All issues
Volume 20, Issue 6
Displaying 1-17 of 17 articles from this issue
- |<
- <
- 1
- >
- >|
-
Masatoshi NAKAMURA, Yasuhiko OISHI1984 Volume 20 Issue 6 Pages 471-478
Published: June 30, 1984
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSWhen the Least Squares (LS) method is applied to parameter identification of general systems, it yields biased estimates. In order to remove this bias the Generalized Least Squares (GLS) method was proposed. But the conventional methods of the GLS have some defects. Namely, it is time consuming to transfer each input and output data into filter data every time the estimation algorithm is iterated, and it takes much time to calculate the coefficients of the error-whitening-filter.
In this paper we propose a new GLS estimation algorithm compensating for these defects. This new method has two advantages. The first is that it easily derives the coefficients of the error-whitening-filter using a simple recurrence formula. The second is an introduction of self and cross correlation functions of input and output data. If these correlation functions are calculated in advance, it is not necessary to transfer each input and output data into filter data using these correlation functions when the estimation algorithm is iterated. Because of these two advantages the calculation time for unbiased estimates can be reduced effectively.View full abstractDownload PDF (993K) -
Takami KOYANAGI, Shin-ichi NAKAJIMA1984 Volume 20 Issue 6 Pages 479-485
Published: June 30, 1984
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSOptimum design of the position control system using radiation relay is presented in this paper.
As the controlled object (a plate with uniform thickness) covers the radiation detector from one side to the other with constant velocity, the amount of incident radiation is changed. This radiation is scaled with constant scaling time and the radiation relay acts and stops the motion of the plate at the first time when the output of the scaler reaches the settling value.
The problem of calculating the mean value and variance of the position where the object plate is stopped is a kind of the first passage time problem and this calculation is troublesome and inconvenient for analysis.
In this paper the simple apploximate equations for obtaining the mean value and standard deviation are presented. These equations can be simplified using the total counts while the plate goes across the window (<nd>·D2B/U).
U: Velocity of the controlled object (cm/s).
<nd>: Mean counting rate per unit aria (counts/s cm2).
D: Length of the window of radiation detector (cm).
B: Width of the window (cm).
By these apploximate equations the optimum scaling time, optimum length of the window, and necessary mean counting rate of radiation are obtained to minimize the standard deviation of the stopping position less than the given value. The mean stopping position can be also obtained.
The limit of the application of these apploximate equations is also presented.View full abstractDownload PDF (866K) -
Toshimitsu USHIO, Kazumasa HIRAI1984 Volume 20 Issue 6 Pages 486-491
Published: June 30, 1984
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn this paper, sufficient conditions for chaos in Piecewise-linear sampled-data control systems are studied by using Shiraiwa-Kurata's Theorem.
First, it is shown that a certain piecewise-linear discrete-time control system is chaotic if the lower-dimensional subsystem has a snapback repeller.
Secondly, it is shown that, under certain conditions, a piecewise-linear sampled-data control system with two fixed points is chaotic for any sampling period which is larger than a critical sampling period.
Finaly, a chaotic region for a second-order sampled-data control system with a dead-zone element is shown.View full abstractDownload PDF (733K) -
Toshihiko ONO, Toshio KUMAMARU, Kousuke KUMAMARU1984 Volume 20 Issue 6 Pages 492-497
Published: June 30, 1984
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn this paper, a new detection and estimation method of a sensor fault, applicable to both linear static systems and linear dynamic systems, is proposed.
The method is based on vector gradient.
To the static systems, a sensor error vector, which indicates the fault of the sensor, is calculated from the outputs of the sensors and the mathematical models of the systems. And from its behavour in the vector space, the dection and the estimation of the gain parameter fault of the sensor are performed, that is, the fault detection by the magnitude of the vector and the estimation, which identifies a faulty sensor and calculates its gain deviation, by the magnitude and the vector angles.
