All issues
Volume 19, Issue 7
Displaying 1-14 of 14 articles from this issue
- |<
- <
- 1
- >
- >|
-
Setsuo SAGARA, Takami MATSUO, Kazushi NAKANO1983 Volume 19 Issue 7 Pages 519-525
Published: July 30, 1983
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIt is well known that the maximum amplitude of the estimation error of a full order observer becomes extremely large as all the poles approach infinity in the left half complex plane (LHP).
In this paper we clarify the asymptotic behavior of a full order observer for the case where a part of the poles approach infinity in LHP.
First, we show that the dimension m of the outputs is equivalent to the number of poles which can approach infinity in LHP without diverging of the initial estimation error. Next, we prove that the full order observer converges to a minimal order observer except the initial state as the m poles tend to infinity in LHP. Finally, we investigate a full order observer estimating a linear function of the state variable under the condition that a part of the poles approach infinity in LHP. The obtained result is considered as an extention of Mita's result under the condition on all the poles.View full abstractDownload PDF (787K) -
Kunihiko ICHIKAWA1983 Volume 19 Issue 7 Pages 526-530
Published: July 30, 1983
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSExact model matching is considered to be the most reasonable design method for the control systems. As the means of exact model matching, two approaches have been developed; one is the time domain method and the other is the frequency domain method. The latter is more general than the former in that it can be applied to multivariable systems. Furthermore, the required calculations are far simpler for the frequency domain method. It has been recognized that the resultant control system consists of input dynamics and linear state variable feedback part in general. In the first half part of this paper, it is pointed out that the resultant control system for the scalar plant can always be transformed into the complete feedback control system.
It is well known that two conditions are necessary and sufficient to the achievement of exact model matching; (1) the plant is of minimum phase, and (2) the relative degree of the desired transfer function is not less than that of the plant. It is the fact that there are many nonminimum phase plants. Furthermore, the plant of minimum phase in continuous time turns to the nonminimum phase plants in discrete time often. The adaptive control of nonminimum phase plant is the recent important subject, and it should be noted that the basic concept for the adaptive control links firmly to the exact model matching theory.
The method to achieve exact model matching for the nonminimum phase plant is presented in the second half of this paper. The basic idea relies on the fact that any transfer function can be represented as the difference of two minimum phase transfer functions. The problem can be reduced to the ordinary exact model matching one by setting up a compensator in parallel with the plant to be controlled. The design method is illustrated by an example.View full abstractDownload PDF (639K) -
Tomoo TAKEGUCHI, Hajime AKASHI1983 Volume 19 Issue 7 Pages 531-537
Published: July 30, 1983
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn practical problems of decision making by the maximum expected utility criterion, there are not a few cases where it is difficult to identify utility functions. For such cases, however, it is relatively easy to find what the decision maker's attitude toward risk is, i.e., whether it is risk aversion or decreasing risk aversion. With such a partial knowledge of utility functions, stochastic dominance (SD) can be used as an approach in ranking alternatives by comparing their structure of probability distributions. In this paper, single-attribute SD rules are first surveyed. Then, by introducing multi-attribute value functions, multi-attribute SD rules are derived with the aid of single-attribute SD rules. Finally, the relations between the multi-attribute SD rules and statistical parameters are examined. It is found that the relations of statistical parameters obtained represent the decision maker's preference by statistical parameters. Therefore, he could readily understand the content of stochastic dominance by those relations and accept stochastic dominance as a decision supporting method.View full abstractDownload PDF (906K) -
Yasuo SUGAI, Hironori HIRATA, Tadashi KURATA1983 Volume 19 Issue 7 Pages 538-543
Published: July 30, 1983
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn the well known Volterra models resources are unlimited. However, if one regards resources as one of the netrients (e.g. nitrogen, carbon, phosphorus, etc), resources are limited in a fixed region. Therefore studying the models with limited resources and its behavior are important.
This paper has studied two cases of material-cycling ecosystems whose species use renewable-limited resource in common; in one case the species utilize resource directly and in the other case indirectly. For example, being concerned with the nitrogen cycle, there is an ecosystem consisting of phytoplanktons and zooplanktons in a bay. In this case zooplanktons utilize phytoplanktons as resource directly. Moreover, in a one-crop area it can be considered that the system consists of resource (plant), the first consumer (an insect using plant) and the second consumers (insects being parasitic on the first consumer). In such case, parasitic insects utilize resource indirectly through the first consumer. These examples correspond to the above two cases respectively.
