巻号一覧
19 巻, 1 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
- >
- >|
-
大久保 重範1983 年 19 巻 1 号 p. 1-7
発行日: 1983/01/30
公開日: 2009/03/27
ジャーナル フリーA linear servo system of the present can let the output of the controlled object follow to the desired value stably in case that the desired values and disturbances belong to the output of linear time invariant system by using the internal model principle. But the desired values are limited to the output of the invariant system, and the modes of the system must be known previously. So there is the great limitation for following signals in a linear servo system of the present. On the other hand, it is possible for MRACS (model reference adaptive control system) which has developed in recent years to let the output of the controlled object follow to the output of the reference model which has an arbitrary input signal. But there is no enough countermeasure for disturbances in MRACS. In this way the advantage and the disadvantage of a linear servo system and MRACS are compensated with each other. So this paper presents the design of the linear servo system which can let the output of the controlled object follow stably to the output of the following model to have an arbitrary input signal in the case that unknown step disturbances enter into the controlled object by using advantages of a linear servo system and MRACS. In the proposed method to separate the whole system into a steady state system and a transient system, state variable filters are used for the steady state system in order to nill the stationary error, and the augmented system is set for the transient system in order to take off the influences of step disturbances and initial values by the stabilization of the augmented system.抄録全体を表示PDF形式でダウンロード (809K) -
中村 達也, 上田 実1983 年 19 巻 1 号 p. 8-14
発行日: 1983/01/30
公開日: 2009/03/27
ジャーナル フリーExtended Kalman filter is applied to the estimation of robot vehicle position with optical and wheel measurements. Measurement of wheel rotation is used to compute its nominal trajectory. A model that a contact point of a wheel with floor changes randomly is introduced to obtain a state equation with respect to the deviation. A fast slit-pattern measurement using a TV camera is formulated as an observation equation. Then, a Kalman filter and a fixed point smoother are constructed. Several experiments are conducted to evaluate the accuracy of estimated trajectory.抄録全体を表示PDF形式でダウンロード (1563K) -
田嶋 眞一, 若林 二郎1983 年 19 巻 1 号 p. 15-20
発行日: 1983/01/30
公開日: 2009/03/27
ジャーナル フリーAn estimation method is proposed of multiplex disturbances which occur in a nuclear power plant. The method is composed of two parts: (i) the identification of a simplified model of multi-input and multi-output to describe the related system response, and (ii) the design of a Kalman filter to estimate the multiplex disturbance.
Concerning the simplified model, several observed signals are firstly selected as output variables which can well represent the system response caused by the disturbances. A reduced-order model is utilized for designing the disturbance estimator. This is based on the following two considerations. The first is that the disturbance is assumed to be of a quasistatic nature. The other is based on the intuition that there exist a few dominant modes between the disturbances and the selected observed signals and that most of the non-dominant modes which remain may not affect the accuracy of the disturbance estimator.
The reduced-order model is furtherly transformed to a single-output model using a linear combination of the output signals, where the standard procedure of the structural identification is evaded. The parameters of the model thus transformed are calculated by the generalized least square method.
As for the multiplex disturbance estimator, the Kalman filtering method is applied by compromising the following three items: (a) quick response to disturbance, (b) reduction of estimation error in the presence of observation noises, and (c) the elimination of cross-interference between the disturbances to the plant and the estimates from the Kalman filter.
The effectiveness of the proposed method is verified through some computer experiments using a BWR plant simulator.抄録全体を表示PDF形式でダウンロード (660K) -
井上 昭, 岩井 善太, 佐藤 誠1983 年 19 巻 1 号 p. 21-27
発行日: 1983/01/30
公開日: 2009/03/27
ジャーナル フリーIn this report, an adaptive observer which has a parameter adjusting law with time-varying adaptive gains is considered. First, a representation of an adaptive observer based on a nonminimal representation of the system and the structure of an adjusting law with time-varying gains are shown. Then it is proved that errors of parameter identification and errors caused by state estimation converge to zero exponentially under the assumption that the system input satisfies the so-called sufficient rich condition. The speed of exponential convergence can be specified arbitrarily as a design parameter of the observer. It is also shown that the observer with the fixed unit gain gives consistent estimations when the measurement of the plant output is corrupted by noise. This fact shows the usefulness of time-varying gains; a gain smaller than the unit gain is effective if the speed of convergence is important and the gain should be changed to the unit gain when the effect of measurement noise should be rejected. The effect of introducing time-varying gains is confirmed experimentally by applying the method to the parameter identification problem of a second-order thermal system. The results of the experiment show that the effect of observation noise on identification processes is extremely reduced by the use of time-varying gains.抄録全体を表示PDF形式でダウンロード (954K) -
鈴木 隆, 新中 新二, 田中 幹也1983 年 19 巻 1 号 p. 28-35
発行日: 1983/01/30
公開日: 2009/03/27
ジャーナル フリーThe adaptive control system is generally best implemented with the digital computer, due to complexity of the controller. In this case, the system design should be done for the discrete-time model obtained from the original continuous-time plant. Such a model is very likely to yield unstable zeros, even if the original plant has stable zeros only. The model reference adaptive technique cannot be applied to such a plant, because it cancels the plant zeros with unstable controller poles. So, other adaptive techniques should be employed.
