巻号一覧
18 巻, 6 号
選択された号の論文の18件中1～18を表示しています
- |<
- <
- 1
- >
- >|
-
中森 誠一, 幡司 明1982 年 18 巻 6 号 p. 541-548
発行日: 1982/06/30
公開日: 2009/03/27
ジャーナル フリーThis paper presents a new method which solves a linear nonstationary tracking problem. The problem is considered by minimizing a quadratic cost function. We express a solution of a state using state transition matrix. Substituting the equation for the state into the quadratic cost function of the tracking problem, a Fredholm integral equation of the second kind is derived by the variational calculus. From the integral equation new algorithm of the tracking problem is obtained.
The advantage gained by the algorithm might be explained as follows. The feedback coefficient is calculated directly by solving a matrix Riccati equation. Therefore, a number of the Riccati equation of the algorithm is less than that of previous well-known work of the LQ problem.
To show the characteristic of the presented algorithm, some digital simulation studies are performed. We apply the presented method to a constant value control of a DC motor.抄録全体を表示PDF形式でダウンロード (979K) -
砂原 善文, 相原 伸一, 小島 史男1982 年 18 巻 6 号 p. 549-555
発行日: 1982/06/30
公開日: 2009/03/27
ジャーナル フリーThe purpose of this paper is to study the optimal control problem for a stochastic distributed parameter system with white Gaussian noise processes as coefficients in dynamics and boundary conditions of the system. By introducing the stochastic integral on a Hilbert space, the mathematical model of the system is first described by a stochastic evolution equation. Secondly, the optimal control signal under the quadratic performance criteria is derived with the aid of stochastic maximum principle. Finally, the existence property of a solution to the operator Riccati equation which plays an important role in the optimal control problem is investigated.抄録全体を表示PDF形式でダウンロード (775K) -
美多 勉, 向田 昌幸1982 年 18 巻 6 号 p. 556-563
発行日: 1982/06/30
公開日: 2009/03/27
ジャーナル フリーNowadays a chance using microcomputers for the design of digital feedback control systems becomes increase. However existing digital control theory has many problems which must be solved before practical use. For example, in the existing digital control theory, the time delay arising from the computation time is not properly accounted.
In this paper a new digital control theory in which the computation time can be accounted is derived and this theory is compaired with the conventional theory in many aspects. And a new method is proposed in which the control and the estimation can be performed independently with two independent processors.
The results are interesting since they require only minor modifications of the conventional theory.抄録全体を表示PDF形式でダウンロード (828K) -
内藤 重保, 鳩山 由紀夫, 野田 淳彦1982 年 18 巻 6 号 p. 564-570
発行日: 1982/06/30
公開日: 2009/03/27
ジャーナル フリーThere have been various researches on identifying the dynamics of manual control characteristics based on the linearity assumption. However, all of these have required the additional assumption of stationarity. That is to say, manual control characteristics have been identified as time average over a certain time span.
Incidentally, the dynamics of human controller would gradually vary through learning as well as fatigue. Therefore, this variation was tried to be identified with respect to the pursuit manual tracking characteristics using discrete weighting function.
Continuous random process was used as reference input, the cutoff frequency of which is approximately equal to 0.8Hz, and CRT was used as the display. Input and output signals of the system were simultaneously sampled with an AD-converter, and processed by a fast algorithm called “a posteriori optimal method for system identification”, which we have developed recently to track variation of system dynamics.
The variation of pursuit manual tracking characteristics has been acquired as sequences of sampled weighting functions, and tendency in learning or fatigue has been shown qualitatively. This is the first result accompanied not only by a common prior certification on the identification error as for the ensemble average but also by a posteriori certification on the error of realized sample.抄録全体を表示PDF形式でダウンロード (927K) -
鹿股 昭雄, 樋口 龍雄1982 年 18 巻 6 号 p. 571-575
発行日: 1982/06/30
公開日: 2009/03/27
ジャーナル フリーThis paper proposes a microcomputer-based reliability evaluation system which injects artificial fault signals into a hardware prototype under examination and periodically monitors its state. A fault injector which consists of a simple hardware, namely shift registers, a decoder and several gates can generate fault signals whose interarrival times follow a well-known Weibull distribution.
