巻号一覧
16 巻 (1980)
- 6 号 p. 783-
- 5 号 p. 615-
- 4 号 p. 469-
- 3 号 p. 303-
- 2 号 p. 147-
- 1 号 p. 1-
16 巻, 4 号
選択された号の論文の25件中1～25を表示しています
- |<
- <
- 1
- >
- >|
-
貝塚 洋, 杉本 末雄1980 年 16 巻 4 号 p. 469-476
発行日: 1980/08/30
公開日: 2009/03/27
ジャーナル フリーIn this paper, we propose a new algorithmic method to determine the maximal parts of the input and the state at each time instant which are reconstructable exactly from the output alone for both continuous and discrete linear time-invariant systems with unknown initial states. Simultaneously this method contains the practical computational method of reconstructing the parts of the input and the state from the outputs.
For these purposes, we provide two new computationally efficient algorithms. The first algorithm is the extension result of one developed by Moylan (1977) which constructs the stable inverse system. This algorithm partitions the parts of inputs and states which become to be known at each iteration step, and at the same time subtracts these parts from the original system so that the original system may be reduced to the so-called “contracted system” of it.
In continuous systems it is shown that no part of the inputs or the states in the contracted system can be determined by the knowledge of the outputs and that the reconstructable parts of the input and the state at each time instant are constant.
On the contrary it is shown in discrete systems that the reconstructable parts of the input and the state vary and monotonously increase as time increases. Then the second algorithm is developed for the purpose of reconstructing these parts, and it is applied to the contracted system.抄録全体を表示PDF形式でダウンロード (902K) -
美多 勉, 吉田 広志1980 年 16 巻 4 号 p. 477-483
発行日: 1980/08/30
公開日: 2009/03/27
ジャーナル フリーAn eigenvector assignability of the closed loop matrix A-BF is considered. The main results are the following.
(1) A new concept“Eigenvector Assignability” is proposed. And a necessary and sufficient condition on the eigenvector assignability is derived.
(2) An explicit form of F, which assigns specified poles and specified eigenvectors, is obtained.
As an application, the overshooting phenomenon of linear regulators and linear observer is investigated. And an optimal eigenvector, in the sense that all the outputs have no large overshoot when the poles are assigned far from the imaginary axis, is derived.抄録全体を表示PDF形式でダウンロード (835K) -
中溝 高好, 小林 伸明1980 年 16 巻 4 号 p. 484-490
発行日: 1980/08/30
公開日: 2009/03/27
ジャーナル フリーThe problem is considered of constructing a stable lower order whitening filter for a linear stochastic system with noise-free observation. Some applications of the filter to the control and state estimation problems are discussed.
It is first shown that a stable lower order whitening filter is just a rearranged reduced-order Kalman filter. Next, from the observations that the stable lower order inverse system may serve as a whitening filter, the condition for the state to be perfectly estimable by a physical realizable filter without pure differentiations is derived. Finally this result is applied to obtain the solution to the so-called stochastic output zeroing problem. It is shown here that this problem exactly parallels the state estimation problem in the dual sense.抄録全体を表示PDF形式でダウンロード (852K) -
馬場 順昭1980 年 16 巻 4 号 p. 491-497
発行日: 1980/08/30
公開日: 2009/03/27
ジャーナル フリーThis paper is concerned with the choice of sampling period for the minimum-time on-line closed-loop computer control of a linear process. The algorithm for this system is based on linear programming.
Generally the minimum time required to reach the desired state decreases as the sampling period T decreases. But choosing a smaller T results increases the minimum sampling number required to reach the desired state and eventually computation time and computer memories increase. On the other hand, computation time and computer memories decrease as the period T increases. But choosing a larger T increases the minimum time required to reach the desired state.
In this paper it is found that the limiting factors in the choice of the period are the computation time, the computer memories and the minimum time required to reach the desired state, and the concrete method of choosing the appropriate sampling period is presented.抄録全体を表示PDF形式でダウンロード (938K) -
石原 正, 竹田 宏1980 年 16 巻 4 号 p. 498-503
発行日: 1980/08/30
公開日: 2009/03/27
ジャーナル フリーThis paper considers the adaptive control of a special class of linear stochastic systems where unknown parameters in the driving vector are Gaussian random variables while other unknown parameters are discrete ones. First, an optimal state-parameter estimator using the parallel computation of the Kalman filters is constructed. Since the optimal control is not obtainable, an algorithm of the suboptimal control based on the well-known OLFO method is derived. In spite of the theoretical importance of the OLFO control, the resulting algorithm requires extensive on-line computation. Therefore, a simple suboptimal control which makes full use of the structure of the estimator is proposed. The performance of the proposed control is compared with that of the OLFO control by using the Monte Carlo simulation. The results show that the proposed control is not inferior to the OLFO control.抄録全体を表示PDF形式でダウンロード (757K) -
安田 一則, 平井 一正1980 年 16 巻 4 号 p. 504-510
発行日: 1980/08/30
公開日: 2009/03/27
ジャーナル フリーThis paper is concerned with the problem of stabilizing a linear large-scale system by means of state feedback.
