The m/n frequency divider founded on the new multivibrator using the switched-capacitor (SC) circuit and the CMOS logic gate is described. In this multivibrator, the electric charge is transmitted between the capacitor of the SC circuit and the coupling capacitor of the multivibrator by the input signal. And with the gate voltage up and down in the staircase, the circuit state is inverted momentarily. By means of the logical product of the multivibrator output, an m/n frequency divider is obtained. (m and n are any positive integer and set on m<n) This divider has a wide synchronization range and a steady operation. And divided frequency ratio is conveniently changed with varying the coupling capacitor of the multivibrator. The calculated values are in agreement with the measured values.
Recently, one of thermal-type IR detectors, a pyroelectric sensor has a great interest in applications to commercial uses because of their working temperatures without cooling, a constant detectivity with independent of wavelength and a cheaper price. The conventional pyroelectric materials are usually normal ferroelectrics with a first or a second phase transition. The working temperatures prefer to enough below the Curie temperature, Tc, for the stable responsivity to temperature. Electric field induced-type pyroelectric sensors have also been proposed. Relaxor ferroelectrics such as Pb(Mg1/3Nb2/3)O3 (PMN) and Pb(Sc1/2Ta1/2)O3 (PST) which have a glassy Curie temperature near room temperature are used as this type sensors. This paper describes the sensor properties of electric field induced-type pyroelectric sensors prepared by using PMN and PST ceramics compared with the conventional type sensors. Material evaluations of PMN and PST ceramics were made to dielectric and pyroelectric properties. PMN shows the excellent induced pryoelectric properties for the sensors over wide range of temperature. On the other hand, PST seems to be inadequate for an IR detector because of a very narrow temperature range to high response. Prepared sensors with PMN and PST ceramics show enhanced pyroelectric activities under DC bias field. Measured sensor voltage-responsivities agree with those of calculated values for the PMN case. The electric field induced-type infrared sensor with the thick or thin film materials seems to be fine as linear array IR detectors for thermal imaging by means of application of higher electric field.
The streamer corona discharge is well known having useful industrial applications such as ozone generation, gas treatment, and plasma processing, etc. In this work, how generate the streamer corona continuously in model air (N279%+O221%) at atomospheric pressure has been investigated experimentally, and the current-voltage and frequencyinterelectrode resistance charachteristics for a plasma reactor based on the positive streamer corona discharge are discussed. The positive streamer coronas are generated and kept to be stable by the positive electrode equipted with the thin stainless steel fins, 0.1mm in thickness, 1.5mm in width, 3mm in length, for making a non-uniform field. The negative electrode is composed by sphere or plane electrode for the uniform field, and DC high voltage is applied during them. The results show that the positive streamer corona inception voltage decrease significantly with the effect of the fins. Increasing the number of fins increase the positive streamer corona current and increase the stability range of the positive streamer corona discharge. Moreover, the frequencyinterelectrode resistance charachteristics in the positive streamer corona discharge has been shown to be in inverse proportion to the magnitude of the interelectrode resistance.
This paper describes a new signal processing method for MUSE system. MUSE system uses two signal processing modes: a moving image processing mode and a stationary image processing mode. An intermediate mode between these two, if one were contrived, would provide higher picture quality for moving objects. Therefore, we examined the feasibility of such a hybrid mode encompassing the attributes of both the moving and stationary image domains. As a result, we have found that a mode which manipulates frames to process moving images is very effective in reproducing less moving objects more clearly. In the following, details of the hybrid image processing method and the results of computer simulation of this method will be discussed.
In this paper, we propose a new design method for a nonlinear control system using neural networks with mixed structure. The plant consists of linear part and nonlinear part. First, the plant is roughly approximated by a linear model. The optimal control gain and observer gain are determined for the approximated system by using linear system theory. We propose a neural network compensator for the approximated control system designed by linear system theory. A new control system has two neural networks for observer and regulator. A neural network for the observer is used to compensate a nonlinearity in state estimation. A neural network for the regulator is used to compensate optimization of the control system. The neural networks have a nixed structure which has recurrent connections only in a hidden layer since the neural networks with recurrent connection can represent a dynamical mechanism. Simulation results show the effectiveness of the orooosed method for nonlinear control problems.
It is required in servo-system design techniques that not only the steady robustness can be guaranteed and an arbitrary dynamics can be realized, but also a low parameter and disturbance sensitivity can be obtained. On the other hand in robot arm trajectory control, it is required that the arm can follow a target signal accurately and quickly. As a servo controller which can realize these requirments, we propose a new type controller called a multi input-output tracking controller. The characteristics of this controller are (1) the outputs can track perfectly target signals with one sampling delay, (2) the system is steady robust to step disturbances, (3) the sensitivity characteristic can be specified arbitrarily. This cotroller can be constructed for discrete time plants with both stable and unstable zeros. We solved the unstable zero problem by applying the 2-Delay input control technique.
The conventional sliding mode control is well known for its robust property to nonlinear uncertainty. However the assumption that the upper-bound of the uncertainty must be known and the excessive switching gain which is determined by the upper-bound always limit its application. For this, Slotine(2) proposed adaptive sliding mode control which partitiones the uncertainty into two parts: one is known structurally but unknown with unknown parameters, the other one is structurally unknown but its upper-bound is known. The first part is identified and compensated, so that switching gain can be decreased. However the upper-bound must be known as usual. For this, in this paper, we improve it making use of the idea of adaptive robust control law in which the upper-bound is estimated.
The most general technique that can be used to solve a wide variety of combinatorial extremumsearch problems, such as the task scheduling for paraller processing, is the branch-and-bound algorithm. However, since these problems are nondeterministic polynomial (NP)-complete, the solution derived within an appropriate searching time is not always optimal. In this paper, from the standpoint of avoiding local minimum solutions. we propose a practical scheduler which searches a wide area of the search tree using plural searching strategies based on the latest starting time and performing a systematic shuffling of the proority list for node selections. Our computer simulation results demonstrate that the systematic shuffling yields to dramatic acceleration in obtaining the candidate solutions. An application of the proposed scheduler to the simulation of the impact drop of a roll milling machine indicates that the total idle time of the processors can be decreased by a factor of ninety when compared to the conventional scheduling approachs. The observation of a mechanical manipulator control application reveales that the minimum number of processors for reaching the critical path length is five, with the conventional scheduling requiring seven.
Backpropagation is most widely used learning algorithm for neural networks. It has some well known drawbacks. The most important drawback is the difficulty in the choice of the values of various learning parameters. Inadequate values will result in a very slow convergence. In the most serious case, backpropagation process will often encounter a local minimum and thus been anable to solve a learning problem. These drawbacks stem from that backpropagation is a gradient based optimization procedure without linear search. In this paper, we propose a new learning algorithm, based on the use of solution method for nonlinear equations which represent the output errors. We basically use Newton method for nonlinear equations since it is superior in convergency. However, Newton method, in dependence on the initial point, does not ensure the global convergence. So, we employ the Homotopy continuation method, which is one of the parametric method, to overcome this drawback. The proposed method is tested for various learning problems. The computational results show that the proposed method is superior to backpropagation in convergency.