All issues
Volume 19, Issue 4
Displaying 1-17 of 17 articles from this issue
- |<
- <
- 1
- >
- >|
-
Shape Optimization of Internally Finned Tube of Heat Exchanger Using FEMYasuo TSUKAMOTO1983 Volume 19 Issue 4 Pages 279-285
Published: April 30, 1983
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSA continuum structure in cooperating physical fields is described by a distributed parameter system using partial differential equations. A shape optimization problem of the structure is usually modelled by minimizing an objective function with shape parameters subject to the governing field equations and several constraints. The problem can be reformulated as a variational problem introducing their adjoint system. The two or three dimensional problem is essentially nonlinear with respect to the shape parameters, even if the governing equations are linear to the fields. It is therefore difficult to solve the problem analytically.
In the presented paper, an iterative optimization procedure is proposed to solve the nonlinear variational problem numerically with the aid of the finite element discretization technique combined with the multiplier method. The procedure is applied, for a numerical example, to a shape optimization problem of an internally finned tube, which is used to improve the heat transfer efficiency. The results show that the proposed scheme can readily be implemented to obtain the optimal shape under several cooperating physical fields.View full abstractDownload PDF (964K) -
Junji NAKANO, Shigemi TAGAMI1983 Volume 19 Issue 4 Pages 286-292
Published: April 30, 1983
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper describes the procedure of testing the presence of a feedback loop for a scalar discrete dynamic stochastic system.
Box and Jenkins proposed a stochastic model for open-loop systems. They employed a transfer function model to describe the dynamics between the input and the output, and an autoregressive moving average model to characterize the input and the disturbance. They also obtained an identification and estimation method. But their method leads to inappropriate estimates if the generating process is a closed-loop system whose input is written as a linear combination of the output. Therefore, at the first stage of the data analysis, the existence of the closed-loop should be tested.
For that purpose, we use the portmanteaulike statistic, which is constructed from the input and the output prewhitened by an autoregressive model fitting with the FPE criterion. This statistic is shown to be asymptotically distributed according to the χ2 distribution of certain degrees of freedom. The test procedure is derived from this result. Two real data are analysed to demonstrate the usefulness of the present method.View full abstractDownload PDF (811K) -
Yasuhiko MUTOH, Kunihiko ICHIKAWA1983 Volume 19 Issue 4 Pages 293-299
Published: April 30, 1983
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn this paper, we discuss the generalized configuration of the discrete model reference adaptive control systems (MRACS) and their boundedness. In many literatures, the design of discrete MRACS can be classified into two methods; the first one uses the property of the positive realness and the second one uses the parameter identification algorithm proposed by Kudva and Narendra. In the latter case, Goodwin and et al. have proposed new algorithm which does not include an augmented error signal. We apply this algorithm to the former case and derive the unified design method without an augmented error signal. Furthermore, by extending Goodwin's argument, we give the proof of the asymptotic stability of the output error and the boundedness of the total system, which also guarantee the asymptotic stability and boundedness of all MRACS that will be derived from this generalized adaptive system.View full abstractDownload PDF (838K) -
Toshiro ONO, Haruo SHIMAOKA1983 Volume 19 Issue 4 Pages 300-307
Published: April 30, 1983
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSQuickness and precision are the important properties of industrial scale or weighing machine used in production lines, and the requirement for quickness and precision of weighing has recently become more stringent than ever.
In this paper a software-oriented approach has been presented to develop a high-speed weighing machine which meets the requirement. The basic point of the approach lies in the idea of microcomputer-assisted real-time data processing of the transient output signal from the measuring system, utilizing its dynamic mathematical model.
First, a procedure has been described to derive the various types of dynamic massestimation algorithm, which play an important role in real-time data processing of the observed signal. A special attention has been paid to the fundamental characteristics of the dynamic mass-estimation algorithm of displacement and velocity sensing type, which utilizes informations on the displacement and velocity of the goods-plate oscillation of a spring scale for real-time calculation of the mass value.
Secondly, discussions have been extended to the development of a simple but effective massestimation algorithm to be used in practice.
Finally, a microcomputer simulation study has been made to investigate the feasibility of the resulting algorithm. It turns out that the algorithm shows several desirable features suitable for microcomputer-assisted real-time data processing and is capable of being put into practice.View full abstractDownload PDF (989K) -
Mikiya KAWAMURA, Hiroe UESAKO1983 Volume 19 Issue 4 Pages 308-313
Published: April 30, 1983
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSAn experimental result is reported on the radiation error correction method for temperature measurement by the thermistor type digital thermometer which has been fabricated by our laboratory.
A principle for the radiation error correction is based on the fact that radiation error varies on each emissivity of a temperature sensing element. That is, using two points of measured temperature values obtained by separate temperature sensing elements having different emissivity and temperature value on the coordinate axis at emissivity equals to zero is extrapolated.
During our experiment, we have employed two kinds of digital type thermometers, one of which is a reflection type equipped with Gold or Aluminum vaporized temperature sensing element of lower emissivity and another black painted absorption type with the emissivity nearly equals to unity. Accurate temperature can be caluculated by an extrapolation formula using the measured emissivities by the two thermometers and a emissivity ratio between two thermometers.
