All issues
Volume 10 (1974)
- Issue 6 Pages 641-
- Issue 5 Pages 521-
- Issue 4 Pages 389-
- Issue 3 Pages 263-
- Issue 2 Pages 137-
- Issue 1 Pages 1-
Volume 10, Issue 3
Displaying 1-18 of 18 articles from this issue
- |<
- <
- 1
- >
- >|
-
Tsuneo YOSHIKAWA, Yuji YOSHIDA1974 Volume 10 Issue 3 Pages 263-271
Published: June 30, 1974
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper deals with an optimal control problem of a discrete-time linear stochastic system with two control stations, each of which can make certain noisy measurements of the state of the system. These two stations cooperate in trying to minimize the same performance index, but do not exchange information. Certain constraints are imposed on the control structure, and so our problem is to determine the optimal values of the structure parameters of the stations.
Weak separability is introduced as an extended concept of the separation theorem, and sufficient conditions for the problem to be weakly separable are obtained. The usefulness of the concept is illustrated by several numerical examples.View full abstractDownload PDF (1046K) -
Yoshimasa KAMEYAMA, Hayatoshi SAYAMA1974 Volume 10 Issue 3 Pages 272-277
Published: June 30, 1974
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSFor state inequality constrained optimal control problems, we have discontinuities in the adjoint variables at the junctions of unconstrained and constrained arcs. Thus it is very difficult to obtain the optimal solutions to these problems.
The purpose of this paper is to show that SUMUP (Sequential Unconstrained Method Using Penalty) for nonlinear programming can be extended to optimal control problems with state constraints.
The penalty function used in SUMUP is the generalized form of the functions for interior and exterior penalty methods and contains functions corresponding to Lagrange multipliers.
The computational features in this method are as follows:
1. We need not begin with a trajectory interior to the constraint set.
2. The integration of the adjoint equation does not become inaccurate.
3. This method can approach the optimal solution from the interior or exterior of the constraint set by choosing the functions corresponding to Lagrange multipliers.
An example is solved and the result presented.View full abstractDownload PDF (729K) -
Suguru ARIMOTO, Toyokazu AONO1974 Volume 10 Issue 3 Pages 278-283
Published: June 30, 1974
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThe problem of performance deterioration in optimal regulators, which is caused by the introduction of observers, is investigated. Time-invariant Luenberger observers with minimal order are used as state estimators for deterministic linear time-invariant systems. A necessary and sufficient condition for the performance deterioration to be made arbitrarily small is given, under the assumption that the observer state can be chosen in terms of the regulator output at initial time.
It is shown, according to this condition, that n-m=1, where n denotes the dimension of the regulator state and m denotes the dimension of the measurable output vector, implies the existence of observers with minimal order satisfying a prescribed small deterioration of the performance, and n-m≥2 implies, in almost all cases, that if the initial state, x(0), of the regulator belongs to a fixed subspace, V, of Rn, the performance deterioration, ΔJ, is bounded away from zero as shown by the form ΔJ≥ρ||x||2, where ρ is a positive constant independent of the choice for observers.View full abstractDownload PDF (736K) -
Haruo YODA, Sadahiro IKEDA, Masakazu EJIRI1974 Volume 10 Issue 3 Pages 284-289
Published: June 30, 1974
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn order to automate production and physical distribution systems involving the handling of a variety of products, it is necessary to develop visually sensitive systems that can “see” objects and adapt to the changes of their forms and positions. To meet this requirement, the authors developed a visual device around a TV camera and a simple servo-controlled handling device connected to a control computer. These devices can automatically sort objects flowing along a belt conveyor by visual information transmitted by the TV camera.
The object recognition process of the device involves the following steps: (1) Image input, (2) Image positioning, (3) Transforming into polar coordinates, (4) Matching process utilizing registered object images and a rotational operation.
The computer program for these devices is about 12 kilowords, including a visual data area. From the experiments, it is found that the maximum recognition time is 1 second, and the positioning error in stacking objects is within ±1mm in the X-and Y-directions and less than 10 degrees in the angular direction.View full abstractDownload PDF (1994K) -
Theoretical Considerations for the Measurement of Body FiuidZen-uemon ABE, Kunio TANAKA, Masao HOTTA1974 Volume 10 Issue 3 Pages 290-297
Published: June 30, 1974
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThe possibilities for the measurements of the blood flow and water contents in the human body by NMR techniques have been analyzed during the last four years. A new technique for this purpose is proposed as one of the new approaches.
