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, 6 号
選択された号の論文の23件中1～23を表示しています
- |<
- <
- 1
- >
- >|
-
西村 敏充, 狩野 弘之1980 年 16 巻 6 号 p. 783-790
発行日: 1980/12/30
公開日: 2009/03/27
ジャーナル フリーMatrix Riccati equations having constant coefficient matrices are known to have periodic solutions under certain conditions. The necessary and sufficient conditions for existence of general periodic solutions (covering non-negative definite, non-positive definite and indefinite solutions) are derived in this paper. Also the specific condition for existence of non-negative (or non-positive) definite and periodic solutions is clarified. Such conditions as well as computational procedures in deriving periodic solutions are demonstrated in the example. Self-excited oscillations inherent to matrix Riccati equations have significant implications, not only from the analytic point of view on Kalman filter and regulator problems, but also from the viewpoint of numerical integration of such equations.抄録全体を表示PDF形式でダウンロード (1034K) -
伊藤 輝生, 高田 茂夫1980 年 16 巻 6 号 p. 791-795
発行日: 1980/12/30
公開日: 2009/03/27
ジャーナル フリーFor optimization methods, penalty methods are conventional, although they are approximate methods theoretically. Futhermore the numerical solution is the solution which satisfies the necessarly conditions approximately. So it is important to show the existence of the solution and, if it exists, the error between the numerical solution and the theoretical extremal.
In this paper, the theorem is represented, with which any numerical solution can be assured of having an extremal, utilizing the interiror-point method or barrier function method. At the same time, this theorem can give the error bound of the numerical solution from the extremal. The author gave the same theorem utilizing the exterior-point method already. Thus any numerical solution can be assured of the existence of the extremal, if the solution satisfies either the conditions of the theorem in this paper or the conditions of the theorem utilizing the exterior-point method. An illustrative example is given.抄録全体を表示PDF形式でダウンロード (606K) -
神藤 久, 岩住 哲朗1980 年 16 巻 6 号 p. 796-803
発行日: 1980/12/30
公開日: 2009/03/27
ジャーナル フリーThe purpose of this paper is to obtain some sufficient conditions for the existence of suboptimal controls in the singularly perturbed linear regulator problems with multiple time scales.
The modal approach is formulated and then the general form with several small parameters is discussed.
The sufficient conditions derived here are controllability and observability conditions of the reduced and boundary layer systems.抄録全体を表示PDF形式でダウンロード (1073K) -
中森 誠一, 幡司 明1980 年 16 巻 6 号 p. 804-811
発行日: 1980/12/30
公開日: 2009/03/27
ジャーナル フリーThis paper designs the new fixed-point smoother, which estimates a stationary stochastic signal, based on the innovations theory by the covariance information of the signal and observation noise processes. Two kinds of observation noises are considered here, which are the stationary white noise and the stationary white plus stationary coloured noise. We design the sub-optimal fixed-point smoother for each noise. The new smoother has the two advantages: 1) The linear stationary stochastic signal can be estimated, with a simple initial-value system, directly by the curve-fitted functions of the covariance information, and by the observed values. So we do not have any trouble of finding an approximating state space model by the spectral factorization method, using covariance information. 2) The calculation of the algorithm for the fixed-point smoothing estimate is easy. The reason is that the covariance function of the stationary stochastic signal and coloured-noise processes are assumed to be semi-degenerate. Semi-degenerate kernels can express the general kind of covariance information in the form of finite sums of products of nonrandom functions.
The originality of this paper lies in the idea that we first derived the fixed-point smoother, which estimates a linear stationary stochastic signal, based on the innovations theory by the covariance information. In comparison with the fixed-point smoother for the stationary white noise case by the Wiener-Hopf integral equation approach using the information in nonstationary systems, the present fixed-point smoother in this paper for stationary systems also has the numerical merit that the given algorithms contain fewer number of differential equations than those by the Wiener-Hopf integral equation approach.
The effectiveness of the new fixed-point smoother was assured by some numerical examples.抄録全体を表示PDF形式でダウンロード (1160K) -
門田 良実, 有本 卓1980 年 16 巻 6 号 p. 812-817
発行日: 1980/12/30
公開日: 2009/03/27
ジャーナル フリーThis paper proposes an identification algorithm and a stability criterion for multivariate autoregressive moving average type (ARMA-type) models based on autocorrelation matrices computed from observed sequences of joint input-output processes.