The linear large-scale system considered consists of a number of interconnected linear time-invariant subsystems with local single inputs and is described by the equations
xi=Aixi+ΣNj=1j≠iAijxj+biui,
i=1, 2, …, N,
with xi∈Rni, ui∈R, where the matrix Ai and the vector bi are given in the controllable cannonical form.
In this paper, a sufficient condition, under which exists the local state feedback ui=kiTxi such that the overall system is asymptotically stable, is obtained. This condition for stabilization is given in terms of the interconnection structure of the system, and includes the results so far obtained by many authors.抄録全体を表示PDF形式でダウンロード (784K) -
深田 悟, 毛利 彰, 高田 勝1980 年 16 巻 4 号 p. 511-516
発行日: 1980/08/30
公開日: 2009/03/27
ジャーナル フリーThis paper presents a new approach to find the optimal feedback gains of multivariable linear control systems incorporating integral action. The control law considered here is composed of PID actions of output error and the feedback of measurement variables. The cost performance index is a quadratic type one constructed by the output error and the control error, and also it may include the time derivative of the control if necessary. Using the parameter optimization technique, the necessary conditions for the solution are given by the nonlinear simultaneous matrix equations which are solved by gradient techniques.
The solution is freed from its dependence on the set points and constant external disturbances by use of the covariance of them or the minimax solution which minimizes the worst case.
The result obtained here is easily applicable to several types of controls, e. g., a PI-control, a PID-control and so forth. An illustrative example of the minimax solution is given for a PID control.抄録全体を表示PDF形式でダウンロード (762K) -
本多 中二, 合田 周平, 田中 浩二1980 年 16 巻 4 号 p. 517-524
発行日: 1980/08/30
公開日: 2009/03/27
ジャーナル フリーThe line balancing problem of assembly processes is one decisive factor in designing an efficient assembly system. It is generally difficult to solve this problem only by using computer because it includes ambiguous and non-quantative factors. A better way of treating this problem is to clarify the respective roles of a man and a computer, and to use effectively the respective functions of them. This idea is fundamental in Computer Aided Manufacturing (CAM).
A method based on CAM is proposed in this paper to treat the line balancing problem and is named as Iterative Structural Modeling of Assembly Processes (ISMAP). ISMAP consists of two processes, one is the ISM program extracting a precedence relation among the work elements of assembly processes, and the other is the line balancing program which has generality of its own, although formulated by a simpler algorithm for mini-computer. ISMAP has been applied to an assembly line for the frames of motor cycles in this paper as a practical application. The result obtained shows that it is possible to balance the physical loads of the workers more evenly without decreasing the system efficiency of the assembly line.抄録全体を表示PDF形式でダウンロード (1158K) -
安部 正人, 鈴木 昿二, 原島 文雄1980 年 16 巻 4 号 p. 525-532
発行日: 1980/08/30
公開日: 2009/03/27
ジャーナル フリーIt has been supposed that the redistribution of empty vehicles is essential in the operation of a city car system lest the service level of the system should go down because of the random arrival of passengers at each station.
This paper deals with the simulation study on the effects of redistribution of empty vehicles on improving the service level of the system under random passenger arrival. For the simulation model, a city car system with 15 stations is assumed and the passenger arrival rate and the time distance between any two of the stations are set to be constant, and also a Poisson passenger arrival is assumed at each station.
