All issues
Volume 6 (1970)
- Issue 6 Pages 483-
- Issue 5 Pages 385-
- Issue 4 Pages 289-
- Issue 3 Pages 195-
- Issue 2 Pages 97-
- Issue 1 Pages 1-
Volume 6, Issue 2
Displaying 1-13 of 13 articles from this issue
- |<
- <
- 1
- >
- >|
-
Mikio NAKATSUYAMA1970 Volume 6 Issue 2 Pages 97-103
Published: 1970
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSSweep oscillator have been widely used for measuring the frequency characteristics of various devices because of their simplicity and convenience. The characteristics of a sweep oscillator, however, have not been analysed precisely. Measured values are different from the values in the steady state, since the frequency of the sweep oscillator increases exponentially. Measured values should be essentially regarded as the transient ones.
In this paper, at first, it is clarified that the frequency of a sweep oscillator is exactly proportional to the exponential function. The sweep oscillator refered in this paper is a kind of Wien bridge oscillator.
Differential equations which are valid in the first or the second order system to be measured with a sweep oscillator, are normalized and are analysed numerically by Runge-Kutta-Gill method. The error of the first order system is sufficiently small even if a sweep period is substantially short. The error of the second order system is larger when Q of the system becomes higher. Q and the resonant frequency measured with a sweep oscillator are heavily different from the ideal values especially when the sweep period is shorter or Q is higher. The true value of Q may be calculated from the measured value of Q by a monograph showed in this paper. The error of the higher order system may be calculated with the same methods.
The frequency characteristics of systems are normally recorded with a pen recorder and the output voltage of the system is usually compressed by a logarithmic amplifier before recording. The error caused by the AC-DC converter of the logarithmic amplifier is also calculated and the measured value of Q is shown graphically. The error caused by the AC-DC converter is sufficiently small.View full abstractDownload PDF (797K) -
Masayuki HATTORI, Ryoichi TAKANASHI, Tsunetomo ANNO1970 Volume 6 Issue 2 Pages 104-108
Published: 1970
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSFriction-torque characteristics of DC servomotor in the range of lower speed have not been measured because the motor, in this range, could not even sustain rotation, not to speak of rotating with constant speed.
In this report, measuring method of such characteristics using rate feedback and experimental results on conventional motors are described.
Furthermore it is clarified that the friction-torque of the DC motor in the range of lower speed are concerned with not only motor speed but also angular position of rotor of the motor.View full abstractDownload PDF (646K) -
Kusuo YAMAGUCHI, Masao SAKURAI1970 Volume 6 Issue 2 Pages 109-116
Published: 1970
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSColour density value which is defined by extinction coefficient at the definite wavelength is usually measured by a photoelectric colorimeter. As the monochromatic light source of a colorimeter usually has some band width, data obtained with such a colorimeter do not obey the Lambert-Beer's law. This paper shows that the disagreement with the Lambert-Beer's law results from the difference between pure monochromatic light in definition and pseudo-monochromatic light in actual measurement, and that the correcting curve can be made by the calculation from spectrum of measured liquid and other optical elements. It was verified that the corrected data with this obey the Lambert-Beer's law.
Using above results, we assembled a prototype of colour density value measuring system which is a self-balancing type with varying light pass area. It has dual beam and a simple sliding iris which gives a standard light attenuation equal to the light attenuation in sample liquid.
Next, a line spectrum of a mercury lamp of which intensity pulsates at the double frequency of the AC power source was used as pure monochromatic light source. In this case, a particular electric circuit is necessary for taking out the difference between two pulsating signals of photoelectric detectors. The measuring system with this pure light source has been tested, and good results were obtained.View full abstractDownload PDF (1044K) -
Iwao YAMAMOTO1970 Volume 6 Issue 2 Pages 117-125
Published: 1970
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSA pneumatic micrometer, developed to apply as a linear deflection type meter, is described. This is based on the measuring system of both a cascade connected pneumatic amplifier, composed of two pneumatic amplifier units with a high pressure supply, and a linear feedback transducer. The transducer is constructed from a set of two bellows and an elastic plate which has a rectangular crosssection, and the measuring range of the meter can be changed by the selection of the effective area of the bellows of the transducer.