It is firstly shown that the identification of ARMA-type models is closely related to the autoregression of joint input-output processes and both the computational complexity and the memory storage capacity for identifying an ARMA-type model can be reduced by using the fast recursive algorithm, the so-called Levinson-Wiggins-Robinson algorithm for joint input-output processes. Secondly a Lyapunovlike equation is derived, which is associated with more general minimization problems that govern the prediction theory of nonstationary random processes. And by employing this equation the stability of an ARMA-type model with an input white noise is examined.
In this paper the stationarity of joint input-output processes need not be assumed a priori but can be determined through the recursive procedure of the LWR algorithm抄録全体を表示PDF形式でダウンロード (764K) -
戸谷 隆美1980 年 16 巻 6 号 p. 818-823
発行日: 1980/12/30
公開日: 2009/03/27
ジャーナル フリーThe use of a pulse sequence modulated by an M-sequence as an input test signal for measuring system response is proposed, and the signal processing procedures in this method are considered.
The response function of a system subjected to a single pulse input can be obtained by the least squares method or the decoding procedure, and the moving average of this response function is obtained by the correlation method.
In the usual response measuring method using an M-sequence signal, there is a possibility that a deceptive response appears in the processed result if some nonlinear delays exist at the switching instants of the test signal. The proposed pulse sequence test signal has an advantage over the simple M-sequence test signal in that this kind of error does not occur.
The paper also shows that the above consideration suggests the structure of the multislit which can avoid the occurrence of the ghost caused by the slit-width errors in Hadamard spectroscopy.抄録全体を表示PDF形式でダウンロード (735K) -
鈴木 隆, 中村 猛見1980 年 16 巻 6 号 p. 824-830
発行日: 1980/12/30
公開日: 2009/03/27
ジャーナル フリーThis paper presents a scheme for designing an indirect model reference adaptive control system. In this scheme, the control input is adaptively synthesized by using the parameter estimates obtained by identifying the unknown plant. The identification is carried out for a nonminimal representation of the plant, which is derived by applying the state variable filter technique to the original plant equation. This makes it possible to avoid the use of plant output derivatives for synthesis of the control input. The parameters of the identification model are adjusted according to the adaptive law which is derived on the basis of Liapunov's second method. The control input necessary for guiding the plant output to coincide adaptively with the reference model output is easily realized by using the plant parameter estimates. The proposed scheme enables us to construct an adaptive control system having a simple structure, because it does not need the stabilizing signal as required in conventional direct schemes. The effectiveness of the proposed scheme has been demonstrated by the computer simulations carried out for a second order plant.抄録全体を表示PDF形式でダウンロード (858K) -
米澤 洋, 大川 不二夫1980 年 16 巻 6 号 p. 831-837
発行日: 1980/12/30
公開日: 2009/03/27
ジャーナル フリーRecently, a large number of design methods for model reference adaptive control system (MRACS) have been proposed by the researchers of stability theorem. Among them, Monopoli proposed an attractive design method for MRACS by introducing an augmented error signals and signals obtained from filters operating on the plant's input and output signals. But the MRACS with an augmented error signal has a complicated construction and algorithms due to introduction of an augmented error signal and new signals, and the filter's parameters could not be precisely determined. Furthermore, the above adaptive algorithms require the information on one plant parameter.
In this paper, a design method for discrete MRACS is proposed based on Lyapunov's direct method. It is shown that the proposed MRACS is simply constructed by the use of only adjustable elements in the feedback and feed-forward paths of a plant. For the adaptive algorithms proposed here, no information on an unknown plant parameter is required. Furthermore, the relationship between the proposed MRACS and the MRACS with an augmented error signal is discussed. Finally, the validity of the proposed adaptive algorithms is demonstrated by numerical examples.抄録全体を表示PDF形式でダウンロード (807K) -
小西 克信, 芳村 敏夫, 添田 喬1980 年 16 巻 6 号 p. 838-845
発行日: 1980/12/30
公開日: 2009/03/27
ジャーナル フリーThis paper is concerned with the asymptotic behavior of an adaptive regulator for single-input single-output linear plants described by LS model with constant parameters. The regulator is composed of three parts: a parameter estimator, a linear controller and a block which determines the controller parameters from the parameter estimates. The asymptotic behavior of the regulator is analyzed by introducing the concept of possible convergence points (p. c. points), and by assuming the unpredictable fluctuation of the estimates is to occur in the neighbourhood of the set of p. c. points. Firstly, the equations to determine the p. c. points are derived. Secondly, it is shown under some regularity conditions that if control lags exist in the plants, the p. c. points give only the true value and the regulator asymptotically converges to the optimal regulator, and that if control lag does not exist, the p. c. points give a curved line passing the true value in the space of plant parameters. Numerical examples for a second order plant are presented to illustrate the effectiveness of the proposed method.抄録全体を表示PDF形式でダウンロード (1082K) -
中西 英二, 南原 興二1980 年 16 巻 6 号 p. 846-851
発行日: 1980/12/30
公開日: 2009/03/27
ジャーナル フリーThe idea of the Smith Predictor, which was originally defined in the s-domain for single-variable time-invariant systems, is extended to multivariable time-varying systems with time delays in the control variables. A new t-domain's predictor is first devised to take the time delay element out of the feedback control loop of the single-variable time-varying system. Next the multivariable time-varying system is decomposed by using the decoupling condition into a set of single-variable systems each of which has the abovementioned predictor. A numerical example is given to explain the design procedure of such a multivariable Smith Predictor and also to demonstrate the considerably improved performance of a multivariable time-varying feedback control system having the multivariable Smith Predictor.抄録全体を表示PDF形式でダウンロード (809K) -
安田 一則, 平井 一正1980 年 16 巻 6 号 p. 852-857
発行日: 1980/12/30
公開日: 2009/03/27
ジャーナル フリーThis paper is concerned with the stabilization problem for bilinear systems by means of a linear state feedback.