This paper will deal with the following four topics connected with our past studies for noninvasive testing of biological materials.
(1) Proposal of magnetic focusing techniques and generation of the focusing field.
(2) Deterioration of the signal level originating frcm the loose coupling between the measuring target and the detector, and minimum limit for measurements of the stationary body fluid utilizing steady NMR.
(3) Improvement on the blood flowmeter utilizing pulse NMR.
(4) Measurement of the stationary body fluid utilizing pulse NMR.
For non-invasive measurements of the nuclear magnetic substances distributed in the human body, a new approach is proposed by utilizing the magnetic focusing technique on the measuring target. And several methods for generating this magnetic focus are discussed theoretically. Moreover, the degree of inevitable deterioration of the signal from the focus induced by a loose coupling between the target in the body and the detector on its surface is calculated theoretically.
Theoretical estimations of a new blood flowmeter are shown by taking advantages of the tag effect by radio frequency pulse on the blood flow.
For this improvement of the blood flowmeter, the techniques are explained to estimate the pseudo-signal intensity originating from the neighbouring tissues around the blood vessel. A method similar to blood flowmetry can also be applicable to detect the stationary body fluids with some modifications.
And some simple assumptions of the minimum limit of detectability are summarized from the view point of bio-medical applications.
In conclusion, the authors are able to explain a few considerations and theoretical prospects on these applications of NMR.View full abstractDownload PDF (1135K) -
Hitoshi SASAI, Toyoho FUKUDA1974 Volume 10 Issue 3 Pages 298-303
Published: June 30, 1974
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSSince the state space of delay-differential systems become function spaces, there is a great difficulty in solving or discussing various control problems directly.
In this paper we introduce approximation systems, which are widely used in computing or discussing control problems in linear delaydefferential systems.
It is shown that such approximations are reasonable. Furthermore, the conditions of controllability of the original systems are obtained in terms of the conditions of approximation systems.View full abstractDownload PDF (602K) -
Tomoaki MORINAGA, Yuichiro YOSHIMOTO1974 Volume 10 Issue 3 Pages 304-310
Published: June 30, 1974
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSOne of the important application fields of fluidics is the process controller. There are servral reports on fluidic process controllers, but none on a controller with PID control modes.
The PID controller herein reported is composed of wall-attachment fluidic elements and operates with a PWM signal and a differential type algorithm. The input error signal is pulse-width modulated, and the PWM signal is delayed first by one period and then by two periods. The output difference is obtained by addition and subtraction of the original and delayed PWM signals, and then integrated by capacity.
The characteristics of the controller are discussed here theoretically and experimentally.View full abstractDownload PDF (1177K) -
Saburo IKEDA, Masaaki OKAMOTO, Masasuke SHIMA1974 Volume 10 Issue 3 Pages 311-320
Published: June 30, 1974
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThe optimal control of a bilinear distributed parameter system described by a first order partial differential equation is considered. The performance index to be minimized is the integrated square error of the state at the outlet of the system. By means of the Maximum Principle for lumped parameter systems, we have previously solved this problem under steady-state initial conditions and constant boundary conditions.
In this paper, by appling the Maximum Principle for distributed parameter systems, we solve the problem under the more general conditions where both the initial condition and the boundary condition are continuous and mildly varying.
Our main results are as follows: In order to control the system optimally, we should first use the bang-bang control until the outlet value is equal to the desired value, and then the singular control in a wide sense to maintain the desired value. Moreover, the singular control in a wide sense is reduced to the repetition of switching between the bang-bang control and the singular one in a narrow sense under steady-state initial conditions and constant conditions. The result in this case coincides with the result in our previous paper.View full abstractDownload PDF (1155K) -
Takao TOMARU1974 Volume 10 Issue 3 Pages 321-326
Published: June 30, 1974
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn this paper a unified recursive algorithm is realized to determine the partial fraction expansion coefficients of of the most general form of rational functions without any kind of distinction.
(1) Fundamental recurrence formulae are derived for the expansion of the product of two kinds of multiple poles, and for the Laurent series expansion of numerator polynomials.
(2) Coefficients a (l, i) of the Laurent series expansion of denominator 1/Q(z), and b(l, i) of the numerator polynomial P(z) are given by the iterative use of these recursive formulate.
(3) The product of the coefficients a (l, i) and b(l, i) constitutes the partial fraction expansion coefficients c(l, i) and c0(l) of P(z)/Q(z).
