All issues
Volume 27, Issue 6
Displaying 1-18 of 18 articles from this issue
- |<
- <
- 1
- >
- >|
-
Yuzo OHTA, Katsuhiko YASUI, Hiromasa HANEDA1991 Volume 27 Issue 6 Pages 620-627
Published: June 30, 1991
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn this paper, the relative stability of control systems with uncertainty is studied. It is shown that relative stability problems such as the degree of stability and stability margin are reduced to the evaluation of the image of characteristic function under perturbation of parameters. To get an approximation of the image, the addition and multiplication of intervals is used. As intervals in the complex plane, the family of polygons are used, and traditional interval analysis in the real line is generalized to the complex plane. Moreover, to get a good approximation of the image, 2 methods are used: one is the method based on the decomposition of extreme points of convex polygons, and the other is the method based on the partition of polygons. The former method includes the method based on the mapping theorem as a special case.View full abstractDownload PDF (1088K) -
Kazuyuki JOYA, Katsuhisa FURUTA1991 Volume 27 Issue 6 Pages 628-632
Published: June 30, 1991
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSA necessary and sufficient condition for all members of a family of polynomials to be Schur stable is given in the case that the variation of the coefricients of polynomials is constrained with lp-norm.
With a computer, the condition becomes to be feasible. Two examples are given to show that.View full abstractDownload PDF (530K) -
Hongye Du, Yoshikazu HAYAKAWA, Seizou FUJII1991 Volume 27 Issue 6 Pages 633-639
Published: June 30, 1991
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn the design of feedback control systems, two kinds of uncertainties must be considered. One is the external disturbance, the other is the plant perturbation, which is caused by the variance of the parameters and the uncertainties of the mathematical model of the plant. The second must be considered seriously by system synthesis because the uncertainties can harm the stability of the colsed-loop system. A controller that satisfies the stability requirement under perturbation is the main aim of robust control.
In the case of centralized control systems, there are a lot of studies which are concerned with the robust control. Especially, the Structured Singular Value (SSV) method, developed by J.C. Doyle, et al., can deal with the robustness and performance simultaneously. This is why the SSV has received increasing attention. For the decentralized controll systems, S. Skogestad & M. Morari gave a sufficient condition for robust performance by using the SSV method.
In this paper, by using the SSV and the parameterization of all the decentralized stabilizers that the authors have obtained already, two kinds of perturbation models (an additive pertutbation and a multiplicative one) are considered.View full abstractDownload PDF (719K) -
Yasuyuki FUNAHASHI, Hisao KATO1991 Volume 27 Issue 6 Pages 640-646
Published: June 30, 1991
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn this paper, a design method for the deadbeat compensator with optimal robustness for MIMO systems is presented. As the compensator is decided among the class of two-degree-of-freedom robust tracking compensators, it is guaranteed that the deadbeat control and internal stability of the whole system is achieved for nominal process. Zero offset is also achieved for the perturbed process which does not disturb internal stability.
Using the Hermite form, the design procedure of this paper is much simpler than the one, proposed in the past, that uses the Smith-McMillan form.
And robustness criterion, which is L2-norm of sensitivity function, is optimized. So it is hoped that the stability region for process perturbation is enlarged and that the influence of process perturbation on the output gets smaller. The minimization of deadbeat steps and the optimization of robustness are achieved simultaneously. It is a merit of two-degree-of-freedom compensation scheme.View full abstractDownload PDF (802K) -
Kazuo AIDA, Toshiyuki KITAMORI1991 Volume 27 Issue 6 Pages 647-654
Published: June 30, 1991
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis study is concerned with a design method of the robust servo system which is optimal in the sense that a performance index, comprising the quadratic forms of both the output error vector and the deviation control vector, is minimized.
The design method is an exact model matching with pole-zero cancellations. The model is an optimal control system without servo compensator, designed with the optimal regulator technique. The initial values of the servo compensator are set according to the initial conditions of the plant and the reference signal generator.
The sensitivity can be reduced utilizing the configurations of cancelled pole-zero pairs. The order of the Butterworth pole pattern is constant regardless of the dimension of the servo compensator. Moreover, this servo system has the same stability robustness, in the sense that the criterion for optimizing is satisfied, as typical optimal servo systems designed using different performance indices.
