All issues
Volume 20, Issue 3
Displaying 1-16 of 16 articles from this issue
- |<
- <
- 1
- >
- >|
-
Takayoshi TOTANI, Kenzou NONAMI, Hitoshi OKAMURA1984 Volume 20 Issue 3 Pages 187-192
Published: March 30, 1984
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThe paper presents a method for realizing the finite-time linear optimal control by means of a time-invariant state-feedback control law and a compensation input, and shows the procedure to compute the required compensation input. One of the advantages of the method is that the control system behaves as a time-invariant optimal regulator if there is no compensation input. The method was applied to the settling control of a bridge crane model with a micro-computer, and the experimental results which were in good agreement with the theory were obtained. It is shown that the method is far more practical as the means for constructing the finite-time terminal control than the one by the well-known time-varying state-feedback control law.View full abstractDownload PDF (753K) -
Takeshi TSUCHIYA1984 Volume 20 Issue 3 Pages 193-199
Published: March 30, 1984
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSMulti-input and multi-output output feedback control system design method based on improved optimal regulator theory is proposed. Steady state errors of the output feedback control system remain zero for changes in the desired signals and the disturbance signals and in spite of the parameter variations of the controlled object, same as in the full state feedback control system previously proposed by the author. The control system structure of the output feedback control system as well as the full state feedback control system are clarified. It is shown that proportional actions and integral actions are included in the full state feedback control system, on the other hand, derivative actions as well as proportional actions and integral actions are included in the output feedback control system. That is, (n-1)th and (n-2)th order derivative actions for error signals and disturbance signals are introduced for n-th order controlled object. In order to improve the transient responses for changes in the deisred signals and the disturbances, feedforward control loops utilizing the informations on the signals, if possible, are introduced for both the systems. Input delay times due to the processing time required to execute the employed control algorithm in a microprocessor are also easily taken into account in this design method, which results in the introduction of the compensating actions for the input delay times.View full abstractDownload PDF (842K) -
Katsushi MURAKAMI, Hideaki KANOH, Masami MASUBUCHI1984 Volume 20 Issue 3 Pages 200-206
Published: March 30, 1984
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSA method of constructing a state observer for a continuous fermentation process is presented in this paper. The process is expressed by a growth model describing the growth rate of the microorganisms and the consumption rate of limiting substrates and by a product model for product synthesis. These models are described by ordinary differential equations to represent a completely stirred tank fermentation system. The growth models we consider here are general models which include the Monod model as a special case.
Firstly, conditions for constructing the state observer for the growth are obtained based on the fact that the state observer derived from growth models have nonnegative solutions as in growth models.
Secondly, conditions for designing the state observer for the estimation of the growth rate of and product rate from the microorganisms are obtained using the results of the present authors on a stability problem for the error equation of state estimation in which the time varying system matrix reduces to a triangular matrix by appropriately choosing the design parameters.
A numerical example is included to show the performance of the observer proposed.View full abstractDownload PDF (937K) -
Jiro SHIMONISHI, Takao HINAMOTO, Sadao MAEKAWA1984 Volume 20 Issue 3 Pages 207-213
Published: March 30, 1984
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper considers the problem of pole assignment for two-dimensional (2-D) systems expressed by Roesser's state-space model. This is done by introducing a 2-D dynamic compensator. First, output feedback law is established and the pole assignment problem is divided into two 1-D pole assignment problems. Next, a sufficient condition for the existence of 2-D dynamic compensator is presented to assign arbitrary poles given in a set of complex number pairs. Finally, an algorithm is proposed for designing a 2-D dynamic compensator and a numerical example is solved to illustrate it.View full abstractDownload PDF (824K) -
Fujio OHKAWA, Yoo YONEZAWA1984 Volume 20 Issue 3 Pages 214-219
Published: March 30, 1984
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSA large number of design method for a model reference adaptive control system (MRACS) have been proposed recently by the researchers of stability theorems. The MRACS is today one of the most feasible approaches for the practical implementation of adaptive control system. However, few papers have so far dealt with design method and stability of the MRACS for a plant with unknown time-delays.