At the first stage, the simulation without redistribution is executed to show conclusively that the service level of the system goes down owing to the random passenger arrival. The simulations with a redistribution process of several redistribution capacities are executed at the next stage. As the results of the simulation study it is pointed out that the appropriate redistribution process should be introduced in order to keep the operation of the system stationary and to have some satisfactory service level of the system under random passenger arrival. Also, as the effect of redistribution is related to the total number of the vehicles in the system, the redistribution capacity should be decided according to the total number of the vehicles in the system.抄録全体を表示PDF形式でダウンロード (1061K) -
大西 俊四郎, 秋田 鴻之助1980 年 16 巻 4 号 p. 533-538
発行日: 1980/08/30
公開日: 2009/03/27
ジャーナル フリーWhen a thyristor junction is heated above its rated temperature, many electrical characteristics of the thyristor are changed and eventually the high temperature may damage or break down the thyristor. Therefore, we have to be vigilant at keeping track of the junction temperature of the conducting thyristor. To estimate the junction temperature of the thyristor, we may apply the forward voltage drop method, or we may calculate the heat resistance, the heat losses and measure the temperature of the case. But, these methods are difficult to apply under an A. C. conduction. The gate triggering method by H. Q. Tserng is hardly applicable except for an A. C. phase control circuit. We use the gate triggering method by the saw-tooth wave voltage to measure the junction temperature and indicate its temperature digitally.
Applying a saw-tooth voltage to the gate of the thyristor of the parallel inverter, the junction temperature Tj is obtained as the function of the gate triggering voltage Vgt and the voltage between the anode and the cathode Vak By G. M. D. H., we can determine the equation to calculate Tj. And, we can indicate Tj of the thyristor in the running state by the microcomputer.
In this paper, we report the method of observing the junction temperature of the conducting thyristor of the parallel inverter using the microcomputer.抄録全体を表示PDF形式でダウンロード (659K) -
服部 晋1980 年 16 巻 4 号 p. 539-544
発行日: 1980/08/30
公開日: 2009/03/27
ジャーナル フリーThis paper deals with radiant characteristics of non-isothermal cylindrical cavities by introducing “effective temperature”.
Radiant characteristics of non-isothermal cavities are represented more simply and clearly by effective temperature than by effective emissivity. It is shown that effective temperature scarcely depends on the wavelength of radiation even for cavities whose temperatures vary considerably along the axes of cavities.
Relation between the temperature distribution and the effective temperature is investigated in detail, and an approximate formula for effective temperature and its parameters are given so that one can calculate the effective temperature of a cavity with an arbitrary temperature distribution.抄録全体を表示PDF形式でダウンロード (791K) -
仙田 修, 伊藤 隆, 蓑輪 善蔵, 渡辺 英雄1980 年 16 巻 4 号 p. 545-549
発行日: 1980/08/30
公開日: 2009/03/27
ジャーナル フリーThe hydrometer for LPG (Liquefied Petroleum Gases) is ordinarily calibrated with reference to the water density under the atmosphere pressure.
However, the hydrometers are used actually under higher pressure (about 0.8MPa) and the following two effects have to be considered to correct their indications.
1) Pressure effect-The volume of hydrometer is reduced by the pressure and consequently the immersion level is changed.
2) Buoyancy effect-The density of gas (LPG) is relatively higher than that of air. The indication of the hydrometer is affected by the buoyancy of the part of the stem emerging out of the sample liquid.
The correction values for five LPG hydrometers are experimentally determined and their indication is compared with that of other methods by measuring the densities of common samples of pure propane and butane.
In the result, the calibration accuracy of hydrometers is enhanced by correcting the indication for the pressure and buoyancy effects.抄録全体を表示PDF形式でダウンロード (616K) -
岡田 三郎, 住本 哲宏1980 年 16 巻 4 号 p. 550-555
発行日: 1980/08/30
公開日: 2009/03/27
ジャーナル フリーIn this paper, to improve the detecting accuracy of a servo-type wave gauge, first a newly developed testing apparatus of wave gauge is reported. Using this testing apparatus, the testing accuracy is shown to be ±0.01mm. By introducing a micro-computer system, the time needed to test is shortened. By recording the data automatically no personal error is conceivable.