The algorithm is suitable both for machine calculation and for figures.
The application of the algorithm to computer programs is discussed.
The accuracy of the calculated values of coefficients clarifies the effectiveness of the algorithm.View full abstractDownload PDF (619K) -
Case with Information Exchange on Control Variables with Finite Unit Time DelayTsuneo YOSHIKAWA, Hiroaki KOBAYASHI1974 Volume 10 Issue 3 Pages 327-332
Published: June 30, 1974
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper introduces a definition of observability of decentralized control systems and gives a necessary and sufficient condition for a decentralized control system to be observable.
Consider a system with two control stations, x(k+1)=Ax(k)+B1u1(k)+B2u2(k), k=0, 1, 2, … yi(k)=Cix(k) i=1, 2.Let the set lZi(k) of data available to station i at time k be lZi(k)={yi(0), …, yi(k), ui(0), …, ui(k-l), uj(0), …, uj(k-l)}, where {uj(0), …uj(k-l)} are assumed to be communicated to station i from station j. (i, j=1, 2, i≠j, 2≤l<∞.)
We define the system to be observable by station i if x(k) can be determined from lZi(k) for some finite k. Then a necessary and suffcient condition for a system to be observable is that (i) {A, Ci} is an observable pair (in the usual sense) and (ii) rank CiBj=rank Bj.View full abstractDownload PDF (674K) -
Hiroshi YAMADA1974 Volume 10 Issue 3 Pages 333-339
Published: June 30, 1974
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn general, conventional hysteresis motors have a small torque-inertia ratio as compared with conventional servo motors, because the hysteresis material has a small hysteresis loop, a large loss caused by harmonic waves based on slots, and a large specific gravity.
The author has proposed a new rotor structure of an axial air gap hysteresis motor in which anisotropic barium ferrite and Al-alloy plates are utilized. The barium ferrite has a large hysteresis loop, a small loss caused by harmonic waves based on the slots, and a small specific gravity. The proposed motor has a small inertia as a print motor, a large starting torque and a large synchronous torque as compared with conventional hysteresis motors.
This paper presents the structure, the induction of the equivalent circuit containing hysteresis and induction torques, and the experimental characteristics of the motor. The experimental characteristics are close to the characteristics of a print motor.View full abstractDownload PDF (1444K) -
Hideaki KANOH1974 Volume 10 Issue 3 Pages 340-347
Published: June 30, 1974
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThe aim of the present work is to study the static and the dynamic behaviors of parallel and counter flow heat exchangers controlled by manipulation of flow rate. It is shown that PI-feedback control system has a unique steady state. The PI-feedback control system has an unstable region in which the limit cycle is caused by the nonlinearity of the manipulation of flow rate rather than by the valve saturation. The stable region may be determined by a parameter plane method. Experimental results verify the existence of the limit cycle and agree well with the theoretical calculation.View full abstractDownload PDF (865K) -
Takeshi NAKADA1974 Volume 10 Issue 3 Pages 348-354
Published: June 30, 1974
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThere are several methods of Pulse Width Modulation (PWM). For any method, the output of the pulse width modulator corresponding to a sinusoidal input consists of the various frequency components, depending on the frequencies of the input and the carrier. Generally speaking, some frequency components lower than the input frequency are mixed therein. When the ratio of the carrier frequency (fc) to the input frequency (fe) becomes small, the signal cannot be accurately transmitted, because these components become large. In communication, it is easy to make this ratio large enough. On the other hand, in control, the carrier frequency (fc) is limited by some function of the equipment in the systems, and it is required to broaden the bandwidth of the systems as far as possible. So it is important to use a method by which a frequency signal as high as possible can be transmitted on a constant (fc). For the above method it is necessary to clarify the limit of the ratio (fc/fe) from the viewpoint of accurate signal transmission.
In this paper, four typical modulators are shown. First their characteristics are analyzed theoretically. Second each modulator is compared from the viewpoints of the frequency characteristics and the so-called “effect of the removal of time independent nonlinearities” this is the significant advantage of the PWM. Finally as the result of the comparison, it is concluded that the method using both edges is the best one.View full abstractDownload PDF (882K) -
Oichi ATODA, Yasuhiro DOI1974 Volume 10 Issue 3 Pages 355-362
Published: June 30, 1974
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIt has been a difficult task for analog electronic circuits to delay signals by a few milliseconds or more. Theoretically a ladder circuit consisting of numbers of lumped inductors and capacitors might act as a good approximation of delay cable if the signal fre-quency is limited below a certain bound, but unfortunately the characteristics of such a circuit cannot really be satisfied because of the imcomplete performance of conventional large inductance coils.
