All issues
Volume 14 (1978)
- Issue 6 Pages 613-
- Issue 5 Pages 479-
- Issue 4 Pages 347-
- Issue 3 Pages 217-
- Issue 2 Pages 107-
- Issue 1 Pages 1-
Volume 14, Issue 5
Displaying 1-22 of 22 articles from this issue
- |<
- <
- 1
- >
- >|
-
Takao HINAMOTO, Jiro SHIMONISHI, Sadao MAEKAWA1978 Volume 14 Issue 5 Pages 479-485
Published: October 30, 1978
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper presents a method for constructing an observer at some control station of a linear discrete-time decentralized control system. It is assumed here that there are no information exchanges among the stations and the system is observable at the considered station.
By using the algorithm proposed in this paper, a minimal-order unbiased observer is constructed while the estimation error is redused to zero with a minimal time. Also, it is clarified that this is a minimal-dimensional and minimal-time observer at the considered station.View full abstractDownload PDF (840K) -
Toshimitsu NISHIMURA, Hiroyuki KANO1978 Volume 14 Issue 5 Pages 486-493
Published: October 30, 1978
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThe matrix Riccati equation with coefficients varying periodically with time is discussed for discrete systems in this paper. The solution of such equation plays an important role in the application of control and estimation theory. The necessary and sufficient condition for the existence of a unique, non-negative definite and periodic solution as well as the asymptotical stability of the associated closed-loop system is derived and it is related to the stabilizability and detectability of the discrete system.View full abstractDownload PDF (1067K) -
Takahiko MORI, Masami ITO1978 Volume 14 Issue 5 Pages 494-498
Published: October 30, 1978
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper presents a new procedure for constructing the inverse of linear time-invariant systems described by system matrices of a state-space form using the concept of a system being properly invertible. This procedure has a great generality seen never before, and gives a new insight into the structure of inverse systems. Also, relations between various inverse systems, especially between an L-integral inverse by Sain and Massey and an inverse system by Silverman, become very clear.View full abstractDownload PDF (632K) -
Yoshifumi SUNAHARA, Akira OHSUMI, Tokuo FUKUDA1978 Volume 14 Issue 5 Pages 499-506
Published: October 30, 1978
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn this paper, a method is presented for the system modelling from noisy data, where the unknown system model is specified by an n-th order linear stochastic differential equation with time-varying coefficients. The goal of the system modelling is to estimate the system order n and to identify the unknown, time-varying coefficients.
A decision rule is first established on the notion of the multi-hypothesis testing. Secondly, a procedure to estimate simultaneously the unknown system order and the unknown coefficients is given within the framework of estimation theory. An algorithm is described for determining the system model from noisy data where the likelihood-ratio function plays a key role. Salient features of the proposed method are emphasized by two examples. One is a simulation experiment which illustrates the extent of the theoretical results obtained. The other is an application of the present method to geophysical data by which a dynamical model is established. This simulation experiment has shown a possibility of an extensive use of this method to identify earthquake waves.View full abstractDownload PDF (902K) -
Yoh YONEZAWA, Tadayoshi FURUYA, Michiko HAYASHIDA1978 Volume 14 Issue 5 Pages 507-513
Published: October 30, 1978
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIdentification of systems realized with minimal order models is considered to be important. In this paper, a new identification method for linear discrete systems is presented.
Here the linear system is first divided into a subsystem and a noise process in order to identify the process dynamics and the noise characteristics separately. By applying a 3-level M-sequence to the subsystem and by observing its output, the Markov parameters of the subsystem are estimated, and a minimal order realization is accomplished. Consequently, the pulse transfer function of the subsystem can be estimated. The minimal order is found in the pulse transfer function of the noise process using Akaike's and Furuta's methods, where the autocorrelation functions of the noise process are introduced by the resultant pulse transfer function of the system.