To the dynamic systems, by transformation of state space equations a gain parameter jump is treated as a bias jump in a observer equation. The fault detection is performed by a generalized likelihood ratio test (GLR test) applied to the residual of a Kalman filter. The faulty sensor and the jump value of its gain parameter are determined by the vector gradient method same as the static systems. By combining the vector gradient method and the GLR test, the algorithm for diagnoses is simplified, keeping the excellent performance of the GLR test.
Satisfactory results are obtained at the simulation of the sensor fault in the 5th order motor drive system.View full abstractDownload PDF (641K) -
Takehisa ONISAWA, Michio SUGENO1984 Volume 20 Issue 6 Pages 498-505
Published: June 30, 1984
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper proposes an index ‘likelihood of failure’ expressed by a fuzzy set instead of failure rate used in the conventional fault tree analysis, and presents a method of fault tree analysis based on fuzzy theory.
Likelihood of failure is the grade of failure caused by a component, human error or environment. A causal relation of common mode failure is expressed in a fault tree by a fuzzy relation. We define operations corresponding to AND, OR and IMPLICATION based on triangular norm or conorm in order to analyze the fault tree. And we present four criteria to evaluate the result of the present analysis. The first is the numerical likelihood of failure which gives the greatest grade of the membership of the fuzzy set. The second is defined as the potentiality of failure. The third is defined as the fuzziness of the first criterion. The last is interpreted as the importance of a basic event.
As an illustrative example, a circuit diagram of a power supply in a nuclear power plant is shown and its fault tree is analyzed, where the top event is a loss of electrical power supply to engineering safety features. It is shown that the present method well gives us the fuzziness of likelihood of failure and the potentiality of failure.View full abstractDownload PDF (933K) -
Yoshinori HARA, Yoichi KAYA1984 Volume 20 Issue 6 Pages 506-513
Published: June 30, 1984
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSAutomatic display of a given system structure is a useful tool in the analysis of large scale systems.
This paper presents a method for drawing a graph representing the system structure, which is basically an extensive application of Tutte's method to a system consisting of components strongly connected with each other.
Steps are, firstly to extract a strongly connected graph from a given system, secondly to condense the graph and extract a proper planer graph, thirdly to locate its vertices on the two-dimensional space so as to construct a graph without any crossing of edges, and finally to complete the entire figure by adding both the condensed vertices and the deleted edges.
Application to a few system models such as a model of climatic effects of carbon dioxide and a social event map in Japan demonstrates usefulness of this method. The graphs envisaged by this method have the following characteristics.
1) A system can be expressed as a graph with straight edges in which the number of edge crossings are almost minimized.
2) A system can be easily disaggregated into sub-blocks, as the graph is drawn on the principle that the closer related vertices are, the nearer they are located to each other.
3) The process is flexible enough to allow some important cycles in the system to be located in the center of the system graph, which may help realize the entire system structure, especially in the case of closely connected social systems.View full abstractDownload PDF (1037K) -
Osami SAITO, Takaharu MORISHIMA, Yuji ITAYA1984 Volume 20 Issue 6 Pages 514-519
Published: June 30, 1984
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThe time-spatial distribution of a residential area is considered to be the main factor of urban economy and this is formed by economic activities of landloads as suppliers and renters as consumers. So their behaviors are very important for analyzing the dynamics of urban growth.
This paper considers the investment behavior of a landload who occupies a lot at distance x from central business district, assuming that the rent is increasing caused by urban growth. If he invests his money in building a rental housing, he must choose the optimal strategies, i. e., construction time, initial quality of housing and its maintenances, to maximize his profit.
