巻号一覧
26 巻, 3 号
選択された号の論文の19件中1～19を表示しています
- |<
- <
- 1
- >
- >|
-
土谷 茂樹, 鈴木 清光, 松本 昌大, 小林 彬1990 年 26 巻 3 号 p. 243-250
発行日: 1990/03/31
公開日: 2009/03/27
ジャーナル フリーThe effects of various window functions on spatial filtering characteristics of a spatial filter velocimetry are studied theoretically and experimentally.
Windows which have transmittance distributions of Gaussian, sinusoidal and triangular types are applied to the spatial filter with a periodic transmittance distribution. The relationships between the types of window functions and parameters involved inthem, and various quantities (peak value, selectivity, side lobe and dependency of the peak value on the orthogonal component of the object velocity) which characterize the spatial filtering characteristics are analyzed.
Then, it is shown that such window functions can be realized optically by combination of an object lens and two kinds of apertures and the effects of these windows on power spectrum of the output signal of the spatial filter are confirmed experimentally.抄録全体を表示PDF形式でダウンロード (1677K) -
指田 孝男, 赤羽 可奈子, 川原井 貢1990 年 26 巻 3 号 p. 251-258
発行日: 1990/03/31
公開日: 2009/03/27
ジャーナル フリーLong term reliability up to 10, 000 hours has been tested for the capacitive-type humidity sensors using cellulose acetate butyrate (CAB) film as the humidity sensitive material. When these sensors using as-received commercial unpurified CAB were tested outdoor in the thermometer screen box, their capacitance values descended abruptly within 3, 000 hours from about hundreds pF to nearly zero pF. The cause of this deterioration was found to be originated from the hydrogen-ion releasable impurity included in CAB.
Then, on the substrate type-humidity sensors, four kinds of tests, i.e., humidity cycle test, high temperature and high humidity test, and so on, were carried out in accelerated conditions.
As a result, the sensors made of the purified CAB with a water-vapor permeable Au thin film electrode on the glass substrate showed long term stability, i.e., fluctuation of their capacitance values corresponded to that of humidity less than 1% RH/1000h. It was suggested that the Au electrode acted as a filter and it protected the elements of sensor from injurious materials present in the environment.抄録全体を表示PDF形式でダウンロード (1102K) -
小山 弘, 渡辺 嘉二郎1990 年 26 巻 3 号 p. 259-265
発行日: 1990/03/31
公開日: 2009/03/27
ジャーナル フリーThis paper describes how to accurately locate and estimate a partial plugging in a gastransport pipeline by an acoustic method.
The acoustic method proposed is one of the indirect estimation methods. This method identifies the inner structure of the pipeline via the acoustic standing waves excited by noise introduced at one terminal of the pipeline.
Experiment and simulation to ensure the method were carried out. The pipe made of vinyl chloride whose length is 4.113m and whose diameter is 16.0mm (cross-sectional area was 201mm2) was employed. To avoid disturbing influences, the pipe was straight and had the uniform cross-sectional area without any connecting point. Results by experiment and simulation demonstrated very similar time responses which show the clear pulses due to a partial plugging. From the shape of pulses and the pulse occurrence times, the location and the degree of the partial plugging were calculated A partial plugging with open diameter of 15.6mm (95% cross-sectional area of the pipeline) was clearly located and estimated with reasonable accuracy.抄録全体を表示PDF形式でダウンロード (847K) -
武藤 康彦, 市川 邦彦1990 年 26 巻 3 号 p. 266-271
発行日: 1990/03/31
公開日: 2009/03/27
ジャーナル フリーThe exact model matching control (EMM) system is a simple and effective control method for a linear multivariable plant when all the parameters are known. The control parameters of EMM system is obtained through Diophantine equation. To solve this equation it is required to calculate the relatively right prime fraction form of the plant transfer matrix, an eliminant matrix, its inverse matrix, etc., which involve complicated operations of polynomial matrices. In this paper, the simple calculation method for EMM control parameters is presented using Markov parameters of the plant instead of relatively right prime fraction form of the plant transfer matrix, which does not include any operation of polynomial matrices.抄録全体を表示PDF形式でダウンロード (616K) -
山本 祥弘1990 年 26 巻 3 号 p. 272-277
発行日: 1990/03/31
公開日: 2009/03/27
ジャーナル フリーAn adaptive parameter estimation method, which can be applied for the estimation of suddenly changed unknown parameters, is proposed in this paper. The method is based on the Schmidt's Orthogonalization which is well known in Linear Algebra, and is called Orthogonal Projection Algorithm. This algorithm is remarkable in its recurrent form, and shows a minimum steps convergence for suddenly changed unknown parameters. A generalized orthogonal projection algorithm is also presented which is a simplified form of the orthogonal projection algorithm. A relationship of this algorithm and the Modified Least Square Method is presented. Finally, some simulation studies are stated in which the algorithm is applied to an adaptive control systems synthesis. It is noted that some considerations are required in applications for actual control systems.抄録全体を表示PDF形式でダウンロード (745K) -
大屋 勝敬, 西村 行雄, 米澤 洋1990 年 26 巻 3 号 p. 278-285
発行日: 1990/03/31
公開日: 2009/03/27
ジャーナル フリーThe authors propose a design method of a model reference control for nonlinear one-input, one-output systems. These systems contain such nonlinear elements as a dead-zone or a backlash at the input and the output. It is supposed that nonlinearities f(u) and g(σ), which are contained at the input and the the output, are divided into a linear part and a bounded nonlinear part such that f(u)=f1u+f(u), g(σ)=g1σ+g(σ). And for the systems, it is assumed that the all states are available directly.