Numerical examples confirm the adequacy of the proposed identification algorithm, and suggest that it is effective to use input signals having different SN ratios to achieve an accurate identification.View full abstractDownload PDF (765K) -
Kazuo AIDA, Shin-ichi NAKAJIMA, Masayoshi SAIGO1978 Volume 14 Issue 5 Pages 514-518
Published: October 30, 1978
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn this paper an identification procedure for linear constant coefficient systems is proposed. If an appropriate mathematical model for the system to be identified is made, the parameters of the model can be determined from the functions of the exponentially weighted time integrals of the input and output of the system or from those of the correlation functions. These functions are equal to the coefficients of the Taylor series expansion of the inverse tansfer function of the system about s=α(α≥0).
This method can also be applied to comparatively complicated systems such as the closed control system, because the coefficients of the Taylor series expansion of the inverse transfer function of the system are relatively simple if compared with those of the transfer function of the system used in the modified moment method.
The efficiency of this method is demonstrated by the digital simulations for two examples.View full abstractDownload PDF (564K) -
Hiroyuki TAMURA, Tadashi KONDO1978 Volume 14 Issue 5 Pages 519-524
Published: October 30, 1978
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn this paper, a revised GMDH (Group Method of Data Handling) algorithm is developed in which we do not require to divide available data into two groups; training data and checking data. In this algorithm, all the data can be used not only as the training data but as the checking data, that is, the Prediction Sum of Squares (PSS) calculated from all the data is used as a criterion for selecting intermediate variables and for stopping the multilayered calculation. Therefore, the identified results do not depend on the heuristics of dividing the data into two groups. Furthermore, the revised GMDH developed in this paper automatically generates optimal partial polynomials in each selection layer. The revised GMDH, therefore, has flexibility much better than that of the basic GMDH in constructing a complete polynomial. The revised GMDH algorithm is applied to a simple illustrative example and compared with the result obtained by the basic GMDH algorithm.View full abstractDownload PDF (715K) -
Hirotaka NAKAYAMA1978 Volume 14 Issue 5 Pages 525-531
Published: October 30, 1978
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESS“Subjective programming” means programming based on the subjective judgment of a decision maker. This paper treats the cases under the assumption that the decision maker's preference is smooth and convex. First, several necessary and/or sufficient conditions of the optimal solutions are presented with the proof of their necessity and sufficiency. Next, the interactive coordinatewise optimization method, called briefly ICOM, for finding the optimal solution is suggested. The most prominent feature of the method is that it dose not require any complete information about the preference attitude of the decision maker but only the preference judgment based on the pairwise comparison among criteria which can be made most easily by the decision maker.View full abstractDownload PDF (997K) -
Takakazu ISHIMATSU, Akira MOHRI, Masaru TAKATA1978 Volume 14 Issue 5 Pages 532-537
Published: October 30, 1978
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn this paper we consider the optimal sampled-data control with stochastic sampling intervals. The admissible control considered here is of a zero-order hold type.
Introducing an augmented system, we obtain the optimal control law which is implemented by a linear state feedback. In case each interval has a general distribution, the feedback gain is obtained by troublesome computation. But in case the distribution is exponential, the feedback gain is obtained more easily. The most practical result is that of the steady case, where the freedback gain is constant.
A numerical example is given for a system whose sampling intervals are uniformly distributed.View full abstractDownload PDF (706K) -
Hajime AKASHI, Yoshiki TAKEUCHI1978 Volume 14 Issue 5 Pages 538-543
Published: October 30, 1978
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSAn algorithm is presented for a general class of nonlinear filtering problems. The process to be estimated is a Markov process with a piecewise-constant component taking values in a finite set of integers. A nonlinear stochastic differential equation modulated by the piecewise-constant component generates another component in a multi-dimensional Euclidean space. The observation is also nonlinear and is corrupted by an additive noise of a Brownian motion process.
