All issues
Volume 21, Issue 11
Displaying 1-18 of 18 articles from this issue
- |<
- <
- 1
- >
- >|
-
Akira MOHRI, Hong Chen CHANG1985 Volume 21 Issue 11 Pages 1137-1142
Published: November 30, 1985
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn this paper, from the viewpoint of eigenvalues and eigenvectors assignment, some considerations about improvement of transient characteristics of output feedback control systems are described.
For a controllable and observable multiinput system with n states, m inputs and l outputs, if n≤m+l-1, then almost arbitrary sets of closed-loop eigenvalues are assignable by using output feedback. Then, in general, some degrees of freedom remain in eigenvector assignment after the eigenvalues have been assigned. Therefore, transient characteristics may be improved by making use of these degrees of freedom in eigenvector assignment.
In this paper, this improvement method is described and its effectiveness is shown by the numerical examples.View full abstractDownload PDF (592K) -
Fi Hyol LEE, Kageo AKIZUKI1985 Volume 21 Issue 11 Pages 1143-1149
Published: November 30, 1985
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn recent year, Variable Structure Conrol System (VSCS) has become to be applied to control of a wide range. The notable feature of VSCS is that the sliding motion occurres on a switching surface. While in sliding mode, the system remains insensitive to parameter change and disturbances. However, when the Sliding Mode Control System receives random input, we haven't the condition of the sliding motion and a design method for an optimal switching surface for the present.
This paper deals with the sliding mode of VSCS which is subjected to white noise and state-dependent noise respectively, and the design of an optimal switching surface. First, we deduce the condition of a control gain for sliding mode in a mean sense. Secondly, we derive an equation of an optimal switching surface, which is based on a quadratic evaluation of a state, from the equivalent control method. Lastly, through digital simulation we confirm the above results and robustness of VSCS disturbed random input.View full abstractDownload PDF (740K) -
Setsuo SAGARA, Miyoichi EGUCHI, Kiyoshi WADA1985 Volume 21 Issue 11 Pages 1150-1155
Published: November 30, 1985
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn this paper, we propose an approach to choose the optimal sampling period for system identification on the basis of the covariance matrix of estimates of pales for a continuous system. This matrix can be easily obtained from the covariance matrix of poles of the corresponding discrete model making use of the simple transformation of poles between s- and z-plane. The covariance matrix of poles of discrete model is calculated from the covariance matrix resulting from the maximum likelihood estimation of parameters of the discrete model, applying the relation between roots and coefficients of the characteristic equation.
Our approach is practical and usefull since the algorithm is so simple that the optimal sampling period can be determined for general systems through several experiments with different sampling periods.View full abstractDownload PDF (720K) -
Kanya TANAKA, Takashi SUZUKI1985 Volume 21 Issue 11 Pages 1156-1162
Published: November 30, 1985
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThere are two typical schemes for constructing the continuous-time adaptive identifier based on the equation error principle. One, which is called Scheme I in this paper, employs (n-1)-th order state variable filters for an n-th order plant model and the other, called Scheme II, uses n-th order filters. This paper discusses robustness properties of these schemes in the presence of parasitics and output noise. As a consequence, the following points are clarified: 1) Scheme II is more robust than Scheme I in the sense that the steady-state variation of the identification result is small. 2) Addition of the proportional term to the integral adaptive law is effective for improving robustness properties in the above both schemes. 3) It is profitable to adopt only the integral output as the identification result in place of the sum of the integral and proportional outputs when the integral plus proportional law is used. Furthermore, a new robust scheme with higher order state variable filters is also proposed here.View full abstractDownload PDF (890K) -
Yoshihiko MIYASATO, Toshiyuki KITAMORI1985 Volume 21 Issue 11 Pages 1163-1170
Published: November 30, 1985
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSDuring the last several years there have been presented a number of results in the field of model-reference adaptive control. In the most of the results, the process is assumed to be a lumped parameter system, which is described by ordinary differential equations. But real physical systems occupy a certain spatial domain and are not concentrated at a single spatial point. Such systems are called distributed parameter systems, which are modeled by partial differential equations.