The design method is considered on the coordinates transformed by a diagonal matrix T=diag[1, α-1, …, α-n+1] where α is the positive design parameter. The feedback control system is constructed so that the object may follow the reference model. The signals utilized in the feedback system are the state deviation between the system and the model and the signal g(σ)(1-ε), where g(σ) is the nonlinear characteristics of the output yP(=g(σ)) and ε(1>ε>0) is the design parameter. In the control system, feedback gains of the state deviations are selected so that eigenvalues λi (i=1, …, n) of the linear part of the system may satisfy following conditions; 1) All real part of the eigenvalues λi are negative. 2) The real part λRi and the imaginary part λIi of an eigenvalue λi satisfies |λRi|-|λIi|≥0. For the control system above mentioned, it is shown that the norm of the output error converges to any vicinity of the origin with any degree of stability by increasing α and decreasing ε. Namely, it is shown that the output error is practical stable.
Finally, computer simulation results 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 (951K) -
島村 徹也, 高橋 進一1990 年 26 巻 3 号 p. 286-292
発行日: 1990/03/31
公開日: 2009/03/27
ジャーナル フリーIn this paper a new system identification method is proposed. This method is based on representing an unknown system by a model with complex coefficients. Much variety of methods have been known which can estimate the parameters of a linear discrete-time system from the input-output records. In their methods usually the real coefficient models are assumed for the unknown system.
In the proposed method, we transform the real input-output records into the analytic signals, which are the complex signals that have no negative frequency components. And then we estimate the complex coefficient model by using the analytic signals. The model with complex coefficients have the meaning only for the positive frequency characteristic of the unknown system and the model order is reduced by a factor of two.
When the complex modeling method is applied to the instrumental variable approaches, which have been often used for the system identification, the obtained complex instrumental variable method requires slightly more computational burden than the conventional one, but considerably improves the estimation accuracy. This fact is verified by several computer simulations.抄録全体を表示PDF形式でダウンロード (930K) -
小林 伸明, 神崎 一男, 中溝 高好1990 年 26 巻 3 号 p. 293-299
発行日: 1990/03/31
公開日: 2009/03/27
ジャーナル フリーThis paper is concerned with the decoupling problem for time invariant descriptor systems. Some decoupling conditions are introduced without using additional variables. A new procedure is given for constructing the decoupling system with the descriptor variable feedback and the proper dynamic compensator. The problem is extended to make it a decoupling problem by only descriptor variable feedback, then a sufficient condition and a design procedure are given. It is also shown that under certain conditions the decoupled system may comprise the maximum and stable slow subsystem. A numerical example is included to illustrate proposed conditions and algorithms.抄録全体を表示PDF形式でダウンロード (826K) -
山本 圭治郎, 石田 堯嗣, 上田 貞靖1990 年 26 巻 3 号 p. 300-306
発行日: 1990/03/31
公開日: 2009/03/27
ジャーナル フリーNew impingement type fluidic differential amplifier functioning as a controllable gain pressure comparator has been developed. The fluidic configuration of the element is that of the summing impact modulator with short nozzle spacing and symmetrical construction, where two jets from opposing nozzles collide axially forming an impingement surface. The sensitivity of the impingement surface position to differential supply pressure can be controlled by the resistance in supply lines, also for the same supply pressure, the impingement position can be shifted by the differential resistance of the supply lines.
An optimum design based on the determination mechanism of the impingement position has been proposed. The working characteristics have been analyzed and compared with experiments.