The a posteriori probability and the conditional filtered estimate of another component evolve with time as shown by calculation for each possible value of the piecewise-constant component. The stochastic linearization technique is used to obtain a feasible algorithm. A numerical example is shown to illustrate the applicability of the proposed algorithm.View full abstractDownload PDF (767K) -
Hironori HIRATA, Hiroshi KOGO1978 Volume 14 Issue 5 Pages 544-549
Published: October 30, 1978
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSA new method from the viewpoint of flow is studied in order to derive the mechanism and rule of large-scale systems. As a practical application of the study, a model of the ecosystem with migration is proposed from the viewpoint of flow. This model explains:
1) How the ecosystem exhibits a hierarchy in energy per unit biomass.
2) How the density-dependent terms in population dynamics arise naturally.View full abstractDownload PDF (677K) -
Skeleton of the System and Extraction of Boundary LayersYoshiki MORIYAMA, Yoji UMETANI, Akira HARADA1978 Volume 14 Issue 5 Pages 550-556
Published: October 30, 1978
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn order to execute a project such as the construction of underwater facilities, submarine dredging equipment and so on, it is necessary to survey correctly and efficiently the wide areas of submarine topography. Although we have various kinds of prospecting instruments in use, we cannot help depending upon few experts who have geological knowledge and experiences rich enough to analyze the pictorical data of them. The present paper aims at organizing a system for discriminating the Alluvium sediment into sludge, clay, silt, sand and pebble, and finally drawing a submarine geological profile by processing the data as automatically as possible.
The methods of sonoprobe, a kind of acoustic prospecting instrument, is briefly described. Fundamental characteristics of the reflected sonic wave from the layer of the sediment are abstracted and summarized by applying useful knowledge and experiences of the experts. These characteristics are utilized to build up image processing procedures for extraction of a boundary between the layers and for recognition of the sort of the Alluvium. A total system is proposed on the block diagram, and the working principle of every element of the system is described. Algorithms for obtaining a picture of the geological profile are shown, emphasizing the methods to eliminate noise and to connect or bridge the boundary lines trimmed by pre-processing.
The authors believe this system will be widely applicable to various areas where efficient surveying of the geological topography within a wide range of the sea is indispensable.View full abstractDownload PDF (3150K) -
In Case of Vickers Indentation MeasurementKen-ichi TANAKA, Hiroshi YANO1978 Volume 14 Issue 5 Pages 557-563
Published: October 30, 1978
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThe measurement errors due to human factors are not studied enough, compared with the errors by a measurement instrument. In these measurement achieved by the observer, the errors due to the observer occupy the major portion of the total error. The research of this paper is concerned with the coincidence method in the eye-piece of micrometer for Vickers indentation measurement. The main cause for the error in this measurement is of a human factor. The human errors consist of the difference among the observers, days and repetitions. The purpose of this research is to decrease these errors. In the measurement the function of sight is most important, since the improvement of coincidence method decreases the errors.
A simulator used in the experiment simulates a microscope for measuring Vickers indentation. In the experiment five coincidence methods were tried by the simulator, and it was found that one of these methods makes the smallest error. In this method an observer watches the tip of the indentation corner and discriminates the existance of the tip when the tip is hidden by the edge of the white target. The observer can judge easily this hidden point, and this fact decreases the error.
The coincidence method is applied to the measurement by a practical Vickers hardness tester, and it is found that this method make a small error practically. This result shows that this coincidence method is useful and the simulator accurately simulates the actual measurement.View full abstractDownload PDF (967K) -
Miaki YAMAMOTO, Kozo TAMURA, Shoichiro YANAGISITA, Sakuki INAGAKI1978 Volume 14 Issue 5 Pages 564-571
Published: October 30, 1978
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSA new ultrasonic flow measurement system has been developed for sewage and waste water flowing in circular conduits with or without free surface. The system is composed of three units, an ultrasonic height gauge to measure depth of water, an ultrasonic doppler velocity meter to measure velocity at a predetermined point in the water and a flow computing unit.
The flow computing unit consists of a meter coefficient generator and a multiplier to calculate the product of the meter coefficient and the velocity.
We derived the meter coefficient as a function of water depth and geometry of the conduit from theoretical investigation and some experimental work. The validity of the derived coefficient was confirmed by some field tests.
The results obtained are summerized as follows:
(1) The meter coefficient is independent of the inclination of the flow surface.