Furthermore, corrected temperature value was easily obtained by making subtraction of the two thermometer readings when one of the thermometers with Gold as reflecting material was doubled the reading by adjusting the gain of Analog to Digital Converter in the thermometer.
Measurement error is resulted to be reduced to 0.2°C within the temperature range of 35 to 45°C. The former measurement without correction of this method had shown 0.5°C error.View full abstractDownload PDF (1900K) -
Tetsu NEMOTO, Tatsuo TOGAWA1983 Volume 19 Issue 4 Pages 314-318
Published: April 30, 1983
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSA pulsed-wire gas flow meter using a singaround system and a constant temperature heating method was developed for the spirometry with a wide range of flow rate.
A wire located at the upstream was heated by a pulse current and a traveled heated air was sensed by the other wire. The sensed signal triggered the next pulse current so that the system could selfoscillate. The repetitive pulse was period was determind by the traveling time of heated air from the heated-wire to sensor-wire.
The current of pulsed-wire was electronically controlled so that heating wire was kept in constant temperature.
The transducer was constructed in a rigid pipe of inside diameter of 35mm. The heated wire was tungsten wire of 5μm and placed a metel mesh screen with gauze to make flow velocity uniform.
Two wires of 5μm tungsten wire used for seneing, one was for detecting temperature compensation.
The range of flow rate was from 0.05l/s to 12l/s with 12mm wire spacing and from 0.02l/s to 0.5l/s with 4mm wire spacing.
The stability of the pulse frequency in ambient temperature air gas was within 2 percent. The influence of each gas composition (CO2, N2, O2, of 100%) was within 5 percent to the air.View full abstractDownload PDF (1227K) -
Kazuo HOTATE, Naohiko OOKUMA, Minoru HIGASHIGUCHI, Noboru NIWA1983 Volume 19 Issue 4 Pages 319-326
Published: April 30, 1983
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSA phase-nulling detection is useful for obtaining a good linearity and scalefactor stabilization in a fiber gyro. We utilize optical heterodyne technique followed by a electronic phase-nulling detection, which provides electric signal frequency proportional to the rotation rate. To reduce the output drift, the optical system of the gyro has little pathlength difference and no frequency difference in the fiber loop between counterpropagating waves. Reduction of the output noise that is due to the backward Rayleigh scattering in the fiber is achieved by direct frequency modulation of a single-mode laser diode, the optical source. The constructed experimental gyro shows a short-time resolution of about 5deg/h rotation rate for an integration time of 1sec, and a good linearity even in fast rotation.View full abstractDownload PDF (1619K) -
The Case of Simple Composite PatternsHiroaki OZAKI, Kenji YAMADA, Akira MOHRI, Masaru TAKATA1983 Volume 19 Issue 4 Pages 327-332
Published: April 30, 1983
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThe pattern recognition method which recognizes the pattern and the size of an object grasped by an artificial hand and identifies the position and the orientation of it is proposed. In this recognition method the pattern of the object is regarded as one composed of simple basic patterns and the composite pattern is recognized by expanding the recognition odmeth for simple basic patterns. This paper treats the case where the composite pattern is composed of a closed convex curve and a convex polygon. So as to recognize basic patterns the recognition method by UVD (Unit-Vector Distribution) is utilized. The experiment using a trial hand with sensors has shown the effectiveness of the proposed method in recognizing objects composed of a right circular cylinder and a right square prism.View full abstractDownload PDF (1344K) -
Masahiro KANEDA, Ichiro AKAHORI1983 Volume 19 Issue 4 Pages 333-335
Published: April 30, 1983
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSDownload PDF (324K) -
Naoya KAWASAKI, Etsujiro SHIMEMURA1983 Volume 19 Issue 4 Pages 336-338
Published: April 30, 1983
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSDownload PDF (416K) -
Takehisa KOHDA, Hiromitsu KUMAMOTO, Koichi INOUE1983 Volume 19 Issue 4 Pages 339-341
Published: April 30, 1983
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSDownload PDF (295K) -
Hiroo YAMAURA, Shoji KIYOHARA, Toyomi OHTA1983 Volume 19 Issue 4 Pages 342-344
Published: April 30, 1983
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSDownload PDF (350K) -
Trade-off with Housing AreaShinobu KITANI1983 Volume 19 Issue 4 Pages 345-347
Published: April 30, 1983
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSDownload PDF (321K) -
Hiroshi YAMADA, Hisanori UCHINO, Sukemitsu NAGASONO1983 Volume 19 Issue 4 Pages 348-350
Published: April 30, 1983
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSDownload PDF (344K) -
Manabu TAKATSU, Yoshitaka TOGARI1983 Volume 19 Issue 4 Pages 351-353
Published: April 30, 1983
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSDownload PDF (344K) -
1983 Volume 19 Issue 4 Pages 355
Published: 1983
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSDownload PDF (44K) -
1983 Volume 19 Issue 4 Pages e1
Published: 1983
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSDownload PDF (93K)
- |<
- <
- 1
- >
- >|