A bilinear system described by the equation
x(t)=Ax(t)+∑ri=1ui(t)Bx(t)+Cu(t)
is stabilizable by using a linear state feedback u=KTx(t), if the pair (A, C) is controllable; however, it is not generally stabilizable in the large. We, in this paper, give a sufficient condition under which the bilinear system is stabilizable in the large, and estimate quantitatively the extent of a stability region around the equilibrium state in the case that the system is not stabilizable in the large. Moreover, the behavior of the solution whose initial state is in the estimated stability region is considered. It is also shown that the stability region derived here is evaluated on a ground tighter than the previous ones.抄録全体を表示PDF形式でダウンロード (683K) -
清水 初男, 吉原 稔, 渡辺 嘉二郎1980 年 16 巻 6 号 p. 858-864
発行日: 1980/12/30
公開日: 2009/03/27
ジャーナル フリーThis paper describes a computation method for time-responses of stiff, nonlinear ordinary differential equations, which are frequently adopted as models of real systems but are difficult to solve by numerical methods. Applying explicit methods to these equations does not always yield stable solutions because of the numerical instability induced by the system stiffness. On the other hand, application of implicit methods makes the computational algorithm very complicated and does not always assure numerical stability.
The new method presented here matches well with equations as such. In it, the stiff, nonlinear equation X=g(X, F) is represented as x=Ax+Bf+ue by separating g(X, F) into the stiff linear term Ax+Bf and the nonlinear term ue. The approximate solutions x0k+1 of xk+1 is obtained by applying an implicit method to the equation x=Ax+Bf+ue during the interval [kτ, (k+1)τ], where ue is at first regarded as a constant value uek. Then an implicit method is iteratively applied to compensate for x0k+1. Here, the approximate value of ue can be estimated by using the approximate solution of xk+1 at each iteration of the compensation process. The proposed iterative method solves equations of a broad stepwidth with high speed.
Finally, this paper evaluates the numerical performance, attesting the performance by computing the time-responses of a chemical process and a mass-damper nonlinear stiffness system.抄録全体を表示PDF形式でダウンロード (875K) -
中村 豊1980 年 16 巻 6 号 p. 865-871
発行日: 1980/12/30
公開日: 2009/03/27
ジャーナル フリーIn this paper, we define a new concept of integrals on fuzzy measure, that we call the extended fuzzy integrals. The extended fuzzy integrals can be represented in various ways, and then is shown to be a natural extension of Lebesgue integral and furthermore to include Sugeno's fuzzy integral as a special case. It is necessary to obtain its explicit representation, so we show two new representations of the extended fuzzy integrals.抄録全体を表示PDF形式でダウンロード (818K) -
本多 中二, 合田 周平1980 年 16 巻 6 号 p. 872-879
発行日: 1980/12/30
公開日: 2009/03/27
ジャーナル フリーThis paper describes the extraction of psychometrical characteristics of face pattern used in Faces Method-a method of representing multidimensional data by facial expressions-and their application in representing system states.
First, the configuration of similarity of the 88 faces constructed by uniform random numbers based on subjects' judgment is obtained by multidimensional scaling and the psychometrical distances between the 88 faces in the configuration are calculated. The 88 faces are subsequently classified into 15 groups by using the distances and for each group a representative facial expression is extracted. The language to interpret the classified 15 faces is examined by using 16 words expressing varying forms of emotion, e.g., ‘surprise’, ‘anger’, ‘happiness’, etc. By the method of equal appearing intervals, one dimensional scaling of the 88 faces is also obtained according to the subjects' judgment in the experiment of grading the facial pleasantness and unpleasantness. It can be inferred therefore from this experiment that the method can be used in data evaluation.