(2) When a velocity measuring point is set at 0.1D from the bottom (D is the diameter of conduit), the measuring error due to the variation of roughness coefficient may be negligible.
(3) The results of the tests show that the accuracy of the system is within ±3% of full scale.
The above results are also applied to rectangular channel, but in this case the velocity measuring point should be set at 0.1B from the bottom (B is the width of channel).View full abstractDownload PDF (1529K) -
Takeshi HIROSE, Shuntetsu MATSUMOTO1978 Volume 14 Issue 5 Pages 572-578
Published: October 30, 1978
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper introduces the operating principle and the experimental result of a vehicular gap measurement system using a pattern recognition technique. This measurement system consists of a sign board at the back of a preceding-car, a TV camera mounted on the hood of a following-car and a processing device installed in the following-car.
This sign board has a specific pattern most unlikely seen in a highway traffic environment. Utilizing the fact that the size of the image of the specific pattern taken by the TV camera is inversely proportional to the gap between the TV camera and the sign board, the vehicular gap is measured. But the image is converted into a video signal, the pattern recognition technique is applied to this video signal in order to select the specific pattern from among other visual patterns.
The pattern recognition process is performed by the processing device. This device consists of a preprocessing circuit, a pulse spacing measuring circuit, a comparing circuit, a judging circuit and a converting circuit. These circuits operate as follows.
(1) In the preprocessing circuit, the changing points between a bright spot and a dark spot in the video signal are extrated in pulse signals.
(2) The pulse spacing measuring circuit computes the time interval between the above extracted pulse signals.
(3) The comparing circuit compares the ratio of the above pulse spacing with that to be expected from the given specific pattern.
(4) The judging circuit selects only the pulse train corresponding to that of the specific pattern.
(5) The converting circuit converts the pulse spacing of the above pulse train into the distance of the vehicular gap.
This measurement system was successfully tested on Chuodo highway and Shin-ohme street. The TV camera with a 150mm lens was used on Chuodo highway so that it was possible to measure vehicular gaps ranging from approximately 10 to 160m with a high vehicular speed up to 80km/h. And using a 50mm lens on Shin-ohme street, it was possible to measure vehicular gaps 3 to 50m long.View full abstractDownload PDF (2406K) -
Tsutomu WADA, Akira SHIMIZU, Shujiro DOHTA1978 Volume 14 Issue 5 Pages 579-584
Published: October 30, 1978
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn a few recent studies, it has been shown that the initial turbulence of the jet changes the jet structure significantly. However, the effect of this initial turbulence on the characteristics of the fluidic devices has never been made clear.
In this paper, therefore, effects of the initial turbulence of main jet on the static behaviour and the switching of a reattached jet are investigated. In order to increase the initial turbulence level of main jet, small high-velocity air jets are injected perpendicular to the main flow through the subjet pipe installed in the main jet duct. The results can be summarized as follows: 1. Increase of turbulence level leads to a greater loss of the jet centerline velocity and a greater spreading of the jet, and consequently caused a shorter reattachment distance and a smaller bubble region. 2. Effect of initial turbulence on the switching of a reattached jet could be explained by the decrease of reattachment distance and the increase of the jet width caused by the increase of turbulence level. 3. Effect of initial turbulence on the jet structure could be characterized by the velocity loss. By taking this velocity loss into the previous theoretical analysis, effects of turbulence on the static behaviour of a reattached jet could be predicted well.View full abstractDownload PDF (848K) -
Fuminori KOBAYASHI, Michio NAKANO1978 Volume 14 Issue 5 Pages 585-590
Published: October 30, 1978