In this paper, a design method for discrete MRACS for single-input single-output system with unknown time-delay is proposed based on plant-input representation. It is shown that the proposed adaptive algorithms gurantee the convergency of the plant's output to the reference sequence by introducing an auxiliary input signal and pole assignment method. For the adaptive algorithms proposed here, only available input-output data of the plant are used, using no anticipative value of the plant output and information on unknown plant parameters. Finally, the validity of the proposed adaptive algorithms is illustrated by numerical example.View full abstractDownload PDF (714K) -
Tetsuro NOGUCHI, Kimio KANAI1984 Volume 20 Issue 3 Pages 220-226
Published: March 30, 1984
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn designing a control system of the plant where only the input and the output can be measured, it is desirable to use the adaptive control synthesis approaches. However, most of those approaches can not apply to the plant with the limitation of input amplitude, because those studies are usually done without any limitation of input amplitude.
For such a case, by introducing the dispersion filter, we have proposed the adaptive control scheme for a time-invariant discrete scalar system with the limitation of input amplitude. However, this scheme required the inversion of the dispersion filter to construct the adaptive feedback loop.
In order to overcome the above difficulties, this paper presents the approach by the use of a polynomial algebraic method for constructing an adaptive control system of the time-invariant discrete scalar system with the limitation of input amplitude. Simulation results of a third-order unstable system with the limitation of input amplitude are included at the end of paper in order to verify the effectiveness of the propose scheme.View full abstractDownload PDF (817K) -
Katsutoshi TAMURA, Kunihiko ICHIKAWA, Kenji KOTAKI, Yasushi AMANO1984 Volume 20 Issue 3 Pages 227-232
Published: March 30, 1984
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn this paper a model reference type adaptive speed control system is designed for D. C. servo motor systems with coulomb friction. Conventional model reference adaptive control systems (MRACS) are not directly applicable to this design problem due to the presence of the friction.
A method of designing MRACS for linear continuous systems with stepwise disturbances is presented and a proof of all signals in the system is given. This method is applied to the design problem mentioned above.
An experimental equipment of MRACS is implemented based on the proposed method, where a D. C. servo motor with an unknown load is controlled plant and an analog computer is used as a adaptive controller and also as a reference model. The experimental results show that the proposed MRACS works effectively and compensates successfully the coulomb friction in the plant.View full abstractDownload PDF (1060K) -
Yoshio NAKATANI, Shinji ARAYA1984 Volume 20 Issue 3 Pages 233-240
Published: March 30, 1984
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSInformation processing approach is used for simulation of human escaping behavior from a building in case of a fire. This simulation model makes much of qualitative knowledge and is written in PROLOG suitable for representing and utilizing this.
The model is for an individual behavior. Information about a fire input from terminal is interpreted and the behavior is determined through ‘rules’ and ‘knowledge’ previously set. ‘Rules’ are represented as production rules, deciding how to react to a certain situation based on ‘knowledge’. ‘Knowledge’ is know-what type knowledge about facts and has frame structure. ‘Rules’ and ‘knowledge’ have general type and personal type. Various types of behavior is possible to realize by changing the personal rules and knowledge.
The effectiveness of this approach is shown through simulations, reproducing an actual disaster which happened at a cabaret ‘PLAYTOWN’ in OSAKA.View full abstractDownload PDF (1058K) -
Tadaharu YOKOKAWA, Iwao YAMAMOTO1984 Volume 20 Issue 3 Pages 241-246
Published: March 30, 1984
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSDescribed in this paper, is a semi-automatic data processing method for curves drawn by strip chart recorders. The main parts of this system is a curve following and transforming mechanism which uses both manual tracing device and photo-diode array sensor with a fiber image guide. A long strip chart, on which an experimental recorded curve is drawn, is fed continuously through a guide bar which is placed vertically to the direction of the chart feeding. A cursor unit equipped with one end of a set of image and light guide fibers is fixed on a slider which can move smoothly on the bar. The slider is connected to a potentiometric voltage generator and another end of the set of the fibers is connected to photo-diode array sensor; the output of the following mechanism is given as the sum of the output of the voltage generator and that of the photo-diode array sensor.
Controlling the slider position as to contain the curve into the range bounded with two markes, the curve drawn on the strip chart is transformed to time varring analogue voltage.
Dependence of both the chart speed and the spatial frequency of the curve on the system performance are discussed by experiments using test pattern data, compared to the pure manual operating system.