This paper discusses an adaptive pole placement technique applicable to the discrete-time plant with unstable zeros. For adaptive pole placement, it is common to use the indirect approach that consists of two steps, plant parameter identification and controller parameter determination. In the proposed scheme, the plant parameters are identified by the constant-trace algorithm that has several desired properties and the controller parameters are determined by a recursive method using the random signal. The scheme ensures that the controller parameters converge rapidly to their correct values whenever the plant input is persistently exciting. It is proved in this paper that the persistent excitation of the plant input is guaranteed by that of the external reference input.抄録全体を表示PDF形式でダウンロード (993K) -
武田 洋次, 入江 寿一, 石川 定夫, 平紗 多賀男1983 年 19 巻 1 号 p. 36-40
発行日: 1983/01/30
公開日: 2009/03/27
ジャーナル フリーThe great advance of the power electronics technics brought about a significant improvement for motor control. Especially, induction motors and commutatorless motors driven by current source inverter have many excellent characteristics on steady and dynamic state, and these driving systems are intended for practical use as ideal variable speed motors. The torque pulsation, however, is considerably large and it has bad influence upon the mechanical system. Many studies have been made in order to reduce the torque pulsation. Some intended to improve the current wave forms, others designed to make a multi-phase system. These methods are very useful to improve the torque pulsation, but there are few reports which examined the experimental results of the instantaneous torque.
Generally speaking, the motor driving system forms a high Q torsional resonator and instantaneous torque on the motor shaft contains a large sinusoidal wave component at the resonance frequency. Cosequently, it is very difficult to measure the instantaneous torque generated by the motor. Many measurement methods have been reported, but they are not sufficient for exactness and laborious.
In this paper, the equivalent circuit of the motor driving system is derived and the system is precisely represented as a second-order equation. A new method is proposed, which compensates the torsional oscillation with band rejection filter. It is proved with the experimental results that this measurement method is very useful and accurate.抄録全体を表示PDF形式でダウンロード (605K) -
小島 昇1983 年 19 巻 1 号 p. 41-47
発行日: 1983/01/30
公開日: 2009/03/27
ジャーナル フリーWhen a fluid flows in a passage, the fluid in contact with the wall has been generally treated as no slip. We can, however, recognize some cases where the slippage occurs under certain conditions.
In order to obtain the slippage coefficient, it is usually necessary to measure the pressure difference, the fluid viscosity and so on, but their measurements are generally very troublesome.
In this paper the author deals with a useful method for the slippage measurement and the slippage effects on the flow velocity distribution in a circular pipe.
The main merit of the method is to be able to obtain the amount of slippage from the pipe radius, the flow rate and the maximum flow velocity independent of the pressure difference or the fluid viscosity.
As a kind of the physical quantity to express the slippage we propose a slippage index number which is obtained by the ratio of the slippage coefficient to the square of hydraulic mean depth.
When the water flows in circular pipe, the variation of the slippage coefficient is experimentally observed for the change of the pipe radii even if inside surface conditions of pipes are same finishing, but the slippage index number does not vary and is kept to be constant. Therefore, the usage of the slippage index number is very convenient because of getting generalized slippage.抄録全体を表示PDF形式でダウンロード (774K) -
石堂 正弘, 菅 和俊, 豊田 実1983 年 19 巻 1 号 p. 48-54
発行日: 1983/01/30
公開日: 2009/03/27
ジャーナル フリーIt is tried to get the vertical temperature profile of the atmosphere by measuring the vertical infrared-radiation profile of the CO2 gas on board a rocket. While the rocket was ascending and descending through the atmosphere, the vertical profile of 15μm band infrared radiation of the CO2 gas was measured by the infrared-radiation detecting instrument. Then making an approximation that the atmosphere has a layer structure composed of 0.6km height, and using the iterative approximating method, we got the vertical temperature profile from the vertical infrared-radiation profile which was obtained in the rocket observation. The results have fairly good agreements with that of CIRA model 1972, although there are some discrepancies in details.抄録全体を表示PDF形式でダウンロード (911K) -
妹尾 誠, 出海 滋, 渡辺 孝1983 年 19 巻 1 号 p. 55-61
発行日: 1983/01/30
公開日: 2009/03/27
ジャーナル フリーA prototype monitor is developed for detection of rubbing, i.e. mechanical rotor contact with a stator, during rotations, in steam turbines and turbine generators. The monitor is designed to process, rotationally synchronized with machine rotation, acoustic signals induced by rubbing in the form of a histogram. It has the ability of improving the signal to noise ratio (S/N) by more than ten times. Verification tests, conducted for actual steam turbines, show that a weak rubbing can be detected for a low S/N.抄録全体を表示PDF形式でダウンロード (1863K) -
藤田 博之, 山口 楠雄, 鈴木 博1983 年 19 巻 1 号 p. 62-69
発行日: 1983/01/30
公開日: 2009/03/27
ジャーナル フリーOne of the most important applications of AE technology is continuous monitoring of structures and real-time estimation of their integrity. The quantitative evaluation of the monitoring performance according to applied objects and environmental conditions is necessary. The performance evaluation can determine quantitatively the trade-off between the estimation reliability of integrity and monitoring costs.
