巻号一覧
20 巻, 2 号
選択された号の論文の15件中1～15を表示しています
- |<
- <
- 1
- >
- >|
-
山下 勝己, 谷口 経雄1984 年 20 巻 2 号 p. 93-100
発行日: 1984/02/29
公開日: 2009/03/27
ジャーナル フリーOptimal linear control laws always require the direct measurement of every state. However they are impractical, since it is difficult to measure all the state variables in many realistic systems. Then the problem is to reconstruct the unavailable states from the available outputs, using an observer.
In this paper, we design an observer, which is a state estimator with decaying error at a desired speed, by using the nonlinear transformation of A. M. Letov. The characteristic feature of the method is to introduce a performance index and determine the observer gain uniquely by minimizing it. Also it is unnecessary to give judicious assignment of eigenvalues, since the standpoint of the method is essentially different from that of eigenvalue assignment.
The proposed observer is applied to the reconstruction of the entire state variables for a single machine connected to an infinite bus, including induction motor load. Then the true responses are compared with those given by the observer and the method is found to be useful in power systems including the effects of dynamic characteristics of load.抄録全体を表示PDF形式でダウンロード (845K) -
中鉢 龍雄1984 年 20 巻 2 号 p. 101-107
発行日: 1984/02/29
公開日: 2009/03/27
ジャーナル フリーThis paper offers an approximate algorythm of multiple-input, multiple-output adaptive high speed model following system by the application of a special form of so-called high feedback gain method in the frequency domain. By this scheme, the initial following errors will decrease very rapidly, and there remain some small residual errors. This residual errors converge assymptotically if the system inputs are step inputs and if the system satisfies some conditions, and don't converge for other cases. To apply this method we need some a priori information of the plant, but the estimation errors of plant parameters are allowed to be considerably large. This system has an adaptive or so-called robust character by the application of high feedback gain method. But the construction of this system is different from the so-called model reference adaptive control system (MRACS), because the construction of this system is linear, and the MRACS usually contains some nonlinear parts. The numerical simulations of this scheme show very good results.抄録全体を表示PDF形式でダウンロード (809K) -
鈴木 隆, 新中 新二1984 年 20 巻 2 号 p. 108-113
発行日: 1984/02/29
公開日: 2009/03/27
ジャーナル フリーIt is shown in this paper that a generalized adaptive law for continuous-time model reference adaptive systems can be derived based on the Lyapunov's direct method and the positive real lemma, which corresponds to the discrete-time generalized adaptive law proposed by Landau et al., and that a variety of concrete algorithms, some of which are new, can be obtained by forcing some constraints to the adaptive law. The convergence properties of the algorithms, such as decreasing-gain algorithm, constant-trace algorithm are also examined by computer simulation, which exhibits that their properties are superior to those of the conventional algorithms.抄録全体を表示PDF形式でダウンロード (725K) -
石原 正, 有賀 信雄, 阿部 健一, 竹田 宏1984 年 20 巻 2 号 p. 114-121
発行日: 1984/02/29
公開日: 2009/03/27
ジャーナル フリーAdaptive control of systems described by an ARX (autoregressive system with exogenous input) model with randomly varying coefficients and d delays in the input is considered. The variation of each coefficient in the model is assumed to consist of white noise process and coloured noise process modelled as the output of a linear shaping filter. The optimal estimation of the state of the shaping filter is possible by using the Kalman filter algorithm. Two simple optimal adaptive control algorithms are derived. One is an algorithm obtained by determining the control input so that the d step ahead optimal prediction of the output coincides with a reference output, and the other is obtained by minimizing a quadratic cost function consisting of the d step ahead output error and the control input at each time step. It is shown that these optimal adaptive controls require the moments of the a posteriori probability distribution of the state of the shaping filter up to d and d+1 respectively. By omitting the higher moments from the optimal algorithms, suboptimal algorithms are obtained. Simulation results are presented to compare the performance of the optimal and the suboptimal algorithms with that of the self tuning regulator of Åström and Wittenmark.抄録全体を表示PDF形式でダウンロード (980K) -
長谷川 健介, 高橋 宏治, 増田 良介, 大野 秀嶺1984 年 20 巻 2 号 p. 122-129
発行日: 1984/02/29
公開日: 2009/03/27
ジャーナル フリーThe authors propose the Mark Flow Graph for the purpose of theoretical systematization of the sequential control.
