θ, and the estimated state (r, θ) by a minimum order observer or an approximate differentiator are fed back. Experimental results show that stabilization was posible in spite of the existence of the dry friction at the bearings etc..View full abstractDownload PDF (1680K) -
Takashi SUZUKI, Seiji TAKASHIMA1977 Volume 13 Issue 5 Pages 433-438
Published: October 30, 1977
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper presents a scheme for designing a hyperstable discrete model reference adaptive control system, by using the augmented error signal concept proposed by Monopoli.
In our scheme, the error equation between the reference model and the plant outputs is transformed into an equivalent feedback system composed of a first order linear part and a nonlinear time varying part, by introducing an augmented error signal. The hyperstability of the adaptive control system can be assured by letting the transfer function of the linear part be strictly positive real and controlling the adjustable parameters according to an adaptive law. The condition for strictly positive real transfer function can be easily satisfied since the transfer function is of a first order in our scheme. The adaptive law can be implemented by solving a third order algebraic equation of the augmented error signal at each sampling instant. The control input to the plant is adaptively synthesized by employing the adjustable parameters and available signals in the system. The usefulness of the proposed scheme is demonstrated by the computer simulation carried out for a second order plant.View full abstractDownload PDF (311K) -
Juichi MIYAMICHI1977 Volume 13 Issue 5 Pages 439-444
Published: October 30, 1977
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSDiscrete control is widely found in various field, including the control by digital computers. Systems whose control inputs are restricted to discrete values are called the quantized control systems.
A state x is called reachable if there is an input sequence which carries the system from zero to x. In quantized control system, the best situation for reachability is that within any neighborhood of x there is a reachable state, in which case x is called to be “almost reachable”.
In this paper, a structure analysis of the state space is developed for the linear, constant and discrete-time dynamical systems over the field of real numbers whose control inputs are restricted to the integer vectors:
x(k+1)=Ax(k)+Bu(k)
where x(k)∈Rn, u(k)∈Zm and A and B are real matrices.
Concerning with reachability, three A-invariant subspaces (X2+X3+X4, X3+X4 and X4 below) are characterized, resulting the following canonical form of the quantized control system.
X2+X3+X4: A-invariant subspace spanned by the set of reachable states.
X3+X4: A-invariant subspace in which every state is almost reachable.
X4: A-invariant subspace in which every state is almost reachable in finite time.
A2, B2: matrices over the ring of all integers.
A32, A3, B3: matrices over the field of rational numbers, A2 has no algebraic integers as its characteristic values.
A4, B4: matrices over the field of real numbers.View full abstractDownload PDF (816K) -
Nobuhide SUDA, Akio ENOKI1977 Volume 13 Issue 5 Pages 445-450
Published: October 30, 1977
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn the present paper is established the theoretical basis for augmentation of bond graphs, that is for assignment of strokes and arrows on bonds. We give a precise definition of “adequateness” of bond graphs and then obtain a necessary and sufficient condition, as well as a simple algorithm, for the adequate augmentation.View full abstractDownload PDF (383K) -
Sadaaki MIYAMOTO, Saburo IKEDA, Yoshikazu SAWARAGI1977 Volume 13 Issue 5 Pages 451-456
Published: October 30, 1977
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThe aim of the present paper is to establish a few theorems for a class of optimization problems of a bilinear distributed parameter system. The word “bilinear system” means that it is a linear system and some coefficients of its system equation are control variables. A system described by a parabolic partial differential equation is considered here. The cost functional to be optimized takes a quadratic form of the state variable. First established is a theorem of Fréchet differentiability of the state variable with respect to the coefficients. Then the existence of gradient of the cost functional is easily proved. Introducing the adjoint system, a representation of the gradient is obtained. Further, on the assumption of the existence of an optimal solution, the maximum principle is proved by making use of the variational inequality. The representation of the gradient will be useful in application of certain algorithms such as the method of steepest descent.View full abstractDownload PDF (675K) -
Nariyasu MINAMIDE1977 Volume 13 Issue 5 Pages 457-462
Published: October 30, 1977
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper deals with a problem of determining optimal periodic test input signals for identification of continuous linear systems under the assumptions that the energy of the periodic input signals is bounded above, that the observed outputs are contaminated with additive white Gaussian noises, and that apriori probability distribution of the unknown system characteristics is white Gaussian and is independent of the observation noises.