Since the position of the gauging nozzle of both the input and output stage is finely adjustable, the amplification and the measuring force of the device can be widely selected by adjusting them. The performance of this device is shown as follows; measuring range: 12, 60μ, output level: 0.6kg/cm2 (linearity 1%) gauging force: 0.4∼0.5g, responce time: 3∼4sec (95% settling time) loading volume: 1000cm3). Applications are also described on surface waviness measurement and use as a comparator using a pair of the measuring system.View full abstractDownload PDF (1714K) -
Minoru HAYASE, Kunihiko ICHIKAWA1970 Volume 6 Issue 2 Pages 126-132
Published: 1970
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis is the experimental work concerned with the optimal servo-mechanism in which the future value of desired function is utilized. The future value of the desired function means the time function of the desired signal over some interval subsequent to each present time. The time function generated by serially connected 10 delay equipments can be used for the experimental studies of this class of the optimal servomechanism. The output of each equipment is the time function of the input with a time delay of Δtf. The output of the final delay equipment is used as the fictitious present value of the desired function. Then, the output signals of the remaining delay equipment can be regarded as the future values of the desired function with respect to the fictitious present value mentioned above.
The experimental results for various kind of desired function have revealed that the optimal servo-system of this class is much superior in tracking property to any other conventional servo-system.View full abstractDownload PDF (3900K) -
Yoshiyuki NAKANO, Shunjiro HAIJMA1970 Volume 6 Issue 2 Pages 133-141
Published: 1970
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSA new sequence control system structure is applied to an automatic burner control system which is installed at a large capacity boiler of a thermal power station and controls lighting off and shutting down sequences of a large number of burners. Concerning the whole system structure, instead of having an exclusive sequence control block for each burner as in the conventional systems, this system has only two sequence control blocks shared among all burners and operates each burner alternately by the arrival order of operation request. This conception avoids idle components and attains a very efficient system.
The sequence control block consists of several linkages of sequence units. Each unit performs a unit sequence in burner operation. This sequence control block construction method simplifies the design steps of the block, the alternation of its functions, the follow-up of the sequence and the detection of troubles, and moreover realizes highly safe control.
A prototype model of this newly developed system was prepared and experiments were performed with actual burners. It was found that the model establishes optimum and uniform means to operate burner systems.View full abstractDownload PDF (2952K) -
Takumi NOMURA, Kahei NAKAMURA1970 Volume 6 Issue 2 Pages 142-151
Published: 1970
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn this paper it is aimed to make an explicit expression of controllability condition of the system which is described by a linear partial differential equation. At first is shown the input-state relation of the system, and next is discussed the controllability condition by virtue of authors' already presented theorem.
The “genuine solution” of differential equation for square integrable input function can not be known in general, but the “weak solution” which is equal to the “genuine solution” almost everywhere in its domain, can be obtained by the method proposed in this paper. In order to use this “weak solution” as the input-state relation, it is necessary to introduce the topology under which every functions equal each other almost everywhere are identified as the same. From this point of view, the L2 topology is introduced into the state space.
In the partial differential equation systems, there are two types of controls, one is spatially distributed, the other is concentrated on the boundary. In this paper are derived input-state relations for these two kinds of control and reduced to congruent equations in l2 space. To these relations in l2 space, is applied the theorem obtained by authors in ref. (2) treating the necessary and sufficient condition for a system to be controllable. And then this condition is expressed explicitly in terms of both the elliptic differential opereator of the differential equation and the operators which describe the control mechanisms. The operators which describe the boundary control mechnisms are bounded operators defined on the boundary, but they become unbounded ones in the input-state relation in l2 space. Therefore, in the cases of boundary controls can not be applied the same reasoning as in distributed controls. However, it is also possible in this case to obtain the explicit expression of controllability condition by the appropriate modification of the theorem.View full abstractDownload PDF (1201K) -
Takashi SEKIGUCHI1970 Volume 6 Issue 2 Pages 152-160
Published: 1970
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper deals with the input observability and the duality between inputs and outputs of the linear dynamic measuring system which is time-continuous and analytic.
First we introduce the necessary and sufficient conditions for the input observability of pulse and pulse series in the time varying system. We obtain slightly different conditions between time-fixed pulse series and time-free pulse series. The necessary and sufficient condition for the total input observability of pulse series is easily obtained from that of time-free pulse series.
In case of the time constant system, the necessary and sufficient condition for the input observability of power series is of the same form as that of pulse series.
The above necessary and sufficient conditions are analogous to those of state observability. Considering more general inputs capable of Laplace transformation, however, we need independent output variables of no smaller number than that of input variables.
If the system has the state observability, then the necessary and sufficient condition for the input obsevability is of much simpler form. This relation is similar to that between state controllability and output controllability.
At the end of this paper, it is shown that the pulse input observability is dual to the output controllability.View full abstractDownload PDF (1161K) -
Noboru KUNIMATSU1970 Volume 6 Issue 2 Pages 161-167
Published: 1970
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn this paper, the optimal control problem for a distributed parameter system described by a liner high order partial differential equation is formulated with a performance index of the final energy of the system.
It is pointed out that a class of scalar-formed high order partial differential equation can be effectively reduced to a vector-formed lower order partial differential equation, by introducing a new state vector composed of variables which are enough to assign the total energy of the controlled system.