The fault injection is occurred whenever a 20-tuple Tausworthe sequence generated from a maximum length sequence of 34 degree coincides with a pre-defined bit pattern. The bit pattern can be time-variably changed under the control of a microcomputer. As a shape and slope parameter of a Weibull distribution and a simultaneous failure rate among injected signals are adjustable, we can take a heuristic prediction as to an ability and a limitation of fault-tolerance.
Last, as an example of application, evaluation tests for a Triple-Modular-Redundant microcomputer system with resynchronization are carried out. Experimental results approximately agree with analytical results.抄録全体を表示PDF形式でダウンロード (673K) -
中村 達也, 上田 実1982 年 18 巻 6 号 p. 576-582
発行日: 1982/06/30
公開日: 2009/03/27
ジャーナル フリーThis paper presents a fast and memory saving method to estimate a measuring point by making use of vertical edge patterns. It is especially useful to locate a mobile robot's position in a room where many those patterns are observed at corners of wall and edges of objects. They are detected with horizontal angle measurement using an optical system, which consists of a TV camera and a turning mechanism.
A matching method is used to understand detected patterns, where positions of some edges are assumed to be known. The most probable measuring point is estimated from Baysian rule under the invariants conservation, similar to projective invariant, which is derived in this paper. This method is a searching process for an optimal correspondence between known edges and detected edge patterns. For each correspondence, it is possible to use analytical equations to obtain the most probable point.
Some pruning rules are introduced to deduce searching efforts and to improve reliability of the answer obtained. They are constraints on visual angles of two edges, the number of noise patterns between edges, probability value and estimated position.
An experimental result is shown to confirm effectiveness of this method.抄録全体を表示PDF形式でダウンロード (1676K) -
芝 亀吉, 田中 弘明, 原口 正次1982 年 18 巻 6 号 p. 583-588
発行日: 1982/06/30
公開日: 2009/03/27
ジャーナル フリーThe temperature of the water flow in a pipe is measured by a newly designed external thermometer with a Schmidt belt.
The thermometer is composed of the Schmidt belt of silicone rubber sheet placed in contact with the pipe surface, the thermocouples for measurement of the temperature τ°C of the contacting surface and the temperature θ°C of the external surface of the Schmidt belt, and of a heating plate.
The temperature range of our experiments is from the normal room temperature to about 100°C.
The result of the experiment shows that the temperature of the water flow in the pipe obtained by this newly developed thermometer agrees fairly well with those directly obtained by the thermoelectric thermometer immersed in the water flow for reference.抄録全体を表示PDF形式でダウンロード (653K) -
高橋 浩爾, 山下 良介, 山口 隆平1982 年 18 巻 6 号 p. 589-595
発行日: 1982/06/30
公開日: 2009/03/27
ジャーナル フリーThis paper deals with the measurement of instantaneous oil flow rate for laminar pulsating flow in a pipeline using a venturi meter.
Assuming that the flow is axisymmetric, laminar, incompressible and sinusoidal and that streamlines in the convergent tube are rectilinear, equations between flow rate and differential pressure were obtained theoretically from Navier-Stokes equations of motion and continuity equation by expressing the velocity profile in the cross section as power series of radial coordinate.
In order to satisfy the assumption mentioned above, in designing the venturi meter, the taper of the convergent tube was made much smaller than that of ordinary venturi meters.
Differential pressure was measured by a diffused type semiconductor differential pressure transducer.
Experimental results agreed fairly well with the theory. The relation between time-mean differential pressure and Reynolds number and Bode diagram of flow rate amplitude with respect to differential pressure amplitude were obtained.