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper describes a novel technique for converting frequency into analog d. c. voltage. Its principle consists in the determination of frequency by a null method utilizing the relation that the product of the frequency and the period of a signal equals unity. The closed-loop configuration inherent in null methods is effective to eliminate errors resulting from crude components, thus achieving a static linearity of ±0.02% FS for input frequencies over two decades, a temperature coefficient of 100ppm/°C, a supply voltage sensitivity of 0.01%/%, and an output voltage fluctuation of 0.01%rms. The frequencies handled by the converter range from 0.01Hz to 10kHz. Having a linear characteristic normalized to be independent of the frequency inputted, the loop yields a response time approximately identical to one period of the input signal for any frequency shift, which is beyond the reach of conventional measurements by null methods. Another prominent feature is the negligible ripples in the stationary output afforded by the holder used at the output stage. The major devices employed in the prototype are about ten ICs consisting of OP amplifiers, an analog divider, analog gates, and logic SSIs.View full abstractDownload PDF (2373K) -
Akio SUGIE, Yujiro INOUE, Hidenori KIMURA1978 Volume 14 Issue 5 Pages 591-597
Published: October 30, 1978
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn this paper, modern control theory is applied to the stabilization of the inherently unstable system of “coupled inverted pendulums”. The system consists of two carts and two pendulums. Each cart, on which one pendulum is mounted, is driven by a motor controlled independently. The two pendulums are coupled mechanically by a spring in order to make, the system multivariable. The system has two control inputs, given as the, voltages to the two servo motors driving the carts. The most remarkable distinction between our system and those inverted pendulum investigated so far is that the former is multivariable. This makes one unable to apply the classical control theory directly.
A linear quadratic optimal regulator is used to derive a stabilizing feedback gain matrix. The obtained control law is implemented via the“function observer” that makes it possible to reduce the order of the controller significantly. Actually, we constructed it by the second order system.
First, a single inverted pendulum system is stabilized and the stabilized region with respect to the initial conditions is obtained. Next, the system of coupled inverted pendulums is stabilized. For each experiment the actual response of the system is shown.View full abstractDownload PDF (1368K) -
Minoru INAMURA, Hiromichi TOYOTA, Sadao FUJIMURA1978 Volume 14 Issue 5 Pages 598-603
Published: October 30, 1978
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSArithmetic processing of remotely sensed multispectral images has been based on statistical methods so far. In the present paper, a method of the image processing is discussed from algebraic point of view.
First, spectral characteristics of energy intensity and reflectance are considered to be expressible by a vector which is called the spectral vector. Next, if (l, m) pixels of the image P are scalar, vector or tensor quantities, then P is called a scalar, vectoral or tensoral image respectively. For example, a monochrome image, a multispectral image and an exterior product image are scalar, vectoral and tensoral images respectively.
Let U and V be vectoral images. The inner product between U and V gives a scalar image W. If U and V are multispectral images, the inner product image W and the inner product operation have the following peculiarities.
1) Let U and V be normalized multispectral images in the same area at different times.
Then, W becomes a monochrome image enhancing most the time invariant region of the spectral characteristics in the area.
2) Let all pixels vlm of the V fix the same vector. In this case, the inner product operation has a property of emphasizing filter about the spectral characteristics.
On the other hand, the exterior product of the vectoral images P1, …, Pr gives an image K of which the pixels have tensor quantities of order r. So an exterior product image K is a tensoral image. If P1, …, Pr are multispectral images, the exterior product image and the exterior product operation have the following peculiarities.
1) Let P1, …, Pr be normalized multispectral images in the same area at different times.
Then, K becomes a tensoral image eliminating the time invariant region of the spectral characteristics in the area.
2) Let all pixels pjlm of the Pj(j=2, …, r) fix the same vectors. In this case, the exterior product operation has a property of eliminating filter about the spectral characteristics.View full abstractDownload PDF (921K) -
Takayoshi NAKAMIZO1978 Volume 14 Issue 5 Pages 604-605
Published: October 30, 1978
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSDownload PDF (252K) -
Tetsuo MABUCHI, Toyomi OHTA1978 Volume 14 Issue 5 Pages 606-607
Published: October 30, 1978
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSDownload PDF (307K) -
[in Japanese], [in Japanese], [in Japanese], [in Japanese]1978 Volume 14 Issue 5 Pages 608-609
Published: October 30, 1978
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSDownload PDF (269K)
- |<
- <
- 1
- >
- >|