The tested model with the opposing nozzles of diameter 2.4mm showed a rather high sensitivity. The differential pressure and flow gains were controllable up to 12 and 2.8×10-6m3/s/Pa, respectively, under the supply pressure of 2, 940Pa.抄録全体を表示PDF形式でダウンロード (800K) -
山本 圭治郎, 石田 堯嗣, 上田 貞靖1990 年 26 巻 3 号 p. 307-314
発行日: 1990/03/31
公開日: 2009/03/27
ジャーナル フリーDynamic characteristics of the newly developed impingement type differential amplifier have been analyzed and tested. The configuration of the amplifier is that of the summing impact modulator with short spacing of opposing nozzles and symmetrical construction.
In dynamic tests, sinusoidally changing supply pressure difference has been produced by a newly devised piezoclystal flapper-nozzle differential valve.
The dynamic characteristics of the supply lines, nozzles, impinging jets, flow after impingement and the output ducts of the amplifier have been analyzed by lumped parameter approximations.
Main factors which govern the dynamic characteristics were elucidated as the resonance characteristics of the supply lines and the first order lag characteristics of the output ducts.
The tested model with the opposing nozzles of diameter 2.4mm has the first resonant frequency of around 300Hz caused by the supply lines, so that it can operate up to 200Hz.抄録全体を表示PDF形式でダウンロード (987K) -
江上 正, 土谷 武士1990 年 26 巻 3 号 p. 315-322
発行日: 1990/03/31
公開日: 2009/03/27
ジャーナル フリーSystematical design method of pulse patterns of single phase PWM (Pulse Width Modulated) inverter is proposed in this paper on the basis of system control theory such as optimal preview servosystem, adaptive control system etc. Merits of the design by means of system control theory are as follows:
(1) Pulse patterns are designed with consideration on the load conditions and then, optimal pulse patterns are designed for the whole system.
(2) PWM inverter with low sensitivity for the system variations such as variations of the source voltage and load changes etc. is theoretically designed by means of the robust control theory.
(3) Output feedback control or partial state feedback control or observer application is effective when measurements of variables such as currents and/or voltages are difficult.
The first step of the procedure is the derivation of the state equations of the whole system with inverter circuit and the load. In the state equation, pulse width during one sampling period is taken to be control input variable of the state equations. In this paper, state equations for a full-bridge type inverter and a half-bridge type inverter are drived. Any control system design method can apply to the obtained state equations. Preview servo system is applied when the future desired signals are known in advance, which is possible in the case of application to UPS (Uninterruptible Power System). In this case, almost perfect generation of the desired sinusoidal voltage output is obtained by obtained PWM pulse patterns. Adaptive control system is effective for system changes or load changes.抄録全体を表示PDF形式でダウンロード (942K) -
江上 正, 土谷 武士1990 年 26 巻 3 号 p. 323-330
発行日: 1990/03/31
公開日: 2009/03/27
ジャーナル フリーOptimal digital servo system with initial value compensation and its application to design of variable speed path control of robot manipulator are proposed in this paper. Optimal servo system is applicable to the servo system with the arbitrary internal model. It is synthesized by extending type-1 optimal servo system synthesis method by means of augmented system.
Strictly speaking, optimal servo system by means of augmented system is not guaranteed to be optimal till initial values of input variables are determined. Initial value compensation for input variables is necessary to synthesize an exact optimal servo system. The initial value compensation method for optimal digital servo system is proposed.
The optimal digital repetitive control system includes a internal model of a generator of periodic signals in the closed-loop. It is a type of optimal digital servo system. In this paper, the optimal digital repetitive control system is applied to design of variable speed path control of robot manipulator. Trajectory planning method taking control ability of designed robot manipulator control system into account is combined with the repetitive control. Nonlinear compensation error and interference from the other arms are regarded as being equivalent disturbance signal and compensation action is applied by utilizing estimated value of the equivalent disturbance in discrete-time model. In simulation result, the effectiveness of proposed servo system with initial value compensation is confirmed.抄録全体を表示PDF形式でダウンロード (810K) -
則次 俊郎, 和田 力, 田寺 孝光1990 年 26 巻 3 号 p. 331-338
発行日: 1990/03/31
公開日: 2009/03/27
ジャーナル フリーRobots have been proceeded to be introduced into operations with constraints from the manipulated object such as deburring, human assisting. In these operations, not only the position but also the operating force has to be controlled, and the proper compliance is required to absorb oscillations or impact forces. A pneumatic robot with the inherent compliance due to the air compressibility can meet this requirement. However, for its practical applications, the improvement of the controllability is essential.
The hybrid control is one of methods to control both the position and the force simultaneously. The purpose of this study is to realize the hybrid control of high performance with the pneumatic robot.