Consequently, the theory developed in this paper can be considered to be accurate to obtain the approximately sinusoidal flow rate from measured differential pressure at Reynolds number up to 450, at Strouhal number up to 0.56 and at flow rate amplitude ratios up to 0.2, if one uses a venturi meter with small taper of the convergent tube.
The velocity profile in the cross section of the convergent tube was found to be quite different from that of laminar pulsating flow in a circular tube, even if convergent angle is very small.
When flow rate amplitude is less than 0.2 times average flow rate, time-mean differential pressure is scarcely dependent upon flow rate amplitude.抄録全体を表示PDF形式でダウンロード (1349K) -
小島 昇1982 年 18 巻 6 号 p. 596-601
発行日: 1982/06/30
公開日: 2009/03/27
ジャーナル フリーThe present research, in the flow velocity measurement by Laser Doppler Velocimeter (LDV), discusses a new data processing method that it is possible to obtain the flow velocity from the results using the frequency distribution of output voltages of the frequency tracker. In this case, the Doppler burst signals are produced with the only natural particles contained in fluid.
If we use the method, then the velocities of particles passing the middle point of a measuring control volume are obtained from the value at the point of maximum frequency in the frequency distribution.
Therefore, when the number of particles contained in fluid is poor or the flow velocity is low and also there is velocity gradient in the fluid flow, it is not necessary to intermix especially particles in fluid to do excellent the signal processing, and the method is successful and convenient.
Generally speaking, the measured flow velocity has a bias toward larger than practical flow velocity near the inner wall in a circular pipe in case of flow velocity measurement by LDV, but the bias is able to dissolve by the method.
The flow velocities obtained from the results of measurement at each measured points put very well on the flow velocity distribution curve, and the flow rate calculated from the flow velocity distribution agree well with the measurand of flow rate.
The adequacy of the new data processing method is described.抄録全体を表示PDF形式でダウンロード (750K) -
新宅 勝郎1982 年 18 巻 6 号 p. 602-608
発行日: 1982/06/30
公開日: 2009/03/27
ジャーナル フリーMany precise positioning techniques which utilize a high stability of a laser wavelength have been realized by various methods. But there is no little gap between these methods and the stability of the laser wavelength.
In this paper we use a method of measuring an analogous luminous intensity of a laser interference fringe which is a sin function of a corner cube (piston) displacement. We made a displacement control system which consisted of a laser interferometer and an electrohydraulic servo mechanism (double nozzle flapper valve and piston cylinder), and examined it experimentally.
In this experiment, we found that the system has a very good performance especially near holizontal axis of the sin function. And we confirmed a possibility of realization of control precision within 0.0008μm.抄録全体を表示PDF形式でダウンロード (1420K) -
中山 鳩夫, 山浦 富雄, 小林 彬1982 年 18 巻 6 号 p. 609-616
発行日: 1982/06/30
公開日: 2009/03/27
ジャーナル フリーA new spatial filter detector is described for non-contact measurement of two dimensional velocity vector … moving speed and its moving direction. The detector consists of two spatial filters realized in two parallelogram-slit arrays of photo-sensors on a monolithic silicon wafer. Each spatial filter detects a cosine component of the velocity for its own axis. Then the moving speed and the moving direction are calculated from the data on the two components.
In an experimental system, the image of a moving surface is projected on the detector surface, and each spatial filter generates a narrow band random signal of which meanfrequency is proportional to each cosine component of the velocity. Each mean-frequency is measured by a 3-level correlation type frequency counter. The moving speed and moving direction are calculated by a microprocessor for the two measured mean-frequencies.
The performance of the prototype equipment is as follows (the object speed about 900mm/S); speed measurement accuracy is 1.0%, direction measurement accuracy is 0.3°, and measurable direction range is -20°∼+20°. In addition, the measuring system is applied to non-contact measurement of zigzagging movement, and good results are obtained.抄録全体を表示PDF形式でダウンロード (2801K) -
香川 利治1982 年 18 巻 6 号 p. 617-621
発行日: 1982/06/30
公開日: 2009/03/27
ジャーナル フリーA pneumatic nozzle flapper is widely used in pneumatic control systems and is one of the important converting elements. Its transient pressure response has been investigated in several papers, but the case with small time constant has not been reported so far. This paper shows experimentally that the time constant of pneumatic nozzle flapper varies with the direction of pressure step and the position of load capacitor.