Some interesting results are as follows. In both cases, a selection occurs among n species and we can know which species survive by the comparison of macroscopic parameter, i.e., the species having smallest macro-parameter value can survive. There exists a difference between two cases with respect to the available utilization of resource. The species utilizing resource directly can use it more efficiently than the species utilizing indirectly.View full abstractDownload PDF (753K) -
Akira KOBAYASHI, Yasuhiko SAKAMOTO1983 Volume 19 Issue 7 Pages 544-550
Published: July 30, 1983
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper presents a programming technique for solving simultaneous homogeneous linear inequalities formulated in paired comparison method based on band function model. And this technique enables the algorithm previously proposed to solve the inequalities without rounding errors.
This technique is based on following properties; if appropriate lengths are chosen for the edge vectors which represent solution domain, then all elements of them will be integral, and all other variables obtained by the computation using them will also be integral. Specific procedures are presented to determine the lengths of the vectors. For practical reasons, values of these variables must be small enough, so upper bound of every variable is determined as a function of the number of objects. And it is revealed that 28 bit is sufficient for the largest variable when the number of objects is 10.View full abstractDownload PDF (803K) -
Yuichiro HARA, Seiji INOKUCHI1983 Volume 19 Issue 7 Pages 551-556
Published: July 30, 1983
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn harmonic analysis of periodic wave, the discontinuity at the boundary of observation interval increases the spectral leakage. Several temporal windows have been developed to improve the spectral resolution. In Fourier transformation, input signal is transformed into complex data consisting of amplitude and phase. This paper describes the phase transmission characteristics of temporal windows. The phase preservability that the phase of input sinusoidal wave is preserved in peak component, can be improved by applying the windows. Phase error caused by the other harmonics depends on the amplitude transmission characteristics of windows, especially on their band-width. The spectral peak moves from n-th bin to (n+1)-th bin with the increase of input frequency. The frequency, at which the peak is shifted from n to n+1, becomes close to n+0.5 through temporal windows. Experiments of pitch-frequency detection using phase compensation method show that Hanning window and Cos window are more desirable than the others.
As a result, the temporal window improves the phase transmission characteristics.View full abstractDownload PDF (768K) -
Tohru IUCHI, Fumio TANAKA1983 Volume 19 Issue 7 Pages 557-562
Published: July 30, 1983
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper describes a newly developed radiation thermometry of a steel plate in a high temperature furnace. The measuring system of this method is composed of a reference radiator set above an object and a radiometer which detects the radiance from the object surface through the small opening of the reference radiator. The reference radiator plays a role of both shielding the stray radiation noise and supplying a standard radiation signal.
This method has the following features;
(1) the measurement is done without cooling of the measured object because the reference radiator is heated in a furnace, thus free from the bad effect on the object.
(2) the maintenance is easily performed because of its simple construction.
(3) the reference radiator can be selected optionally according to the furnace condition.
Thus, this method is especially applicable to the temperature measurement of a steel in a high temperature furnace more than 1000°C such as a slab heating furnace.View full abstractDownload PDF (720K) -
Measurement of Horizontal Seismic MotionShigeru KUROSU, Juhei YAMADA, Hiraku TOMIZAWA1983 Volume 19 Issue 7 Pages 563-568
Published: July 30, 1983
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis work is concerned with an experimental device which exhibits the characteristics of a horizontal seismometer with a long natural period using a gyroscope, the experimental device having demonstrated periods of 45 seconds and longer. This seismometer is called the gyro-transducer throughout this paper. The principles of operation described here have previously been presented. The gyro-transducer achieves long-period seismometer action through a gyroscope, an ordinary simple pendulum and an electronic feedback. With suitable device parameters the performance equations will be the same form as those of a conventional seismometer, the natural period and damping being functions of an amplifier gain. Experimental results obtained in the laboratory provide useful data to apply the gyro-transducer to a practical seismic observation system. The gyro-transducer is of interest, however, whether or not it finds use as a seismometer. It will be suggested that the principles described here should be applied to the highly precise gyroscope and the very sensitive sensing elements.View full abstractDownload PDF (2054K) -
Shuta Murakami, Mikio Maeda, Yoshiyuki Nagata1983 Volume 19 Issue 7 Pages 569-573
Published: July 30, 1983
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper deals with the speed control of automobile using microprocessor. The purposes of the speed control are to reduce the fatigue which the driver recievesfor long time driving, to save the gasoline consumption, and to keep on driving at the constant speed. For these purposes, two types of controller, that is, the finite settling controller and the state feedback controller based on optimal control are designed.