First, section 2 shows that his optimal strategies are obtained as explicit forms by Kuhn-Tucker theory. Consequently the relationship between his strategy and housing market is clarified and his behavior can be explained intutively. Secondly, section 3 considers the subsidy which can control the investment of a landload. This paper asserts that the subsidy must not be paid uniformly, but depended on conditions of locations, i. e., the distance x from central business district. By using above subsidy policy, the selfgovernment body can control the spreading of residential area which is an important factor for urban environment.View full abstractDownload PDF (720K) -
Minoru HIGASHIGUCHI1984 Volume 20 Issue 6 Pages 520-524
Published: June 30, 1984
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThe quaternion representation of the vehicle attitude in 3-dimensional space has superior property in a configuration of geometrical conception and algebraic treatments. Especially, a quaternion as a number having a scalar component and 3-dimensional vector component provides very useful relation in its algebra, then the coordinate rotation of 3-dimensional orthogonal coordinate can be described as the 3-dimensional vector multiplications and additions. This vector corresponds to angle and axis of rotation that translates two 3-dimensional coordinate frame. Then the 3-dimensional transformed vector is described by vector algebraic formula using quaternion component that is conformed by original vector and rotational axis vector.
Applying this idea to the least square estimation of the vehicle attitude from observations of lines of sight, the problem is reduced to the most likelyhood estimation of the position of axis from intersectional lines of planes which bisect line of sight directional vectors that assumed as if they were placed in a 3-dimensional coordinate frame. And the attitude is given by a rotational translation about this median line of intersections between bisecting planes. In algebraically, the quaternion expressing attitude of a moving vehicle is the proper vector corresponding to the minimum proper value (nearly to 0) of a positive definite matrix whose property is also discussed. This provides new algorithm for a quick and simple estimation method of the vehicle attitude from lines of sight observations.View full abstractDownload PDF (582K) -
Kamekichi SHIBA, Hiroaki TANAKA, Masaji HARAGUCHI1984 Volume 20 Issue 6 Pages 525-528
Published: June 30, 1984
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIt is known in the previous paper that the temperature of the steady flow of water in a pipe can be measured by outer process with two Schmidt's belts.
In this paper some results of further experimental studies are reported.
The sensors used in this paper are the same as previously used.
It is considered that the temperature measurement becomes very convenient under some special conditions. We take as special conditions, the ratio (τ1-θ1)/(τ2-θ2) is equal to n1/n2, where n1-n2=1.
The results of experiments show that the temperature t°C of liquid in the pipe is obtained by this method, that is
t=τ2-n2(τ2-τ1)
where, τ1, τ2 are the temperatures of the inner surfaces of both Schmidt's belts.View full abstractDownload PDF (420K) -
Sumio KOBAYASHI, Shunji MIYAHARA1984 Volume 20 Issue 6 Pages 529-535
Published: June 30, 1984
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSWe have developed a powder flow meter, using an open-ended microwave resonator. This flow meter has the following advantages: (1) No disturbance to the powder flow is caused. (2) Both the powder velocity and concentration could be measured simultaneously. In this paper the measuring principle, the theoretical analysis of the circular waveguide resonator lined with a solid dielectric, and the results of the application are described.
The sensor part of the flow meter consists of a circular waveguide inserted in series with the powder transporting pipe line. For the selected microwave frequency, at which microwave propagates in the waveguide but not in the transporting pipes, the sensor part operates as a resonator. When powder (dielectric) is flowing through the resonator, the resonant frequency changes. The average of the frequency shift gives the concentration of the powder. As there exist several axial standing-waves in the resonator, the sensor acts as a spatial filter. Therefore the velocity of the particulates is detected using the variation of the frequency shift.
The flow meters were applied to the flow measurement of the pulverized coal injected into a blast furnace. The flow meter showed a good performance, and the accuracy of the flow meters was ±7%. (Full scale: [800kg/h])View full abstractDownload PDF (1753K) -
Masaaki ADACHI, Takahiko INARI, Mitsuhito KAMEI1984 Volume 20 Issue 6 Pages 536-541
Published: June 30, 1984
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis report is concerned with basic research for in-process monitoring of surface profile using diffuse reflection of laser light. The experiment for surfaces machined by an electrical discharge method is conducted. The dependence of diffuse intensity of a reflected laser light on the surface profile is studied by comparing it with rms roughness that is derived from spatial filtering of surface profile measured with a stylus method. The experimental results show that the diffuse intensity is approximately proportional to the rms roughness of a certain filterd surface profile. The theoretical investigation for these results is also presented.View full abstractDownload PDF (818K) -
Hiroshi HARADA, Hiroshi KASHIWAGI1984 Volume 20 Issue 6 Pages 542-547
Published: June 30, 1984
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSWhen we use a high-order M-sequence for generating pseudo-random numbers, we sometimes encounter the case where the selection of the initial n-tuple affects the statistical properties of the generated numbers. For example, if we choose as the initial n-tuple such an unsuitable n-tuple as the initial n-tuple of the characteristic M-sequence or its cyclotomic phases, the resulting numbers does not become random for a while. In these cases, if we know the phase difference between an unsuitable n-tuple and the n-tuple we use, we will see whether the n-tuple is suitable for initial n-tuple or not.