By deriving Radon-Nykodim derivatives of the relevant probability measures, the amount of the mutual information between the observed outputs and the unknown system characteristics is calculated in terms of the Fourier coefficients of the test input signals. It is shown that the amount is upper bounded by the energy of the input signals and depends solely upon the auto-correlation of each component of input signals. It is also shown that the bound is really the least upper bound, and that there exist no test input signals that achieve the bound.
A particular consideration is given to the M-sequence signals. It is shown that, as the pulse interval of the M-sequence signals decreases, the corresponding amount of the mutual information tends to the possible upper bound under amplitude as well as energy constraint.View full abstractDownload PDF (725K) -
Takeo OJIKA, Yoshikazu NISHIKAWA, Hiroaki SHIMAZUTSU, Masashi OKUDAIRA1977 Volume 13 Issue 5 Pages 463-469
Published: October 30, 1977
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper proposes a computational algorithm for the optimal control of coupled nonlinear systems. The dynamical system under consideration is composed of several subsystems with relatively strong interconnections and/or relatively strong nonlinearities. The objective is to minimize a performance index of quadratic type.
The basic idea of the present algorithm is to remove the interconnections and nonlinearities from system variables, and then the overall problem is decomposed into a number of smaller and linear subproblems. At the same time, the performance index is modified by using the interaction variables. Parameters, called weights, are introduced in the modified performance index. The values of these parameters have significant influence on the rate of convergence of the algorithm, and they are major factors determining the total computing time necessary to reach a converged result.
On the other hand, the interaction variables are adjusted directly by a gradient algorithm until the interaction balances are satisfied.
In the paper, a sufficient condition for convergence of the iterative algorithm is discussed in detail, and several features of the present algorithm are illustrated by examining an example.View full abstractDownload PDF (379K) -
Tetsuhiro SUMIMOTO, Saburo OKADA1977 Volume 13 Issue 5 Pages 470-474
Published: October 30, 1977
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper deals with a method for analysis of tidal flow patterns in the Seto Inland Sea hydraulic model. Usually the patterns are analyzed through the observation of the drift of floats on the surface of the water. Observations are made in respect of a number of floats, over a wide area for long periods. One experiment yields numerous data. The analysis of such data by the method hereto requires very much time and is laborious.
A number of drifting floats are photographed at intervals of a predescribed time with a camera from above. Then the images in a roll of developed film are projected by a projector one after another with proper optical enlargement onto a screen from behind. The positions of the individual floats are read by the coordinates reading device, and punched out in the form of time-sequence data by a puncher. The paper tape is led into a minicomputer, which analyzes the data and issues to an automatic drafting machine a command to draft. Thereby the paths of a number of drifting floats are drawn out so speedily and accurately that an analysis of tidal flow on the scale of the Seto Inland Sea hydraulic model can be carried out by a few operators in a short period of time.View full abstractDownload PDF (886K) -
Taketoshi NOZAKI1977 Volume 13 Issue 5 Pages 475-481
Published: October 30, 1977
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis automatic driving system is composed of four elements; i.e. the road environment where obstacles such as guardrails and other automobiles exist, a recognition device of the obstacles, a controller which controls an automobile by the information from the recognition device and the automobile which runs according to the control signals. Described in this paper are the control method of the automobile and the driving tests. The control method must be examined from a viewpoint of improving safety of the system. Therefore, a criterion of the safety is defined, and the influences on it by various factors are investigated by means of a computer simulation. The factors are the recognizing distance, the vehicle speed, etc.. A method for the control is suggested on the basis of the results. Then, the driving tests have been performed on the road where obstructive rubber-cones are placed in the form of a lateral convexity or a line. The automatic driving car has shown good behavior of avoiding the convexity and contouring the line, and the technical possibilities of the system have been shown.View full abstractDownload PDF (1911K) -
Motohiro TAKIUCHI, Yasushi HASHIMOTO1977 Volume 13 Issue 5 Pages 482-488
Published: October 30, 1977
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThe leaf temperature was measured by means of an infrared thermometer, which utilized PbTiO3 pyroelectric infrared detector, in connection with plant physiological information.