In this paper the authors propose a method for forming a delay line using resistors and active elements instead of classical inductors and capacitors. The admittance across one active element is proportional to the minus of the square of the frequency. Each active element consists of several transistors and a few additional capacitors and resistors, and the “active delay line” requeres no more coils than an active filter does. This is just the reason why the active delay line could be almost as ideal as a theoretical one.
Two active delay lines are actually manufactured, one of which is now used as the delay element of a correlation flow velocity meter. Satisfactory results are obtained in experiments on these delay lines, which imply practical applicability of the active delay line.View full abstractDownload PDF (2423K) -
Tomoaki MORINAGA, Masatake SHIRAISHI1974 Volume 10 Issue 3 Pages 363-369
Published: June 30, 1974
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThe usual type of pressure-reducing valve always causes offset due to its proportional control mode. The improvement of its characteristics, on which one of the authors formerly reported, is bounded within a limit, so far as the proportional control mode is adopted. For drastic improvement, the integral control mode must be adopted.
A new type of self-operated pressure regulating valve reported here has on-off integral control mode which is realized by the use of a push-pull type nozzle-flapper, restrictions and diaphragms. The controlled output pressure is almost constant within a remarkably wide range of flow rate, compared with the case controlled by the usual type of pressure-reducing ing valve. Steady state errors for the change of set point and supply pressure are also very small, and the amplitude of steady state pressure cycling is practically of negligible order. The cycling phenomena are discussed here theoretically by the descri-bing function method and the main cause of the cycling is revealed to be the friction on the valve stem, rather than the on-off character.View full abstractDownload PDF (784K) -
Hirobumi OHTA1974 Volume 10 Issue 3 Pages 370-377
Published: June 30, 1974
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSMulti-input/multi-output model reference adaptive control systems (MRACS) are investigated in this paper. The main purpose of this analysis is to guarantee the stabilty of the systems by means of Popov's hyperstability thorem, while most MRACS already designed by Liapunov's second method are single-input/single-output systems.
An analysis by means of the hyperstability criterion is first given by Landau where two important problems are left behind. One of these is a problem containing differentiators, and the other is developing an analysis under the assumption that state variables are measurable. It is therefore one of the objectives of this paper to present a method for designing MRACS without using differentiators. The result is given in Theorem 2 and includes all the results obtained by Butchart and Shackcloth, Parks, Winsor and Roy, and Porter and Tatnall via Liapunov's scheme. It is essentially owing to the positive realness of the transfer function matrix of the system, that MRACS are constructed without using derivatives of output errors. The result in the case where state variables are not measurable is presented in Theorem 3 where a dynamics similar to Luenberger's observer is introduced in the adaptive mechanism, corresponding to that given by Dressler, who has investigated an approximate parameter adaptation rule by means of sensitivity consideration.
The two results above are directly applicable to the design of MRACS and give insights into a general method for studying this type of adaptive system. Finally comparisons between the results here and those by Liapunov's method are given together with examples.View full abstractDownload PDF (1038K) -
Kazuhiko KUDO, Shin-ichi SATO, Masaru TAKATA1974 Volume 10 Issue 3 Pages 378-384
Published: June 30, 1974
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn the recognition of the nature of an object by artificial fingers, the selection of sense elements and the processing method of the information from them are important. In this study a sense element using a strain gauge advanced recognition than that using on-off was made and tested to get the sense information in the form of analogue voltage for more switches. This sense element is superior in line arity and bears repetition of the load. It is seen that this element can get information necessary to classify the shape, size and material of the objects.
The concept of “attributes” to describe the nature of the object is suggested, and a method of classification of each attribute is also presented. It is easy to add the new attribute, and the amount of the calculation for classification is reduced in this method. The linear discriminant function method is one of the most popular recognition theories. On the other hand there is a theory using the Bayes minimum risk decision function, and this can be translated to a Chebychev polynomial discriminant function. We compare both methods using the same data. We could not find significant differences between the two methods in the correct rate of answers, but the latter is preferable from the point of view of the time for calculation.View full abstractDownload PDF (1467K) -
Masasumi KOKAWA, Kahei NAKAMURA, Moriya ODA1974 Volume 10 Issue 3 Pages 385-386
Published: June 30, 1974
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSDownload PDF (316K)
- |<
- <
- 1
- >
- >|