This paper proposes a new method to calculate the phase difference between any two n-tuples. Using this method, the authors obtained the phase difference between the initial n-tuple of the characteristic M-sequence and the n-tuple containing all 1 in case of the characteristic trinomials of the form f(x) =xn+xk+1. The authors show that the phase difference is obtained as a function of n and k in case where (1) k=1, (2) k=n-2l, (3) k=2l (l: ineger).View full abstractDownload PDF (650K) -
Relation between Intestinal Movements and the ContentsYoji UMETANI, Norio INOU1984 Volume 20 Issue 6 Pages 548-553
Published: June 30, 1984
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSVarious modes of movements are observed in animals' small intestines, such as propulsing or mixing the contents in them. Although we can easily supposed that the property of the contents have much effect on the modes of intestinal movements, until now there exist very few papers confirming about it.
The purpose of this paper is to clarify the relation between intestinal movements and the contents. The strain gage method was adopted for measuring the intestinal movements in vivo, improving conventional transducers and their mounting technique. Output signals from six transducers mounted on a dog's small intestine spaced at about 1cm intervals were recorded after feeding different fluid foods (minced meat, soup).
Performing numerical analysis of the recording data on a digital computer, the following results were deduced.
In case that the contents with high consistency such as minced meat enter in the intestine, the intestine shows neither periodic nor propagative movement of muscle contraction as if mixing the contents. On the other hand, in case that the contents with low consistency such as soup, it shows either periodic or propagative movement as if transporting the contents.
From these results, the authors pointed that the small intestine has a adaptive capability varying its modes of movements according to the intraluminal contents.View full abstractDownload PDF (1400K) -
Hirokimi SHINGU1984 Volume 20 Issue 6 Pages 554-560
Published: June 30, 1984
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis report describes a new design concept of control mechanism which makes it possible to remove an interaction of TDG (Dynamically Tuned Dry Gyro) without worsening the transient response characteristics.
At first, we propose a method how the removal of the interaction of TDG, which is required for the success of high accuracy mission, can be done by using subtorquers with main torquers. As theoretical analysys, the tranfer function of the loop between signal generator and subtorquer, which is called NIL (Noninteracting Control Loop), is derived so that the interaction is reduced to zero in time variant input, and in order not to exert bad influences on the response, the use of the NIL along with the DRL (Direct Axis Compensation Rebalance Loop) which is useful to the removal of the interaction in time invariant input, is designed. Next, it is examined to what degree interaction can be reduced by this method and the parameter optimization is performed by simulation. And it is clarified that the interaction can be reduced by less than one tenth at the input frequency of less than 10Hz, and the gain of frequency response of cross tranfer function decreases remarkably. Finally, judging from the analysis and simulation of this reseach, it is concluded that fundamental design concept concerning noninteracting control of TDG was established.View full abstractDownload PDF (862K) -
Hirohito OHTSUKA, Tomiji HISAMURA1984 Volume 20 Issue 6 Pages 561-563
Published: June 30, 1984
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSDownload PDF (296K) -
Eitaro AIYOSHI1984 Volume 20 Issue 6 Pages 564-566
Published: June 30, 1984
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSDownload PDF (340K) -
Yoshihiko TAGAWA, Tadashi YAMASHITA1984 Volume 20 Issue 6 Pages 567-569
Published: June 30, 1984
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSDownload PDF (266K)
- |<
- <
- 1
- >
- >|