A numerical example illustrates the efficiency of this servo system.View full abstractDownload PDF (846K) -
Hisashi TSUCHIDA, Nobuhide SUDA1991 Volume 27 Issue 6 Pages 655-662
Published: June 30, 1991
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSWe discuss the construction of optimal servosystems with quadratic stability for uncertain systems with structured, time-invariant and bounded uncertainties. The aim is to design a control law such that the feedback system has three features: its output asymptotically tracks any constant reference input, it minimizes a certain quadratic performance index, thus inheriting the desirable properties of optimal regulator design, and the stability and the tracking property are both ensured for any uncertainty in a given bounded set. Based upon the observation that the quadratically stabilizing controller by Petersen, if exists, is an optimal regulator, in some way or another, for any of the uncertain system under consideration, such a control law together with a sufficient condition of its existence is derived. Further, by extending a particular optimal servosystem proposed by Suda and Ikeda to the uncertain system, it is shown that, under a sufficient condition, weaker than the so-called ‘matching condition’, it is possible to construct an optimal servosystem with quadratic stability which has a set of gain tuning parameters. Through an example, it is demonstrated that the speed of response is easily selected by adjusting these tuning parameters.View full abstractDownload PDF (876K) -
Takashi SUZUKI, Hiromitsu SUZUKI1991 Volume 27 Issue 6 Pages 663-670
Published: June 30, 1991
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIt is necessary for practical applications to design the adaptive control system which behaves robustly in the presence of unmodeled dynamics such as parasitics. As such robust schemes, the adaptive laws with σ-modification and with relative dead zone have been proposed. The latter scheme is attractive in the sense that the global stability can be guaranteed, but some problems such as how to specify the width of dead zone and the output error between the plant and the reference model arising from the use of dead zone are left to be resolved.
This paper presents a robust design method of the direct model reference adaptive control system in the presence of parasitics. In this method, the controller parameters are adjusted by an adaptive law with relative dead zone and the output error is reduced by a fixed feedback compensator. The stability of the overall system is analyzed using the singular perturbation theory and the Liapunov's direct method and then it is shown that the adaptive system is able to behave robustly to the sufficiently small parasitics if the width of dead zone is specified appropriately.View full abstractDownload PDF (1029K) -
Kenji IKEDA, Seiichi SHIN, Toshiyuki KITAMORI1991 Volume 27 Issue 6 Pages 671-677
Published: June 30, 1991
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper presents a model reference adaptive control system (MRACS) for an SISO continuous-time linear plant in the presence of bounded disturbances. The assumptions made regarding the plant and disturbances are (i) the plant is minimum phase, (ii) an upper bound on the plant order is known, and (iii) the maximum amplitude of the equivalent disturbance at the controlled output is known.
Because the plant relative degree is not known a priori, the adaptive algorithm is based on the switching of parameter adjustment rules, one of which ensures the global stability of the MRACS. And “dead zone” method is applied in the adaptive algorithm for bounded disturbances.
In the stability analysis, the followings are proved: (1) the parameter adjustment rules are switched at most finite times, (2) the adaptive parameters are bounded, (3) the time derivative of the parameters are square integrable, (4) state variables of the control system are globally bounded on [0, ∞), and (5) the output error enters into a certain bound.
Numerical analysis is performed to compare the proposed method and the method in which no dead zone is used. It will illustrate the robustness of this method for bounded disturbances.View full abstractDownload PDF (821K) -
Naoki MIZUNO, Seizo FUJII1991 Volume 27 Issue 6 Pages 678-685
Published: June 30, 1991
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper presents a redesign method of discrete-time adaptive control system for continuous-time systems. Using appropriate approximate discrete-time models of the continuous-time system, called limiting-zero model and modified delta-model, we avoid the problem of the unstable zeros which arise in the case of rapid sampling of the continuous-time system having the relative degree greater than one. In this paper, first, we discuss the motivation and principle of the limiting-zero model of the continuous-time system. Next, it is shown that, in modeling error at high frequency range, the limiting-zero model is superior to delta-model. Moreover, the modified delta-model is introduced by regarding the phase-shift characteristics of the limiting-zero polynomials. Finally, in order to confirm the effectivenese of the adaptive control system using the proposed models, computer simulations are carried out for a continuoustime system having the relative degree equal to 3. It is shown that the proposed models can be successfully applied.View full abstractDownload PDF (1080K) -
Yoshihiko MIYASATO1991 Volume 27 Issue 6 Pages 686-692
Published: June 30, 1991
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThe model reference adaptive control (MRAC) problems for lumped parameter systems have been actively investigated. For distributed parameter systems, the MRAC problems have been recently studied by some reserchers, who used infinite dimensional compensators to solve those problems. Hence, their methods are not useful for practical applications. Afterwards, the finite dimensional adaptive controllers for distributed parameter systems were proposed. In one of them, the nonlinear control laws and the adaptive laws with dead-zones were used, and the uniformly bounded adaptive control systems were proposed for distributed parameter systems of parabolic type.