The principal contribution of this paper is to present a procedure for the design of model-reference adaptive control for an unknown distributed parameter system of parabolic type. We assume the spatial average of the states over some effective sensing region to be the output of the process, and adaptively synthesize the input so that the tracking error between the output of the process and the output of a prescribed reference model, which is a lumped parameter system, is regulated to zero asymptotically.View full abstractDownload PDF (907K) -
Kazuaki ANDO1985 Volume 21 Issue 11 Pages 1171-1176
Published: November 30, 1985
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSPadé approximants of exp(-sT) have been widely used to realize dead-time elements. The companion form realization, however, is numerically ill behaved, if we use a high order approximant, because coefficients of lower order terms of the monic denominator polynomial become very large. In this paper, the realization based on Cauer type continued fraction expansion is shown to be well behaved. A simple method of getting coefficients of the expansion of the Padé approximant, using the continued fraction expansion of the exponential function, is proposed. By simulation, it is shown that the ballancing of the overshoot and the undershoot is attained if we choose the degree L of the mumerator smaller 2 to 5 than the degree M of the denominator, and that the rise time is inversely proportional to L+M.View full abstractDownload PDF (567K) -
Tatsuya MASUDA, Katsuhiko FUJII1985 Volume 21 Issue 11 Pages 1177-1183
Published: November 30, 1985
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSA strongly connected stucture, in which many cycles exist and get entangled each other, is one of the most difficult system structures that can not be fully analyzed by the former methods.
This paper presents an effective structure analysis method for such a strongly connected structure. This method is outlined as follows. Firstly, all of the elementary cycles which form the base of strongly connected structure are picked up from the cycle set. Secondly, a new order relationship among these elementary cycles is defined based on two concepts, i. e., interval order and pseudo-interval order. The ordering results thus obtained are shown as a hierarchical directed graph in which the elementary cycles and the order relationships correspond to vertices and edges, respectively. The graph gives us a great deal of useful structural information buried in the strongly connected structure.
In this paper we demonstrate the effectiveness of this method by applying it to two examples.View full abstractDownload PDF (1014K) -
Toshiro ONO, Haruo SHIMAOKA1985 Volume 21 Issue 11 Pages 1184-1190
Published: November 30, 1985
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSUnder weightless conditions, existing available devices or systems for mass measurements are not usable because they are gravimetric. Therefore, we propose in this paper a new method of mass measurement usable under weightless conditions applying the concept of dynamic measurement method. In the present method, the goods-plate of a spring scale holding the goods to be measured has been displaced a small distance from rest position. Then, the goods-plate being released, it starts undamped oscillation. The mass value of the goods can be obtained through processing the observed data both on the acceleration of the goods-plate of spring scale and on the deflection of the spring by means of a software executed on a microcomputer.
In the present paper, we describe the two dynamic mass-estimation algorithms which play the important roles in mass-measurement under weightless conditions. The one is based on the continuous-time model of a measuring system and the other is based on the discrete-time model of a measuring system. In derivation of the two algorithms, we paid special attention to the dynamics of the two sensors which detect the acceleration and the deflection, respectively.
Moreover, a computer simulation is given to indicate the feasibilities of the two algorithms.View full abstractDownload PDF (881K) -
Kamekichi SHIBA, Tadashi ICHINOSE, Jin-ichi KITAMURA, Ryoji KOBAYASHI1985 Volume 21 Issue 11 Pages 1191-1195
Published: November 30, 1985
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThe capillary viscometer for gases of steady flow type reported here is a device of measuring the viscosities of gases in a steady state.
The measuring principle of the present viscometer is based on the previously developed Hagen-Poiseuille's law for gases. The viscosities can be obtained by a constant flowrate of a liquid and the pressure difference of both ends.