From results of computer simulations and road tests by the automobile (TOYOPET CROWN '78) equipped with microprocessor, we discuss which controller is better to control the speed of automobile from the standpoints of rise time, overshoot, fluctuations in manipulated variable, saving gasoline consumption, and comfortability of driving.
Finally, the long time road test was implemented by using the state feedback controller based on optimal control, and the comfortable and stable driving was realized.View full abstractDownload PDF (611K) -
Tetsuo SHIOTSUKI, Katsuhisa FURUTA, Masahiro OKAZAKI1983 Volume 19 Issue 7 Pages 574-581
Published: July 30, 1983
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn spite of many advantages to use R/C helicopters the practical use of them is rare because of their instability. In this paper we propose a new construction of training simulator of the helicopter based on model following servo controller and present the experimental results.
The helicopter model has the freedom of six dimensions, and for the movement in 3-d space, a new type machine which does fix no motors on moving stages is proposed. The idea used is to use endless belts and diffrential gear mechanism which enable the simulator to move 3-d spce arbitrarily. The principle of model following servo controller (MFS) is briefly explained and its characteristic is examined by using high gain feedback control technique. When the gain of closed loop increases, the transfer function from model input to plant output approaches to that of reference model. This meanes that the transient characteristic of model following type controller can be improved by using MFS controller. The results of the experiment for real helicopter model (SH-3D) are compared for the model following type controllers of various type, 1) Kreindler's method, 2) Davison's method, and 3) MFS. And the result is obtained that MFS is the most suitable controller to control real plant to follow a linear time invariant model.View full abstractDownload PDF (885K) -
Osami SAITO, Yoshihiko NAKATA, Tatau NISHINAGA, Takuo SUGANO1983 Volume 19 Issue 7 Pages 582-588
Published: July 30, 1983
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSRecent progress in integrated circuit technology requires the high quality silicon crystal. A significant property of the silicon crystal is the uniform distribution of intentionaly doped impurity. However, the usual silicon crystal has the nonuniformity such as the impurity striation, which is not desirable to produce the VLSI. The impurity striation is considered to be formed during the crystal growth with the convection. To understand the mechanism of the nouniform impurity incorporation, authors have been planned the spherical silicon crystal growth in space without convection.
To perform the experiments in space successfully with a few attempts, preparatory experiments for various conditions are indispensable. But the circumstance of nongravity can not be realized on earth in long time, and only the examination with computer simulation might be available.
This paper deals with the computer simulation of spherical crystal growth in space. First, the crystal growth process is discribed by partial differential equations with the moving boundary. Secondly, these equations are solved by using the implicit difference form with the variable grid method. Finally, the simulation results provide us the fundamental knowledges about the temperature distribution, the total experiment time and the accuracy of temperature control law to design the experiment system.View full abstractDownload PDF (881K) -
Tsutomu MITA, Yasuhiro OHOYAMA, Katsuhisa FURUTA1983 Volume 19 Issue 7 Pages 589-591
Published: July 30, 1983
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSDownload PDF (339K) -
Takashi SHIGEMASA, Yasuo TAKAGI, Yoshinori ICHIKAWA, Toshiyuki KITAMOR ...1983 Volume 19 Issue 7 Pages 592-594
Published: July 30, 1983
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSDownload PDF (368K) -
Naoki MIZUNO, Seizo FUJII1983 Volume 19 Issue 7 Pages 595-597
Published: July 30, 1983
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSDownload PDF (373K)
- |<
- <
- 1
- >
- >|