The detecting accuracy of various shapes of surface detectors are experimentally compared. The wave level error due to the change of water conductivity varying with water temperature and water quality is calculated. Testing the compensative method of water conductivity confirmed that a successful compensation is achieved. As the result of testing the servo-type wave gauge, the detecting accuracy is improved to be within ±0.05mm.抄録全体を表示PDF形式でダウンロード (1378K) -
石井 泰1980 年 16 巻 4 号 p. 556-561
発行日: 1980/08/30
公開日: 2009/03/27
ジャーナル フリーA new method to measure the mass flow rate of fluid is described. A Pitot tube placed in the flow is vibrated along the direction of the flow with a constant frequency. A small pressure variation having the same frequency is produced at the port of the Pitot tube by the interaction of the velocity of the vibration and that of the flow. This pressure variation is picked up by a condenser microphone attached to the bottom end of the Pitot tube. Using a phase-sensitive detector, the output of the microphone is converted to a DC voltage proportional to the flow velocity multiplied by the fluid density. In order to verify the principle, preliminary experiments, using the air as the measured fluid, were done. Based on the results of the experiments, a measuring device was made and some measurements were taken, which showed the effectiveness of this measuring method.抄録全体を表示PDF形式でダウンロード (1997K) -
富永 昌治, 納谷 嘉信, 永田 昭次郎1980 年 16 巻 4 号 p. 562-567
発行日: 1980/08/30
公開日: 2009/03/27
ジャーナル フリーThe responses of measurement systems are often observed as curves. In this paper, a singular-value decomposition technique is applied to the following three problems in the analysis of response curves: (1) extraction of components or features, (2) effective representation of curves, (3) estimation of unknown curves. The analysis is based on that a response curve can be decomposed to become a linear combination of basic-component patterns, and the predominant members of these patterns provide a good approximate representation of the original response curve. The complicated estimation problem of interpolating the curves, then, can be reduced to a simple interpolation of weighting coefficients needed to combine the component patterns.
Next, the proposed method is applied to the measured curves of γ-ray response spectra. The results show that the curves of response spectra can be represented by utilizing only two principal-component patterns. Moreover, we discuss about the accuracy of the estimated spectral curves and the improved method, and illustrate the goodness of the interpolated results.抄録全体を表示PDF形式でダウンロード (843K) -
原田 耕介, 小島 佑介, 藤井 義博, 竹内 信次郎1980 年 16 巻 4 号 p. 568-572
発行日: 1980/08/30
公開日: 2009/03/27
ジャーナル フリーWe present in this paper a new and direct method of measuring the inner-states of the blast furnace. The key point of this method is that a number of magnetometers installed in the furnace detect the layers of the ores and coke through a disturbance of the magnetic field distribution. By processing the outputs from these multi-magnetometers, we can have various important clues for the control of the blast furnace, such as the falling speed and the thickness of the layers, etc. We develop the magnetometer suitable for the application, which is made of a small Mo permalloy film.抄録全体を表示PDF形式でダウンロード (548K) -
小林 史典, 坂本 泰彦, 中野 道雄1980 年 16 巻 4 号 p. 573-578
発行日: 1980/08/30
公開日: 2009/03/27
ジャーナル フリーA new configuration of phase-lock(ed) loop achieving a peerlessly rapid performance is presented in comparison with the conventional PLL's having limited responsiveness. The principle proposed in this paper makes a step further than an improvement on the existing structure; namely, a time-optimal condition is considered in terms of the input/output relationship regarding the circuit as a black box, and a PLL is devised anew realizing the condition and being free from the traditional arrangement. The outcome PLL features two distinctive properties in operation; i. e., the employment of period rather than frequency as the control variable by the use of a voltageto-period converter, and the replacement of the loop filter by time-discrete elements. A prototype comprising 10 IC's demonstrates that the time-optimal PLL can track in 2 cycles any abrupt frequency shift smaller than 1 octave and it provides an FM demodulation output free of ripples in a frequency range of 0.01Hz to 100kHz.抄録全体を表示PDF形式でダウンロード (1489K) -
李 哲榮, 寺野 寿郎1980 年 16 巻 4 号 p. 579-584
発行日: 1980/08/30
公開日: 2009/03/27
ジャーナル フリーThe story formation defined here means a process of connecting the given sentences, which are logically independent of one another, and creating a new story of which the plot is reasonable. Such a work will be very essential near future to automatic writing of scenario of abstract, but it is not easy because creation is an innate ability of a man. The method suggested here is as follows: (1) a man first makes a fuzzy graph where the verteces are the given short sentences, (2) next he finds the fuzzy labels of verteces and the fuzzy relations among the verteces, (3) according to a given strategy, the computer searches its subgraphs of which the plots are semantically reasonable.
These subgraphs corresponds to the backgrounds and episodes of the candidate stories. A new fuzzy graph is mode of these episodes and backgrounds, and the same procedures are repeated until they are converged. As the result some candidate stories are induced. Though the final selection of the candidate stories is done by the computer, the man can easily intervene in it, because a fuzzy integral is adopted here as the criterion function and its fuzzy measures are difined subjectively.