The outlet of the capillary tube is open to atmosphere. This simple trial viscometer enables us to measure the viscosities of gases easily.
Experimental results show that the obtained values are in good accordance with known reasonable values.View full abstractDownload PDF (1241K) -
Akira YAOITA1985 Volume 21 Issue 11 Pages 1196-1201
Published: November 30, 1985
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis report proposes some schematic optical systems for the temperature measurement of hot gases using double optical path method. As for the temperature measurement of hot gases with optical method, it is necessary to determine the emissivity and the radiance of them. The discussion here extends to three systems for measurement of the emissivity and two ones for the radiance. In each system, there are two optical paths which are prepared for the determination of the emissivity and the radiance in terms of radiative intensities observed through the both paths. These paths are assumed to have automatic balancing mechanism with the optical zero method.
Firstly shown are the basic equations of the emissivity and the radiance which are measured by the proposed optical systems, under the condition that the hot gases are in the state of thermal equilibrium and have axial symmetry. Nextly considered are several actual systems for temperature measurement which are composed of particular ones of the above-mentioned emissivity radiance measuring systems, and shown are the basic equations of the hot gas temperature which are determined using these actual systems.
Finally assessed are the characteristics of these systems using the above basic equations.View full abstractDownload PDF (925K) -
Position and Rotating Angle Measurement by Use of Optical Rotating FieldAkira KOBAYASHI, Shinji OOYAMA1985 Volume 21 Issue 11 Pages 1202-1209
Published: November 30, 1985
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSRecently the needs for distance and attitude measurements are increasing in material handlings by robots or flexiblemanufacturing systems, et al. In these measurements, optical or super sonic methods are mainly used since non contact measurements are desirable.
This paper proposes an idea of signal field which can be applied to various kinds of active measurement. First, both the generating method and the properties of rotating signal fields are considered, and an optical rotating field is composed experimentally by infra-red light emitting diodes IR-LEDs (940nm wave length) which are located quadrilaterally. On the other hand, the detecting system is set on a plane, and is composed of five photo-diodes; four of them are positioned at each corner of a square on the plane, and the remainder is at the center of the square.
A signal processing method is proposed for obtaining six variables which are used to determine the position and the attitude, then the methods are experimentally examined.
With compensating the angular sensitivity characteristics of the sensors, the following experimental results are obtained;
(1) As for the measurent of positions, the measuring ranges are from 500mm to 750mm (Z-direction), from-100mm to+100mm (both X and Y directions), and those absolute errors are within±5mm.
(2) As for the measurement of rotating angle, the measuring range is from-180 degrees to+180 degrees, and its absolute error is within ±5 degrees.View full abstractDownload PDF (1458K) -
Yoshikazu MATSUI, Tomokazu ARAI1985 Volume 21 Issue 11 Pages 1210-1216
Published: November 30, 1985
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis research concerns the trial manufacture of an angular position detector utilizing the Position Sensitive Detector (PSD) and its application to the electro-mechanical line tracer with a constant speed.
The PSD is a planar type PIN photo-diode utilizing the “Lateral Photo Effect” on the semiconductor surface, and it enables the accurate and fast detection of the position of a light spot on its active area with the aid of operational amplifiers.
The angular position detector manufactured here, has a simple structure; it comprises a light source, the PSD, and a slit. The advantages are listed up as follows: i) Free from problems arising from mechanical contact, ii) wide in the effective angle of rotation, iii) excellent in temperature characteristics, iv) free from the noise effects, v) a fast response, vi) a small moment of inertia, vii) various non-linear output voltages are obtainable directly by simply changing the slit shape.
The present detectors are two types of output voltage in proportion to the angular position θ, and two simultaneus output voltages in proportion to sin θ and cos θ.
Furthermore, the line tracer manufactured here, as one application of the angular position detector, can automatically follw figures drawn in a thin black line on a paper. It is applicable to automatic welding or cutting machine, etc.