In the case with steady flow through the load capacitor, the time constant is larger in general than that of side-capacitor, and the time constant of negative pressure response is smaller than that of positive one. These differences are considered to be caused by the heat conduction between air and the capacitor wall.抄録全体を表示PDF形式でダウンロード (1014K) -
小山 紀1982 年 18 巻 6 号 p. 622-627
発行日: 1982/06/30
公開日: 2009/03/27
ジャーナル フリーThe frequency response of annular jet proximity sensors were examined by using quasisinusoidal vibration of a electro-magnetically controled plane flapper. The dynamic response of annular jet sensors was not high, that is, the corner frequencies of the tested three types of annular jet sensors were in the range of 2Hz to 9Hz.
To improve the frequency response, a compensation circuit was designed. The circuit is composed of a pneumatic coupling condenser circuit and a pressure adder. The coupling condenser circuit is made by a combination of two diaphragms, a chamber with a vent and a back pressure sensor and the pressure adder is made by three restrictors.
It was shown theoretically and experimentally that the frequency response of the annular jet sensors was improved explicitly by combining with the compensation circuit. The responses was improved up to 40Hz within the level of ±1dB in each sensor.抄録全体を表示PDF形式でダウンロード (666K) -
大前 力, 丸本 勝二, 内藤 祥太郎1982 年 18 巻 6 号 p. 628-634
発行日: 1982/06/30
公開日: 2009/03/27
ジャーナル フリーA new speed control system is described in which a microprocessor is used to regulate the speed of a DC shunt motor. A distinct feature of this speed regulating control circuit is to achieve an optimum efficiency of the DC drive comprising the power electronic converter, DC motor and control circuitis. A microprocessor based controller is developed, which suitably controls both the armature and the field currents of the DC motor with a view to minimizing the total system losses and achieving at the same time, the required load torque.
An experimental DC drive incorporating the microprocessor-based speed regulator was tested on a 15kW DC shunt motor. It was found that a high efficiency speed control was obtained both at transient and steady state operations.抄録全体を表示PDF形式でダウンロード (1502K) -
井上 紘一, 幸田 武久, 熊本 博光, 高見 勲1982 年 18 巻 6 号 p. 635-640
発行日: 1982/06/30
公開日: 2009/03/27
ジャーナル フリーAn optimal logical structure is developed for a multichannel safety monitoring system. Each channel monitors a plant state, e.g., temperature or pressure. When a state becomes abnormal, the corresponding channel sets off an alarm signal and activates protective actions. Each channel consists of one or more sensors that are exposed to two contradictory failures; one is “failed-dangerous” and the other is “failed-safe”. The problem is to obtain the optimal coherent logical structure of sensors for each channel minimizing expected damage. The optimal channel structure is proven to be the k-out-of-n: G system, and the problem is formulated in nonlinear integer programming (NLIP). The NLIP problem is then solved by the extended Lawler & Bell method. An illustrative example is given.抄録全体を表示PDF形式でダウンロード (685K) -
兼田 雅弘, 赤堀 一郎, 浜田 博1982 年 18 巻 6 号 p. 641-642
発行日: 1982/06/30
公開日: 2009/03/27
ジャーナル フリーPDF形式でダウンロード (267K) -
ケース・スタディー井上 紘一, 守安 隆, 八木 正英1982 年 18 巻 6 号 p. 643-644
発行日: 1982/06/30
公開日: 2009/03/27
ジャーナル フリーPDF形式でダウンロード (257K) -
1982 年 18 巻 6 号 p. 646
発行日: 1982年
公開日: 2009/03/27
ジャーナル フリーPDF形式でダウンロード (169K)
- |<
- <
- 1
- >
- >|