In the present paper, we apply that method to distributed parameter systems of hyperbolic type, and construct adaptive control systems by finite dimensional controllers. It is shown that the resulting control systems are uniformly bounded for any bounded reference signals which the output of the process must follow, and that the tracking error converges to a small residual set whose magnitude can be determined freely. Finally, some simulation results show the effectiveness and simplicity of the proposed method.View full abstractDownload PDF (721K) -
Junichi IMURA, Toshiharu SUGIE, Yasuyoshi YOKOKOHJI, Tsuneo YOSHIKAWA1991 Volume 27 Issue 6 Pages 693-699
Published: June 30, 1991
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis papar is concerned with a robust control for robot manipulators in the case where joint torque sensors are available. First, we derive a dynamic model of the robot manipulator with joint torque sensors, in which the total nonlinear multivariable structure is explicitly described. This model gives a good indication that we can construct the control systems of robot manipulators with joint torque sensors using the same method as in the conventional case without the sensors. Second, based on this model, we propose a robust trajectory tracking control scheme which achieves the specified tracking accuracy in the presence of the modeling error including motor system, where joint torque sensor information is fully exploited to compensate the uncertainty of link and load parameters. Furthermore, an illustrative simulation result is given to show the effectiveness of the proposed control method.View full abstractDownload PDF (995K) -
Akihiko HASEGAWA1991 Volume 27 Issue 6 Pages 700-704
Published: June 30, 1991
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThe strip temperature dynamics of the heating furnace of Continuous Galvernizing Line (CGL) consists of dead-time and time lag against the change of various plant variables. The exact identification of these characteristics is difficult because the dynamics is time-varying depending upon operational conditions and long-term shifts of plant characteristics. In view of these facts, the author applied the Self-Tuning Generalized Predictive Control to the strip temperature control of the heating furnace and obtained fine control performance, and established a new strip temperature control for the heating furnace of CGL or CAPL (Continuous Annealing and Processing Line). This control has been successfully applied at No.5 CGL in Nagoya Works of Nippon Steel Corporation and the basic idea of the control is transferred to No.2 CAPL in Nagoya Works and No.2 CAPL in Kimitsu Works.View full abstractDownload PDF (661K) -
Taketosi KAWABE, Yoshinori YAMAMURA, Kimio KANAI1991 Volume 27 Issue 6 Pages 705-710
Published: June 30, 1991
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper presents a controller for an electric motor-driven servo positioning system intended for use in automobiles. The controller features a low sensitivity design to provide robustness against eternal disturbances and the non-linearity of the positioning mechanism. The discrete plant model denoted by the sift operator has a zero near the unit circle in the z-plane that can cause oscillation in the electric motor current, resulting in electricor sound noise when the controller is designed to contain the inverse system of the plant. The use of delta operator to denote the discretized plant model make it possible to ignore t is zero.View full abstractDownload PDF (1727K) -
Masami SAEKI, Masatoshi HOSHINO, Kazuaki ANDO1991 Volume 27 Issue 6 Pages 711-713
Published: June 30, 1991
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn the loop transfer recovery procedure by H∞ norm proposed by one of the authors, it is shown that the degree of recovery cannot be improved by generalizing the class of controllers from the class of transfer functions consisting of an optimal gain and an observer to that of real rational stable transfer functions.View full abstractDownload PDF (346K) -
Ryou KONDO, Shinji HARA, Takuji KANEKO1991 Volume 27 Issue 6 Pages 714-716
Published: June 30, 1991
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSA parameter space design for H∞ constraints is presented in this paper. A certain H∞ constraints must be satisfied for robust control system. Recently, various theories for H∞ control design have been studied, while they are not applicable to the problems for fixed structure controllers such as PI compensators. This paper provides a parameter space design method for H∞ control problem using Routh-Hurwitz type criteria for the characteristic polynomial of a Hamiltonian matrix with respect to H∞ norm. Furthermore, the method is extended for the gain constraints in the specified frequency band.View full abstractDownload PDF (343K) -
SHUMING Chen, Yoh YONEZAWA, Yukio NISHIMURA1991 Volume 27 Issue 6 Pages 717-719
Published: June 30, 1991
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn this paper, a robust adaptive control method for continuous time plant in the presence of slowly-varying disturbance is proposed, Which makes the output error be small by the way to sample the plant with short sampling period and to use an adaptive adjustment law with dead-zone.View full abstractDownload PDF (352K) -
Hiromitsu OHMORI, Hiroshi ITO, Keiji OHKUBO, Akira SANO1991 Volume 27 Issue 6 Pages 720-722
Published: June 30, 1991
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper presents a robust design method of MRACS including a fixed compensator for an SISO LTI continuous-time plant in the presence of bounded disturbances. It is shown that the design problem of the fixed compensator is formulated as the strong stabilization problem with the sensitivity reduction based on H∞-optimal theory.View full abstractDownload PDF (424K) -
Yasumasa FUJISAKI, Hideki ASADA, Akira KITAMURA1991 Volume 27 Issue 6 Pages 723-725
Published: June 30, 1991
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSWe propose a robust controller for a hot strip mill looper system with the change in rolling conditions. The tranfer function of the looper system was derived using a statistical method and its variation characteristics were shown. We designed a robust controller based on an internal model control structure. This controller was applied to the hot strip mill looper system and its robust stability was confirmed.View full abstractDownload PDF (395K)
- |<
- <
- 1
- >
- >|