A technique of assigning data characteristics to typical facial expressions is developed using the above results. This technique is primarily to determine a mapping function of the data space to the facial expression space by GMDH. The representation of the dynamic data of one heat exchanger plant model is taken as a practical illustration of the technique. The transition of the plant from a stationary state to an abnormal state is depicted by using facial expressions showing pleasantness, normal, surprise, and anger.抄録全体を表示PDF形式でダウンロード (1213K) -
花泉 弘, 稲村 実, 豊田 弘道, 藤村 貞夫1980 年 16 巻 6 号 p. 880-885
発行日: 1980/12/30
公開日: 2009/03/27
ジャーナル フリーA new noise reduction method, which is named APR (Adaptive Peak Rejection) for MSS (Multi-Spectral Scanner) image data is developed. Although many of the signal components of image data are lost if they are processed by a spatial low pass filter or by a running mean method, the APR method loses very few signal components of the image data. For, the APR method uses the scan overlap positively, which has not been used in the usual methods.
And APR not only reduces as much random Gaussian noise as can be reduced by a running mean method using scan overlap but also rejects spiky noise almost perfectly. Moreover, APR does nothing on the image data which include no such noise, and has a simple algorithm. Thus it is concluded that APR method is most suitable to pre-process the MSS image data.
In this paper, the scan overlap and APR method are discussed. And the superiority of APR method to the running mean method using scan overlap is proved by means of actual MSS image data processing.抄録全体を表示PDF形式でダウンロード (3091K) -
渡辺 嘉二郎, 十河 定俊1980 年 16 巻 6 号 p. 886-891
発行日: 1980/12/30
公開日: 2009/03/27
ジャーナル フリーThis paper describes a method of detecting a leak in a gas-transport tube by an indirect detecting approach. In this method, the tube is regarded as an acoustic tube (like a wind instrument) and the acoustic characteristics are made use of to detect a leak location. When the tube is made up of many unit tubes connected lengthwise and when one end of the tube is excited by an acoustic white noise, the reflection coefficients and cross-section areas of the unit tubes are obtained by the partial auto-correlation of the signal measured at the other end of the tube.
The location is detected by comparing the coefficients or the areas of the unit tubes with leak to those of normal tubes. Namely, if there appears a leak in one of the unit tubes, the coefficient or the area of the unit tube will be different from that of a normal tube. Then the location is found by sensing the difference.
This paper shows a few experimental results. The experiments are performed with two kinds of tubes, under several cases of leak.
(1) The length=60.75m, the diameter=15mmφ, the acoustic damping is large and the diameters of leak are 3mmφ, 1.5mmφ and 1mmφ. The location of 3mmφ and 1.5mmφ are clearly detected, but the 1mmφ leak location is not always clearly detected.
(2) The length=6mφ, the diameter=25mmφ the acoustic damping is less and the diameters of leak are 5mmφ, 3mmφ, 1.5mmφ, 1mmφ and 0.5mmφ.
All cases of leak except 0.5mmφ are detected.
Then it becomes clear that the method can detect a leak location by an indirect detecting approach with a relatively high accuracy.抄録全体を表示PDF形式でダウンロード (2335K) -
林 叡, 宮本 義弘, 藤原 裕己, 伊藤 誠1980 年 16 巻 6 号 p. 892-897
発行日: 1980/12/30
公開日: 2009/03/27
ジャーナル フリーIn this paper, the oscillating jet flow in a slit-edge system is experimentally investigated during an edge tone operation. The experiment throws a light upon the relation between the pressure gradient and the deflection of the oscillating jet in the vicinity of the edge, upon streak lines and path lines of the jet center and upon the direction of the jet flow near the edge.
Experimental results are summerized as follows:
(1) In the vicinity of the edge, the phase angle of the pressure gradient across the oscillating jet advances about 80 degrees to that of the jet deflection independent of the slit-edge distance and the mean velocity of the jet.
(2) The pressure force acting on the jet depends on not only the jet deflection but also the flow direction of the jet in the vicinity of the edge.
(3) The oscillating pressures in the same side of the jet center line are in phase, but those of the opposite side are out of phase.抄録全体を表示PDF形式でダウンロード (1219K) -
山本 巖1980 年 16 巻 6 号 p. 898-904
発行日: 1980/12/30
公開日: 2009/03/27
ジャーナル フリーDescribed in this paper, is a Schmitt trigger device based on a pneumatic amplifier with a positive feedback system. Two sets of bellows are located on the same axis in en enclosure and divide the enclosure into three chambers. A nozzle opens at the movable end of one of the sets of bellows and faces the movable closed end of the other set. The nozzle is pneumatically connected to a constant pressure source through a stationary orifice at the fixed end of the bellows mentioned first, and the enclosure opens by an exhaust vent to the atmosphere. The orifice, the nozzle and the closed end form a pneumatic amplifier, which is arranged to have bi-stable transfer characteristics. Input pressure is introduced in the set of bellows having the movable closed end, and the back pressure of the nozzle is used as the output of the device.