The performances of the line tracer are as follows: i) A following speed of 12mm/s max, ii) the width of line capable of being followed, 0.15 to 2.0mm, iii) a tracking error of ±0.3mm in width.View full abstractDownload PDF (869K) -
Kiyoshi OHISHI, Kouhei OHNISHI, Kunio MIYACHI1985 Volume 21 Issue 11 Pages 1217-1222
Published: November 30, 1985
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSGenerally, the speed controller of dc motor is requested to have a quick and accurate response for speed reference and a small speed fluctuation due to the imposed load torque. For this purpose, the drive system of dc motor mounts the speed sensor. However, this speed sensor can not be always implemented because of the limit of the space or the enviromental conditions.
Therefore, from the practical view point, it is desirable to realize a quick speed control of dc motor without speed sensor. This request can be realized by a new method proposed in this paper, which is based on the discrete observer theory. As the motor control system, in which the armature current is sensed, can be treated as an observable system, both the motor speed and imposed load torque can be estimated by the discrete minimum-order observer. In this paper, it is shown that only armature current is a sufficient information for the speed control. This dc motor drive system based on speed estimation method is realized by the microprocessor. Numerical and expermental results are also shown in this paper.View full abstractDownload PDF (894K) -
Tsutomu MITA, Eiji KATO, Yasushi AOKI1985 Volume 21 Issue 11 Pages 1223-1230
Published: November 30, 1985
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThere is a new control method called “learning control” proposed by Uchiyama and Arimoto et al., However, in this control method, there is no precise proof on the convergence provided that the trial time interval is assumed to be infinite, and further, plants which can be controlled are restricted to having transfer functions of relative order zero. In this paper, we improve this control method and give precise proof on the convergence, and the results are applied to the trajectory control of robot manipulators.View full abstractDownload PDF (978K) -
Hidetoshi WAKAMATSU, Toshiyuki KITAMORI1985 Volume 21 Issue 11 Pages 1231-1238
Published: November 30, 1985
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSA method of controlling a respiratory system is proposed for the maintenance of a desired alveolar CO2-concentration with the elimination of the effect of irregular metabolic rate change. This method is based on the model matching technique under the condition that a respiratory system is only partly described by Volterra series. That is, the contribution of air ventilation rate change to alveolar CO2-concentration is regarded to be partially described by an input-output relation using finite order Volterra series; while that of metabolic rate change cannot be recognized at all. Based on such a limited knowledge of the controlled object, two kinds of control systems; one with cascade compensation and the other with feedback compensation are designed by computer-aided partial model matching technique with a given reference model. They provide a robust control against possible parameter deviation caused by the difference of individual patients, measurement error and chronic change which are characteristic in biological subsystems. That assures an application of the proposed method to a clinical control of respiration.View full abstractDownload PDF (1314K) -
Osami SAITO, Takeyuki TAKAHASHI, Ken-ichi ABE1985 Volume 21 Issue 11 Pages 1239-1246
Published: November 30, 1985
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSComputer Aided Design (CAD) systems are essential tool in applying the control theories to the practical problem. Recently various CAD systems of control ssystems have been developed. However all these systems are based on numerical calculation, Hence they are not available for the problem in which the parameters of mathematical models are given as symbols.
This study developed a CAD system which can handle symbolic parameters by using symbolic and algeblaic manipulation system REDUCE 3.1. It is shown that this CAD system can treat the wider problems than the conventional one, for example, the investgation of generic property of the system and designing a controller by exact model matching technique.View full abstractDownload PDF (988K) -
Nobuaki KOBAYASHI, Takayoshi NAKAMIZO1985 Volume 21 Issue 11 Pages 1247-1249
Published: November 30, 1985
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSDownload PDF (317K) -
Makoto SHIMOJO, Masatoshi ISHIKAWA1985 Volume 21 Issue 11 Pages 1250-1252
Published: November 30, 1985
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSDownload PDF (850K)
- |<
- <
- 1
- >
- >|