Sequential control systems are regarded as the event-conditon systems which are activated by mutual connections between events and conditions. These systems have such fundamental features as asynchronism, ordering, concurrency and conflict, and unsafeness and deadlock will happen to them. Petri net is one of the effective means to represent the systems with these properties and phenomena.
The authors modify the concept of Petri net to develop the design and analysis tool for sequential control systems taking the safeness and notation of input/output functions into consideration. Mark Flow Graph (MFG) is proposed as such a new kind of graph. The paper deals mainly with the definitions and rules of MFG, and also with the basic subjects about its structure, structual properties and deadlock phenomenon.抄録全体を表示PDF形式でダウンロード (980K) -
吉田 要1984 年 20 巻 2 号 p. 130-136
発行日: 1984/02/29
公開日: 2009/03/27
ジャーナル フリーThis paper concerns with an approximate optimization technique of geometric programming (GP). The optimal solution is obtained by solving the dual problem which is derived from the given primal problem. And the calculation volume to solve the dual problem is measured by the term called “degree of difficulty”.
Until now, we already proposed a method to calculate an approximate optimum of GP problem. But, the approximate optimal solution corresponding to the approximate optimum is not obtained.
In this paper, next two results are shown, namely, i) the construction method of an approximate optimal solution and ii) an inequality which can be used to estimate the difference between the approximate optimum and the objective function value at the approximate optimal solution. This inequalty includes the decomposition gain, which is introduced here and can be used to judge whether the approximate optimal solution is good or not. At last some numerical examples are shown to illustrate the above results.抄録全体を表示PDF形式でダウンロード (784K) -
吉川 恒夫, 杉江 俊治1984 年 20 巻 2 号 p. 137-142
発行日: 1984/02/29
公開日: 2009/03/27
ジャーナル フリーIn this paper we study robust servo problems considering dynamics of sensors which measure plant outputs. Firstly we derive a necessary and sufficient condition for the system with sensor to be a robust servo system which achieves asymptotic tracking of regulation under perturbations of the plant. Secondly, based on this result, we obtain a condition on plant and sensor under which robust control can be realized. Finally we give a synthesis procedure of robust servo systems with a numerical example.抄録全体を表示PDF形式でダウンロード (755K) -
倉掛 幸隆1984 年 20 巻 2 号 p. 143-148
発行日: 1984/02/29
公開日: 2009/03/27
ジャーナル フリーIn order to reduce the vibration of the system excited by the external force, the dynamic absorder is generally used. When the exciting frequency is fixed, the un-damped dynamic absorber whose the natural frequency equales to the exciting frequency is used. In this case, the amplitude of the system is reduced to zero only at that frequency. On the other hand, when the exciting frequency changes with the time over the wide frequency range, the damped dynamic absorber is used. Then the vibrating system is kept on the optimum damping condition by ajusting well the coefficients of viscus damping. The result not satisfactory compared with the former one.
If the natural frequency of the dynamic absorber can be changed in wide range and if the natural frequency can be controlled so that it can follow to the exciting frequency which is changing with the time, it is expected that the amplitude of the system reduced to zero.
Magnetic dynamic absorber is composed of two side magnets and a movable magnet (absorber mass) and its natural frequency can be changed by adjustingly the length of the distance between side magnets and the magnitude of their magnetic force.
In this paper, it is analized that the natural frequency of the magnetic dynamic absorber depends on the distance between side magnts and their magnetic coefficient. Next when the exciting frequency of the system is changed with the time, the device whose the natural frequency is controlled so that the frequency follows to the exciting one well is made and the experiment is done. The result shows that the amplitude of the system could be sufficiently reduced to less value here than in the optimum damping condition.抄録全体を表示PDF形式でダウンロード (898K) -
芝 亀吉, 小池 敏弘1984 年 20 巻 2 号 p. 149-155
発行日: 1984/02/29
公開日: 2009/03/27
ジャーナル フリーThe capillary viscometer by discharging time is one of the commonly used viscometer. However, the measuring principle of the viscometer of this type is not yet established theoretically, especially the mean effective head of the viscometer is customarilly assumed to be independent on viscosity without theoretical verification.