When the input pressure reaches up a threshold in the increasing period of the pressure, the nozzle contacts the movable closed end abruptly by the regenerative feedback action of the sets of bellows and the output pressure jumps to the same pressure level as the source. When the input pressure reaches down another threshold in its decreasing period, the nozzle leaves suddenly the closed end by a similar feedback action and the output pressure falls down from the source pressure level into a low pressure level.
Experiments were conducted with an apparatus made on trial. Discussions are given on how the operating conditions, such as the transfer gain of the sets of bellows, the source pressure, the arrangement of the moving parts and fluid resistance of the vent, are related to the threshold pressures.抄録全体を表示PDF形式でダウンロード (932K) -
松本 弘, 佐藤 美雄, 天日 康博, 久野 勝邦, 福島 弘一郎1980 年 16 巻 6 号 p. 905-912
発行日: 1980/12/30
公開日: 2009/03/27
ジャーナル フリーWith the recent structural change of electric power supply and demand system, designers of thermal power plants have put much emphasis on their control characteristics such as quick startup and load following in a middle load operation.
In general, fast transient operations tend to speed up life consumption of a turbine rotor metal due to thermal stresses. It is one of the most important subjects for a designer of turbine control systems to reduce the time required for startup and load following. From a viewpoint of the control theory, this kind of subject falls on the minimum-time control problem with constraints in process variables and controls. But it is extremely hard to find real time optimum controls for the processes that have a large degree of nonlinearities and many state variables.
We developed a new computer control method for a turbine with which the following functions are provided.
(1) an estimation of boiler outlet steam conditions by means of autoregressive modeles.
(2) a prediction of turbine inner steam conditions, heat transfer coefficients of rotor surface, and thermal stresses, based on the estimated steam conditions.
(3) a minimization of the time required for startup and load following by determining the desirable rate of speedup and load variation.
And we evaluated the control characteristics of the newly developed method through computer simulations. The results showed a high accuracy of stress control within allowable limits, and proved that the method will reduce the startup time by 25% and the load following time by 30% compared with a conventional off-line scheduled control method. The control system may answer a need of future power plants for safety and high reliability.抄録全体を表示PDF形式でダウンロード (1161K) -
Hot-Film 流速計による実測データ解析の方法山口 隆美, 吉川 哲夫, 吉川 昭, 菅原 基晃1980 年 16 巻 6 号 p. 913-918
発行日: 1980/12/30
公開日: 2009/03/27
ジャーナル フリーA method for analyzing the turbulence phenomena in the aorta is presented. The blood velocity measured with a hot-film anemometer in the canine ascending aorta is referred in the discussion. All the experimental data were recorded on magnetic tape and processed on a digital computer after analog-to-digital conversion for the analysis.
The turbulence velocity is defined as the difference between a sample velocity recording and the ensemble average of sample recordings. To discriminate the fluid dynamical fluctuations from physiological deviations, we used a threshold to exclude recordings with extraordinary amplitude and waveform due to the arrhythmia of the heart from the ensemble of the sample velocity recordings. Small deviations of the recordings due to respiratory changes were corrected with a coefficient which equates the time integral of each individual velocity recording to that of the ensemble average velocity. The mean turbulence intensity was calculated as the square root of the time mean of the ensemble average square of turbulent velocity during deceleration phase of the ensemble average velocity. The mean turbulence intensity defined above is shown to be related closely to the maximum of ensemble average velocity in each dog.抄録全体を表示PDF形式でダウンロード (839K) -
藤井 信夫1980 年 16 巻 6 号 p. 919-920
発行日: 1980/12/30
公開日: 2009/03/27
ジャーナル フリーPDF形式でダウンロード (269K) -
若松 秀俊1980 年 16 巻 6 号 p. 921-923
発行日: 1980/12/30
公開日: 2009/03/27
ジャーナル フリーPDF形式でダウンロード (412K) -
森 武宏, 福間 則夫, 桑原 道義1980 年 16 巻 6 号 p. 924-926
発行日: 1980/12/30
公開日: 2009/03/27
ジャーナル フリーPDF形式でダウンロード (347K)
- |<
- <
- 1
- >
- >|