The infrared thermometer was operated with a band pass filter of 9μm to 12μm, in which region the spectral reflectance of leaf is relatively low (less than 1% for tobacco leaf) and the averaged emissibity of leaf estimated by using Weber's method is nearly 1.0. The stability of the infrared thermometer proved to be better than 0.3K in terms of zero drift in a series of continuous operation and its best resolution of temperature was 0.1K. A high sensitive impedance meter was used simultaneously as for the measurement of water content in plants.
The leaf temperature response of tobacco plant to light inputs and humidity inputs in various condition was measured in a controlled growth cabinet. The water content in stem, the transpiration and the gas concentration were measured simultaneously. As the result;
(1) When the leaf was irradiated, oscillating changes in the leaf temperature in connection with changes in the water content of stem and in the transpiration were clearly observed.
(2) The leaf temperature changed more rapidly at a higher air temperature.
(3) The amplitude of leaf temperature change was greater at a lower air humidity.
(4) A comparison between the results of temperature measurement obtained by thermocouple and those by infrared thermometer showed that the effect of the adhesive that was used for thermal contact of thermocouple on the leaf was not negligible.
The results as a whole shows that the information obtained through the measurement of leaf temperature by means of infrared thermometer can present a key to make clear the phenomena in plant growth.View full abstractDownload PDF (891K) -
Minoru INAMURA, Ryuji TAO, Takashi KATSUMA, Hiromichi TOYOTA1977 Volume 13 Issue 5 Pages 489-494
Published: October 30, 1977
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThe methods of Remote Sensing (R/S) can be classified into three kinds, 1) the measurement of the reflection of the sun beam (Passive R/S), 2) the measurement using millimeter wave or laser rader (Active R/S), and 3) the measurement of the infrared radiation. By these methods, one can obtain information on the measured object concerning its surface temperature, 2) its effective emissivity, and 3) its effective reflectivity.
The surface temperature, in, effect, contains the total information of the under-surface structure. The authors performed a fundamental experiment for extracting such an under-surface information by R/S, which they name “Dynamic Remote Sensing.”
In the first place, we decided the spatial transfer function of the medium (sand in our experiment), and then, filtering the surface temperature pattern, calculated the under-surface temperature pattern, from which we estimated the form of the sample in the medium
In the second place, we analysed the relation between the thermal input (the temperature in the bottom) and thermal output (the surface temperature) in the analogy of electric circuits, calculated the heat capacity and the thermal conductivity of the sample, and estimated its substance.
As a result, the present study will guide us to a new method for the exploration of groundwater or minerals and undestractive tests.View full abstractDownload PDF (585K) -
Masakazu HARADA, Osamu OYAMA, Atsushi KANEURA1977 Volume 13 Issue 5 Pages 495-501
Published: October 30, 1977
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThe probability distributions of switching delay in the circuit of wall attachment devices can be described as similar to a weibull distribution. Histograms and probability distributions are investigated by 500 or 600 measurements in each test. The effects of tubing to swiching probability are measured statistically in cascaded OR-NOR devices. The minimum switching delay is evaluated by extrapolation 0.55ms by a series of 100 measurements.
The triggering pressure levels to open and to close the gate of electronic time counter have a close connection with the switching probability distribution. Noise effect in output signal can be eliminated by setting the triggering level at 90% of the output height is obtained after the effect of reflection is taken away.
The reflected compression waves cause a high rate of switching probability density observed also in the initial switching. On the other hand, reflected expansion waves cut down the probability density or switch back the main jet after switching in a particular case. The tube size to match the nozzle size of the device is calculated to be 1.25mm in the evaluated signal height. The minimum switching delay was obtained with a 2mm tube.View full abstractDownload PDF (2565K) -
Yukio MATSUMOTO1977 Volume 13 Issue 5 Pages 502-508
Published: October 30, 1977
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThe author previously presented a method of analyzing the dynamic characteristics of pneumatic transmission lines with steady flow by dividing the lines into finite number of sections and connecting the characteristics of the sections each other. The transfer functions thus obtained, however, are too complicated to yield the relation between input and output signals.