Also pointed out is that a high order system with complex boundary conditions, such as a homogeneous but time-variant condition or a non-homogeneous condition involving controls, can be transformed to a system with simpler boundary conditions such as a time-invariant or a homogeneous one, respectively, by introducing the extended definition of operators. Then, Dynamic Programming technique is used to derive the optimal condition, which proves to include the already known results given by Russel and Komkov.View full abstractDownload PDF (697K) -
Masahiko ONO, Kazuo HORIUCHI1970 Volume 6 Issue 2 Pages 168-174
Published: 1970
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper is concerned with nonlinear control system with finite memory, the input-output relation of which is governed by nonlinear Volterra integral equations. Investigation concerning global properties of this system is made by the method of functional analysis, that is, existence and uniqueness problems of output signals are reduced to mapping problems in signal space, where n-dimensional direct product space of L2-Banach space is constructed as signal space (n denotes the number of output signals). And, with the help of well-known contraction mapping principle, it is shown that, under several basic assumptions imposed on the system, there always exist output signals with finite total energy, corresponding to any input signals under the same condition, that the upper bound of output signals for almost all time t can be constructed concretely, and that the procedure of calculation of output signals can be established by the method of successive approximation.
But, on the other hand, contraction mapping principle requires a very strict restriction such that Lipschitz coefficient associated with mapping should be less than unity. Then for the purpose of removing this restriction, extended. L2-Banach space is introduced with denotation L2(σ), which is defined as the whole of measurable functions satisfying the following condition: ||x||(σ)={∫∞0|x(t)|2e-2σtdt}1/2<+∞(σ≥0) and by constructing n-dimensional direct product space of L2(σ), and regulating the parameter σ, the problems mentioned above can again be reduced to contraction mapping principle, even for more general systems. In addition to this, by using the properties of norm||·||(σ), it is shown that there always exist output signals with finite mean power, corresponding to any input signals under the same condition.
As results, very comprehensive control systems can be treated even for signals with infinite total energy, and so, even in this case, upper bound of output signals can be constructed concretely.View full abstractDownload PDF (748K) -
Ichiro SUGA1970 Volume 6 Issue 2 Pages 175-181
Published: 1970
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSDesigning method of linear control systems by means of the theory of optimal regulator problem with a quadratic performance index has many advantages compared with conventional design methods. For example, under reasonable restrictions this method always assures a stable control system by straightforward computations.
However, if some characteristics (overshoot, settling time, etc.) of the system to be constructed are previously required, there are no general rules to reflect these requirements on the parameters of the quadratic performance index.
One shortcoming of the regulator problem is that signals from outside are only considered as impulsive disturbances added to each integrator, since reference or command is not assumed to exist in the regulator problem. This is not convenient to desigh servosystem.
This paper discusses an extended regulator problem effectively applicable to the case where design requirements are given and a stepwise reference exists in addition to the impulsive disturbances. In this problem there are introduced a model fulfilling the required characteristics and a modified quadratic performance index containing the value of control function necessary to keep the plant at a steady-state.
The optimal system constructed by this method consists of the both of a feedback controller and a feedforward controller, and its step response can satisfactorily be kept close to that of the model without offset.
An illustrative example is also presented.View full abstractDownload PDF (890K) -
Yoshifumi SUNAHARA, Kohji OHMINATO1970 Volume 6 Issue 2 Pages 182-188
Published: 1970
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSRecognizing the existence of identification errors due to variations of system parameters, an approximate method is described of state estimation for nonlinear dynamical systems with state-independent noise. Guided by the state space representation concept in control theory, a mathematical model of the dynamical system is given by the nonlinear vector stochastic differential equation of Itô- type.
The first part is devoted to the verification of the existence of a unique continuons solution for the nonlinear vector stochastic differential equation containing a random parameter. Both the α and β stochastic sensitivity equations are established through the precise definition of stochastic sensitivity.
The remainder of this paper deals with an application of the stochastic sensitivity concept to the state estimation problem. Effects on the state estimator dynamics due to variations in system parameters are considered by a numerical example.View full abstractDownload PDF (865K) -
Etsujiro SHIMEMURA, Tatsuo KITAJIMA1970 Volume 6 Issue 2 Pages 189-194
Published: 1970
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSA computational algorithm for the optimal control of the coupled system of lumped parameter and distributed parameter subsystems is developed by using a gradient method in a function space.
In this method, an approximation of a distributed parameter subsystem by a lumped parameter one is not necessary, and in principle an arbitrarily close solution to the true optimal control could be obtained. Examples are given to show the effectiveness of the developed method.View full abstractDownload PDF (651K)
- |<
- <
- 1
- >
- >|