To execute the hybrid control, both position and force control systems have to be constructed. Generally, in this kind of robot the force control is relatively easy, but the satisfactory position control performance is not easy to obtain. Firstly, to overcome this deficiency, a sliding mode control with a pressure feedback compensation is proposed and applied to the position control. It can stabilize the response by inhibiting the chattering being a serious problem in the usual sliding mode control, and realize the continuous path control which has been considered to be difficult for a pneumatic robot. Next, the hybrid control is implemented using this position control manner in combination with usual PI type force control. As a result, the satisfactory performance can be obtained.
Further, it is pointed out that to obtain the higher control performance some considerations for the friction are necessary.抄録全体を表示PDF形式でダウンロード (977K) -
井上 正夫1990 年 26 巻 3 号 p. 339-343
発行日: 1990/03/31
公開日: 2009/03/27
ジャーナル フリーFor a manipulator with redundant degrees of freedom, a new control algorithm that is simple enough to be installed in microprocessor is proposed. The basic idea of the algorithm is that the solution of the kinematic equation of the manipulator is derived in the iterative manner not using a pseudo-inverse but a transpose of the Jacobian matrix. The algorithm may minimize the global loss function. The analysis displays that the control law makes the joint coordinates converge to the solution that is derived by the pseudo-inverse of the Jacobian matrix. Some numerical simulations and experiments using a manipulator with 7 degrees of freedom justify the control law.抄録全体を表示PDF形式でダウンロード (578K) -
熊本 博光, 山品 元, 坂本 勇1990 年 26 巻 3 号 p. 344-351
発行日: 1990/03/31
公開日: 2009/03/27
ジャーナル フリーA recursive rulebase system is proposed to automate, under fixed and/or free boundary conditions, fault tree (FT) generations for systems with loops such as controls, return currents, and emergency shutdown features. A system for which FTs are generated is uniquely modeled as a semantic network via a manually constructed block diagram which is similar to and more general than a GO chart. Each block represents a piece of basic equipment controling a particular attribute of a flow appearing in the system. Several typical pieces are catalogued to facilitate the block diagram representation. With each piece a mini-semantic network is associated to uniquely create the system semantic network. Events are classified into flow-triple, equipment-suspected, and equipment-failure, where the flow-triple is a triple, (flow, attribute, value). Event development rules are classified into four types accordingly, and typical rules are obtained systematically from truth tables. Procedural aspects of a manual FT generation are clarified as a recursive algorithm consisting of two modules: “rule addition and removal” and “event addition and removal”. Three-valued logic is used in the algorithm to remove the added rules and the events from the FT, thus enabling FT simplifications so as to reflect boundary conditions, loops, obvious events, etc. These logic values are propagated upwards to the top event.抄録全体を表示PDF形式でダウンロード (1067K) -
平沼 賢次, 河野 通夫1990 年 26 巻 3 号 p. 352-354
発行日: 1990/03/31
公開日: 2009/03/27
ジャーナル フリーThis paper considers the block decoupling by dynamic compensation. A numerical algorithm to design the dynamic compensator is given using orthogonal transformations. This method is numerically reliable.抄録全体を表示PDF形式でダウンロード (362K) -
森 泰親, 馬 培軍, 大嶋 健司1990 年 26 巻 3 号 p. 355-357
発行日: 1990/03/31
公開日: 2009/03/27
ジャーナル フリーThis paper deals with a method of synthesizing minimum variance servo system based on internal model principle for a class of linear stochastic discrete-time systems. The control law which minimizes a quadratic performance index is derived.抄録全体を表示PDF形式でダウンロード (347K) -
田中 博, 川上 用一1990 年 26 巻 3 号 p. 358-360
発行日: 1990/03/31
公開日: 2009/03/27
ジャーナル フリーThis paper describes ground testing methods for an on-board antenna pointing control system. Gravity compensation method is suggested and its validity is confirmed by experiments. It is shown that control performances can be estimated by ground tests.抄録全体を表示PDF形式でダウンロード (343K) -
戸恒 明, 松波 道夫, 坂本 正文1990 年 26 巻 3 号 p. 361-363
発行日: 1990/03/31
公開日: 2009/03/27
ジャーナル フリーThis paper discusses the performance characteristics of a hybrid-type stepping motor with unequal stator and rotor tooth widths.
Effects of tooth width on the static torque characteristics, torque-speed characteristics, dynamic performance and step accuracy are investigated when excited with two-phase on.抄録全体を表示PDF形式でダウンロード (299K)
- |<
- <
- 1
- >
- >|