Here, starting from the fundamental equations, transmission line matrices about pressure and mass flow are derived in the form which facilitates the comparison of the cases with and without steady flow. And moreover, by considering the effect of steady flow, terminal volume matrices are also derived.
By combining the above mentioned line and the terminal volume matrices, the dynamic characteristics of volume terminated pneumatic transmission lines with steady flow are analyzed.
Frequency characteristics derived by this method are compared with the experimental results and a good agreement have been obtained.View full abstractDownload PDF (734K) -
Yajiro MORITA, Masashi ISHIGAMI, Akira KOBAYASHI, Mikio SHIIZUKA1977 Volume 13 Issue 5 Pages 509-516
Published: October 30, 1977
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSOur technology can provide us with much measured data concerning the multidimensional properties of an industrial product. Studying the data we are to evaluate whether the product is good, bad, of the first class or of the second class. Any evaluation of this type inevitably depends on our subjective judgement. Be it subjective or not, we may express the evaluation as introducing an order relationship among the measured entities.
In this paper the pattern measurement is defined. In treating the problem of classifying measured objects according to a certain ordering rule, this paper discusses some fundamental problems of the pattern measurement, the properties of a partially ordered set, and numerical expression of the “goodness of measurement”.View full abstractDownload PDF (462K) -
Teruyuki IZUMI1977 Volume 13 Issue 5 Pages 517-522
Published: October 30, 1977
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper presents the method of generating N-dimensional normally distributed random signals from a single binary random signal.The main constituents on the basis of this method are the n-stage shift register whose state represented by a vector α(t) is varied irregularly by transmitting the binary random signal, and the N weighted adders whose k-th output is xk=α(t)tωk, where ωk is the weight vector of the k-th weight adder. It is theoretically shown that, if ωk is proportional to a k-th degree orthogonal polynominal Pkn-1, the cross-correlation function of xk and xl for different arbitrary k, l is zero, the probability density function of xk is a normal distribution when n is infinite, and xk and xl are statistically independent. Therefore it is possible to generate the N-dimensional normally distributed random signals with an arbitrary covariance matrix by combining the {xk; k=1, 2, …, N} linearly.
It is made sure by a digital simulation that this method can be put to practical use.View full abstractDownload PDF (740K) -
Masakazu MATSUBARA, Yoshiyuki NISHIMURA, Masao IMAEDA1977 Volume 13 Issue 5 Pages 523-530
Published: October 30, 1977
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSSome digital simulation studies of photochemical smog dynamics are undertaken by use of a model in which the mass transfer and the chemical reaction kinetics are described by the travelling air parcel model and by the Eschenroeder-Martinez model, respectively. Of various factors governing the formation of photochemical smog, uncontrollable factors such as meteorological and geographical conditions are fixed at some standard values while the initial concentrations of nitrogen oxides (NOx) and hydrocarbons (HC) at the base of the air parcel are chosen as controllable factors. One daytime simulation from 7.00 to 17.00 is repeated for various sets of initial concentrations and the maximal value of the generated ozone concentration [O3]max and the corresponding time tmax are determined. Then, contour curves of [O3]max and tmax are traced on a plane whose coordinate axes correspond to the fictitious initial concentrations of NOx and HC which will be attained if they diffuse uniformly within the air parcel.
Distinctive features of photochemical smog dynamics which are clarified through the above simulation studies suggest some useful strategies for the smog control. The generated secondary pollutant concentration depends much more on the primary pollutant composition than on its total concentration. In consequence, the suppression of total primary pollutant emission does not always result in the secondary pollutant concentration decrease, but it is much more effective to bias the primary pollutant composition toward either one of NOx and HC in order to decrease the secondary pollutant concentration. The automobile traffic control will cause the change in the average engine operating mode together with suppression of total primary pollutant emission and so a marked change in the primary pollutant composition will result in. Therefore, an insufficient traffic control will be apt to be the cause of a secondary pollutant concentration increase.View full abstractDownload PDF (1120K)
- |<
- <
- 1
- >
- >|