This paper proposes a new kind of on-line identification methods of continuous time-delay systems from sampled input-output data. In order to track the time-varying system parameters as well as time-delay, the recursive linear least squares (RLS) method is combined in a bootstrap manner with the genetic algorithm (GA) which has a high potential for global optimization. The time-delay is coded into binary bit strings and searched by the GA, while the system parameters are updated by the RLS method. Furthermore, this method (GALS method) is hybridized with the sequential nonlinear least squares method to improve the speed of convergence. Simulation results show that both the GALS and hybrid methods are efficient in the case that the system changes abruptly, and among them the hybrid method has superior convergence property to the GALS method and yields excellent estimation results in the case that the system changes with time continuously.

抄録全体を表示

This paper considers a modeling method of the temperature control system for a chemical plant which produces polythlene in a batch reactor. In this plant, the reaction in a tank is complex and has many nonlinear factors. Thus, it is difficult to make the model which approximates an output of this plant precisely. It is well-known that neural metworks are useful for many problems in nonlinear systems because of the ability of learning and nonlinear representation. We apply the neural network method to model a temperature control system in this plant. Finally, the simulation results are illustrated to show the effectiveness of the modeling method used here for the temperature control system of a batch reactor.

抄録全体を表示

To achieve a reasonable controller design for a distributed parameter system, a low order finite dimensional approximation model is indispensable which adequately reflects principal characteristics of the plant.
In this paper we consider an approximation scheme, based on the Galerkin criterion, for linear stable timeinvarient distributed parameter systems. First, we illustrate the construction of the model via the Galerkin method employing a trial function with a force term. Next, we show that the static error of the model can be compensated and reduced to zero by using a function from steady-state step response in the force term, and that the reduced order finite dimensional model has frequency response which is consistent with the original one in a wider range.
A numerical study for an example shows the merit of the scheme in terms of the error plot in frequency domain. We demonstrate the feasibility by considering the achievable control performance in designing a compensator based on the model with the static compensation.

抄録全体を表示

In this paper, we consider a problem to estimate process parameters using input-output data collected from the process operating in closed-loop control system. When orders and delay-time of the process are known correctly, under some conditions of identifying experiments, it is reported that accurate identification results can be obtained by applying prediction error method. To get accurate estimates, it is necessary to know orders and delay-time of the process. It is difficult to determine them in closed-loop identification, because ill-condition for identification are easily caused by selection of unsuitable order or delay-time. Furthermore, the procedures to select orders and delay-time in open-loop identification aren't always available in closed-loop identification. The purpose of this paper is to determine a delay-time under suitable assumption that order of the process are known as the first step.

抄録全体を表示

A resonant frequency of piezoelectric vibrator with high electromechanical coupling varies remarkably depending on electrical impedance connected to the electrical terminals. In this paper, this effect has been applied to realize a frequency controllable ultrasonic transducer made of composite piezoelectric vibrator. The characteristics of the frequency controllable ultrasonic transducer is investigated theoretically by the new transmission-line model for a thickness mode piezoelectric vibrator which has been recently derived including the electrical impedance connected to the electrical terminals. It is further confirmed experimentally that the resonant frequencies can be controlled satisfactorily by a variable electrical impedance.
The results are summarized as follows:
(1) In the case of the two layered piezoelectric transducer, the resonant frequency has been controlled continually in wide frequency range (180kHz_??_580kHz). In this case, the effective attenuation for the variable resonant frequencies are less then 7dB for the frequency range of 330kHz_??_470kHz when the acoustic load is water.
(2) In the case of the bolt-clamped type piezoelectric transducer, the resonant frequency has been controlled continually from the fundamental resonant frequency (27kHz) to the second harmonic resonant frequency (60kHz) by varying the values of electrical inductance L_e. The characteristics of electro-acoustic efficiency have a minimum at 52kHz, where the minimum value was above 20%.

抄録全体を表示

Optical information processing is expected to realize high performance and a large capacity of computing for its natural parallelism, non-interfering transmittance and wide bandwidth. However, carries make the optical computation difficult with its natural parallel processing of each digit. In this paper, the optical adder in the modified signed-digit (MSD) number system is discussed. The MSD system does not cause carries, but is difficult to deal with negative numbers. We propose a method with not only optical intensity and optical phase information but also polarization. We prototype the logic elements for the MSD adder using the optical Fourier transform, to show that this would realize a MSD adder and function as a subtraction. We further discuss on the parallelization by multiple optical logic units and multiple frequencies.

抄録全体を表示

This paper describes the development of a touch sensor which has a mono-structure, but has three kinds of sensing functions for discrimination of material properties. The proposed sensor is used to discriminate typical materials such as iron, aluminium, copper, bronze, gum, wood, acrylic resin, styrene foam and etc..
The sensor consists of two spiral metallic lines on a thin insulating substrate and has four terminals. When the two lines are used as two coils, the sensor has the property of inductance. When the two lines are used as two parallel electrodes, the sensor has the property of capacitance. When one spiral metallic line is used as heater and another one is used as resistor, the sensor has the property of temperature sensor.
The experiments show good results to discriminate each sample of materials and are expected as a step to develop an artificial skin tactile sensing system in future.

抄録全体を表示

This paper proposes a hierarchical modelling method of timed petri nets for large scale manufacturing systems. This method is able to reduce a direct net to a transition with conserving a property of completion time. Therefore, this method enables planner to make a hierarchical decision of managing manufacturing systems, especially in large scale manufacturing systems.
This paper shows that the proposed method is capable of hierarchical modelling of systems and is useful in decision making of management of large scale manufacturing systems.

抄録全体を表示

One of the most serious problems for carrying large-sized power equipments into a substation/power plant efficiently is to find a good wiring route for roller work which executed by putting each equipment on rollers and pulling it horizontally with a wire rope. As roller work is too complicate, we have no systematic study on wiring route decision ever before. So we leave this work in the hands of field overseers without any detailed study.
In this paper, we propose some simulation methods for wiring route decision, discuss how to put it to practical use, and then briefly describe visualization techniques. The proposed methods give a wiring route under the restriction of winch and hook positions, and reduce the number of times of changing pulley positions. The proposed wiring route decision method provides suitable environment in which the installation process can be examined sufficiently; efficient, economical installation works can be done in safe and quick.

抄録全体を表示

In excavating tunnels for subways and power cables, shield tunneling machines, which have many cutters, are used. We previously proposed an effective detection system, for the machines, which detected anomalous objects ahead, to prevent the cutters from getting damages and also to prevent the construction delay caused by the accident. The system had one sonic transmitting transducer and several sonic receiving transducers to get information on the anomalous objects. Not so many observations being available in the detection, optimal observation policy is of a practical importance. From this viewpoint, we next considered the optimal location of the receiving transducers on the cutter plane of the shield tunneling machine and also the optimal observation policy, which took into account the rectlinear and revolving motions of the cutter plane caused by the movement of the tunneling machine, for the case where the anomalous objects were lumped ones. This paper proposes the optimal location of the receiving transdusers and optimal observation policy for detecting anomalous plane object and shows that the same result as before is obtained even for the anomalous plane object.

抄録全体を表示