The error accompanied with the manual operation system is rejected by the semi-automatic system and the processing speed increased approximately four times as much as that of the manual operation.View full abstractDownload PDF (975K) -
Susumu SAITO, Toshimichi SAKAI1984 Volume 20 Issue 3 Pages 247-253
Published: March 30, 1984
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThe pipe as a fluid signal transmission line has many natural frequencies of the first, the second, …that depend on its length. At those frequencies, the amplitude of the signals shows peaks. For this reason it is difficult to operate ac fluidics in wide frequency range.
In this study, rarefied gas was used to improve the characteristics of the transmission of pulsating fluid signals by removing the defects of the amplitude change by the resonance as stated above. This principle is as follows; The viscosity of gases is independent to the pressure and the density is proportional to the pressure, then the kinematic viscosity will become large in rarefied gases, and as a result, the peaks of the amplitude at the resonant frequencies are reduced.
In carrying out this experiment, fluid amplifiers and a sinusoidal fluid signal generator were placed in a closed vessel, and their supply ports were opend to atmospher, and then the pressure of the vessel was lowered by a vacuum pump. The experimental results showed that the effects of the resonance could be reduced considerably at the pressure as less as 0.1×105[Pa].
The frequency characteristics of the transmission lines were well explained by the distributed parameter theory which was modified to apply for the pulsating fluid signal with do flow.View full abstractDownload PDF (858K) -
Michio KONO, Yoji MURAKAMI, Takashi MORISHITA, Kiyoshi MISHIMA1984 Volume 20 Issue 3 Pages 254-259
Published: March 30, 1984
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper deals with decoupling control of controllable pitch propeller (CPP) propulsion system, which has two independent inputs and two outputs. Inputs are the propeller revolution reference and the pitch angle. Outputs are the propeller revolution and the horsepower. Each input affects both outputs. By applying a decupling control law to this system, whole system is divided into two single-input single-output subsystems. Each input affects only the corresponding output. Since an entire state vector is not available in CPP propulsion system, an observer or a Kalman filter is used in order to estimate the state vector. We apply this method to CPP propulsion system of T. S. Shioji Maru. The decoupling control law, the observer and the Kalman filter have been implemented in the minicomputer. The actual sea test have been carried out, and experimental results are given.View full abstractDownload PDF (693K) -
Takayoshi TOTANI1984 Volume 20 Issue 3 Pages 260-265
Published: March 30, 1984
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThe paper presents a procedure for designing terminal motion trajectories which are used when a manipulator hand carries an object from one place to another in the operating space. For constructing the space trajectory, the mathematical model of time trajectories which has been obtained by analyzing human hand motions, is used as the components in the three axis directions. The trajectory parameters are determined by specifying the positions of the terminal points and the direction vectors of the space trajectory at the terminal points. In the design procedure, the quantities of overshoot and undershoot in each direction can be specified as well. In this paper, some diagrams which are necessary to select the trajectory parameters are given and design examples of plane trajectories are shown.View full abstractDownload PDF (725K) -
Its Experimental ConsiderationsAkira MOHRI, Hiroaki OZAKI, Shigetoshi SATO, Tatsuo KAMACHI1984 Volume 20 Issue 3 Pages 266-271
Published: March 30, 1984
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper experimentally discusses an adaptive decentralized control of a manipulator on the concept of the following hierarchical structure. The higher level computer plans a path and calculates individual approximate actuator inputs to drive a manipulator to follow the path. Based on the calculated inputs each of the lower level computers controls the actuator so as to attain its own desired joint trajectory corresponding to the planned path. An adaptive control law is obtained and tried to control individual actuators. The used manipulator has a shoulder and an elbow joints driven by AC two-phase servomotors and is controlled to follow a planned path in the cases with and without load in the hand. State feedback control law and PI control law are also used to compare with the adaptive control law. In the experiments the adaptive control law has been favourably compared with the other two control laws in the trackability.View full abstractDownload PDF (1175K) -
Kazuo MUROTA1984 Volume 20 Issue 3 Pages 272-274
Published: March 30, 1984
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSDownload PDF (397K) -
Masakazu IMAI, Seiji INOKUCHI1984 Volume 20 Issue 3 Pages 275-277
Published: March 30, 1984
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSDownload PDF (343K) -
1984 Volume 20 Issue 3 Pages 279
Published: 1984
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSDownload PDF (143K)
- |<
- <
- 1
- >
- >|