In this paper, considering some simplified models, it is proved theoretically that the mean effective head of a capillary viscometer depend on viscosity with exception of long enough capillary tube, even when the inlet correction of the capillary tube is assumed constant.抄録全体を表示PDF形式でダウンロード (673K) -
金住 順二1984 年 20 巻 2 号 p. 156-161
発行日: 1984/02/29
公開日: 2009/03/27
ジャーナル フリーPresent study deals with the problem of “pulsation error” in the measurement of mean flow-rate of a pulsating stream by the use of a float (area) type flowmeter. This error appears as the “deviation” of the flow-rate value indicated by the mean float-position from the true value of the mean flow-rate, in a pulsatile flow. The cause and magnitude of the “deviation” are examined theoretically based on the equation of motion of the float. A relevant experimental study is also carried out. It is found that the “deviation” in a sinusoidal pulsation of small amplitude may be expressed in a product form of the magnitude of the mean flow-rate, square of the amplitude of the flow-rate pulsation and a characteristic value which is determined by the flow Reynolds number, pulsation frequency and a damping parameter.抄録全体を表示PDF形式でダウンロード (657K) -
動物の腸管運動測定用の歪ゲージ型トランスデューサ梅谷 陽二, 伊能 教夫1984 年 20 巻 2 号 p. 162-166
発行日: 1984/02/29
公開日: 2009/03/27
ジャーナル フリーThe strain gage transducer has been widely used for measuring the force exerted in animal intestinal contraction in vivo in the field of phisiology and medicine.
The purpose of this paper is to make a practical transducer for measuring the displacement of animal intestinal constriction, proving that the conventional transducer is not suitable for detecting the force nor the displacement.
The authors show that output of the transducer does not always correspond to contracting force of the intestine, and that the intestinal movement is influenced by stiffness of the transducer as strongly as it becomes higher.
Then, after evaluating the influence of stiffness to the movement by using a mechanical transducer-intestinal muscle model, they conclude that only the displacement can be measured by this transducer, and that for such measurement the influence of stiffness can be neglected by reducing it to 1/50 (ca. 5N/m) compared with that of conventional ones.
Finally, they manufactured an improved strain gage transducer for measuring the displacement of the intestinal diameter where the stiffness of it is approximately 2.5N/m and verified by animal experiments that it can practically be used in physiology and in medicine.抄録全体を表示PDF形式でダウンロード (1759K) -
田川 善彦, 山下 忠1984 年 20 巻 2 号 p. 167-174
発行日: 1984/02/29
公開日: 2009/03/27
ジャーナル フリーThe human walking characteristics have been actively analyzed in recent years by a computer simulation. In this kind of study special considerations are required in the following aspects: (1) the model has to be carefully defined because the human body has many degrees of freedom; (2) if some experimental results are required as inputs in the simulation, the characteristics that show de finite patterns in the human walking have to be selected as inputs to obtain the simulation result similar to the human characteristics.
In our study a model was first defined to represent an average human body structure by 10 massive elements and 9 ideal joints between massive elements: the massive elements represented an upper torso with head, a lower torso, 2 arms, 2 thighs, 2 thanks, 2 feet, respectively; the physical constants of each element such as length, mass, and so on, were calculated according to the corresponding body element. The equations of motion were simplified by linearization and by neglecting terms of smaller magnitude. The motions of the point of application and of the lower torso were selected as input information in our simulation.
The simulation was carried out for two cases of massive legs and massless legs respectively. The simulation results were represented in a normalized form to compare with experimental results available from literatures. The both simulations showed the characteristics similar to those of the experimental result.抄録全体を表示PDF形式でダウンロード (950K) -
井上 昭, 岩井 善太, 真野 和生1984 年 20 巻 2 号 p. 175-177
発行日: 1984/02/29
公開日: 2009/03/27
ジャーナル フリーPDF形式でダウンロード (383K) -
松宮 恒夫, 鈴木 泰弘, 高木 亀一1984 年 20 巻 2 号 p. 178-180
発行日: 1984/02/29
公開日: 2009/03/27
ジャーナル フリーPDF形式でダウンロード (305K) -
木下 源一郎1984 年 20 巻 2 号 p. 181-183
発行日: 1984/02/29
公開日: 2009/03/27
ジャーナル フリーPDF形式でダウンロード (943K)
- |<
- <
- 1
- >
- >|