Some numerical examples show us the validity of this method.抄録全体を表示PDF形式でダウンロード (813K) -
富塚 誠義, 黒須 茂1980 年 16 巻 4 号 p. 585-590
発行日: 1980/08/30
公開日: 2009/03/27
ジャーナル フリーThis work is concerned with the extended signal quickening, a technique for manual control of higher order plants with little or no damping. It has been shown that adding first and/or higher order derivatives to the error signal and displaying the combined signal are effective in facilitating human control over such plant. This technique was first proposed by Birmingham and Taylor, and is called “Signal Quickening”. The signal quickening is further extended to incorporate future reference trajectory variation into displayed signals so as to minimize the tracking error. The design of extended signal quickening systems is based on the discrete time optimal control theory.
The experiments for a triple integrator plant with a pure time delay indicate the effectiveness of the proposed method for achieving high quality tracking. The extended signal quickening is also shown to be effective under the following realistic conditions: 1) imperfect measurement of plant state variables, 2) presence of input noise.抄録全体を表示PDF形式でダウンロード (763K) -
永井 正夫1980 年 16 巻 4 号 p. 591-596
発行日: 1980/08/30
公開日: 2009/03/27
ジャーナル フリーA vehicle magnetically levitated with attractive forces is designed to have a very narrow clearance between the electromagnets and the guideway. It is a very important concern that a collision may take place between the magnets and the guideway. To deal with this problem, we must analyze the dynamical characteristics of vehicles running over the guideway having irregular surfaces. Practically, the guideway is composed of two tracks for suspension. But in the paper so far published on this problem, the authors have dealt with only one track irregularity. Although it is geometrically and dynamically important to consider the torsion of the guideway plane surface which is caused by two track irregularities, they have overlooked it. This may be because no guideway model of multiple irregularities based on the experiment has been established.
Therefore the purpose of this paper is to describe the mathematical model of the guideway surface irregularities which are the statistical inputs of the vehicle dynamical model. Main contents are as follows. (1) The theory of a stationary gaussian random process can be related to the guideway surface irregularities, such as vertical, rolling and torsional irregularities which are all measured by the high speed instrument vehicle. (2) The statistical model with power spectral densities proposed here is reasonable and practical enough to be applied to the experimental result by V. Bormann. (3) The time series of guideway irregularities can be realized by an AR-MA model. The procedure to obtain the AR-MA model is shown, and the resulting time series is also shown to be practically efficient.抄録全体を表示PDF形式でダウンロード (804K) -
岡田 徳次1980 年 16 巻 4 号 p. 597-602
発行日: 1980/08/30
公開日: 2009/03/27
ジャーナル フリーThis paper analyzes the static motion of multijointed fingers and proposes a control procedure for the fingers to locate an object at a desired position and in a desired orientation precisely. In the analysis, each finger is assumed to have a ball-like form. Cooperative motions of two fingers which generate simultaneous finger operation are considered only in the situation that the relative positions of the grasping fingers and the grasped object are kept the same as those in the initial state. Cooperative motions are classified into the next three types according to the fingers used. 1) two fingers of Fm and Fm, 2) two fingers of Fm and Ff, 3) one finger of Fm, where Fm means the finger which has an ability to control both the position of its fingertip and the orientation of its most distal link, and Ff means the finger which has an ability to control only its position. A sequential control of 2) and 3) realizes an equivalent result of 1). Experimental results of tilting a board have shown that the control procedure obtained from the analysis makes an autonomous precise handling of an object easy.抄録全体を表示PDF形式でダウンロード (1512K) -
柴田 浩, 中沢 高照, 畑 四郎1980 年 16 巻 4 号 p. 603-605
発行日: 1980/08/30
公開日: 2009/03/27
ジャーナル フリーPDF形式でダウンロード (261K) -
森 武宏, 小亀 英己1980 年 16 巻 4 号 p. 606-607
発行日: 1980/08/30
公開日: 2009/03/27
ジャーナル フリーPDF形式でダウンロード (750K) -
伝田 健夫, 小林 精次1980 年 16 巻 4 号 p. 608-609
発行日: 1980/08/30
公開日: 2009/03/27
ジャーナル フリーPDF形式でダウンロード (207K) -
渡辺 嘉二郎, 野口 佳一, 清水 忠晴1980 年 16 巻 4 号 p. 610-611
発行日: 1980/08/30
公開日: 2009/03/27
ジャーナル フリーPDF形式でダウンロード (231K)
- |<
- <
- 1
- >
- >|