A model for the AE monitoring of structures has been developed, which is composed of submodels such as fracture, AE generation, propagation, measurement and estimation of degradation. The model is thought to be useful for the estimation of fracture behavior in structures from detected AE waves. In this paper, examples of sub-models and calculation method applied to the fatigue degradation are given. Following conclusions have been obtained from simulation results;
1) Based on the data obtained by material tests, the integrity of structures could be estimated from the relatively small number of AE events detected.
2) The upper limit of monitoring area is calculable for various AE generation patterns and stress levels.抄録全体を表示PDF形式でダウンロード (856K) -
高森 年, 杢田 淳一, 米持 政忠1983 年 19 巻 1 号 p. 70-76
発行日: 1983/01/30
公開日: 2009/03/27
ジャーナル フリーThis paper presents a propagation model of pressure reflection waves in the exhaust pipe of 2 stroke-cycle engine and a design method of the energy saving type exhaust pipe (ESTEP) by the model simulation.
The following two kinds of pressure reflection waves at the exhaust port are effective for ESTEP.
1) The positive pressure reflection wave in the stroke from scavenging port closing to exhaust port closing (SC∼EC)
2) The negative pressure reflection wave in the scavenging port (SO∼SC)
The positive pressure reflection wave is useful for preventing a flux of fuel gas and the negative pressure reflection wave for activating a flux of exhaust gas.
It has been experimentally confirmed that the model which was decided after the consideration of effects on composition of exhaust gas, temperature and bias pressure, gave a good prediction on behaviours of the pressure deflection waves.
According to the model simulation, the ESTEP having the both reflection waves has been designed and the fuel consumption experiments have been successfully done. The result shows 22% saving of fuel consumption in comparison with an exhaust pipe on market.抄録全体を表示PDF形式でダウンロード (831K) -
浅田 春比古1983 年 19 巻 1 号 p. 77-84
発行日: 1983/01/30
公開日: 2009/03/27
ジャーナル フリーA direct-drive arm is a new mechanical arm in which the shafts of articulated joints are directly coupled to the rotors of motors with high torque. Since the arm does not contain any gears and transmission mechanisms between the motors and their loads, the drive systems have no backlash, small friction and high mechanical stiffness, all of them are desirable for fast, accurate and versatile robots. First, we describe the prototype robot we have developed, and the characteristics of the arm is compared with conventional indirect-drive arms. Second, we discuss a basic feedback control for single-link drive systems in frequency domain. Third, we apply a feedforword compensation to the control of m.d.o.f. motion in order to compensate interactions among multiple links, and Coriolis, centrifugal and gravity forces. Finally, the steady-state characteristics are discussed with respect to servo stiffness and positioning accuracy. The experiments show excellent performances of the direct-dirve arm in terms of speed and accuracy. Throughout the paper comparison with indirect-dirve methods is made to contrast the advantage of the direct-drive method.抄録全体を表示PDF形式でダウンロード (1767K) -
小林 伸明, 中溝 高好1983 年 19 巻 1 号 p. 85-87
発行日: 1983/01/30
公開日: 2009/03/27
ジャーナル フリーPDF形式でダウンロード (339K) -
倒立振子の制御市川 惇信1983 年 19 巻 1 号 p. 88-90
発行日: 1983/01/30
公開日: 2009/03/27
ジャーナル フリーPDF形式でダウンロード (442K) -
大川 不二夫, 米澤 洋1983 年 19 巻 1 号 p. 91-93
発行日: 1983/01/30
公開日: 2009/03/27
ジャーナル フリーPDF形式でダウンロード (414K)
- |<
- <
- 1
- >
- >|