電気学会論文誌C(電子・情報・システム部門誌)
Online ISSN : 1348-8155
Print ISSN : 0385-4221
ISSN-L : 0385-4221
117 巻, 9 号
選択された号の論文の25件中1~25を表示しています
  • 中田 芳樹, 岡田 龍雄, 前田 三男
    1997 年 117 巻 9 号 p. 1173-1180
    発行日: 1997/08/20
    公開日: 2008/12/19
    ジャーナル フリー
    In order to investigate the behavior of species in the laser-ablation plume during the pulsed-laser deposition (PLD) process, the time-resolved two-dimensional laser-induced fluorescence spectroscopy (2D-LIF) and Rayleigh-Mie scattering (2D-RMS) systems have been developed. Results on the PLD process of hightemperature superconducting thin films of YBa2Cu3O7-xare presented. Firstly, the effects of Doppler and collisional quenching on the sensitivity of 2D-LIF are examined. Then the spatial distributions of Cu atoms observed by the techniques as well as emissive particles in the laser ablation process of YBCO are presented. The effect of the heated substrate on the propagation process of plume was observed for the first time. 2D-RMS was utilized to visualize the behaviors of particles directly ejected from the target and of the particles condensed in the gas phase.
  • 波長・出力・基板温度選択による構造及び物性制御
    西尾 悟, 佐藤 博保
    1997 年 117 巻 9 号 p. 1181-1186
    発行日: 1997/08/20
    公開日: 2008/12/19
    ジャーナル フリー
    Amorphous organic semieonductive thin films with electric conductivities ranging between 10-5 and 101 Scm-1 are prepared on several temperature-controlled substrates by cxcimcr laser ablation (ELA) of 3, 4, 9, 10-pcrylenetetracarboxylie dianhydride (PTCDA), with 193nm (ArF), 248nm (KrF) and 308nm (XeCI) beams. The structure, electric conductivity and carrier species of the films prepared strongly depend on the ablation wavelength, fluence and substrate temperature. In thermoelectric power measurements, conversion of carrier species from n-type to p-type is observed with increasing fluence of a 308nm beam from 0.2 to 4.0 Jcm-2pulse-1. A film prepared on a substrate at 300°C by ELA with a 308nm beam partially contains polyperinaphthalene (PPN) structure with electric conductivity of 10-2-10-1 Scm-1.
  • 林 健一, 桑原 尚, 徳永 康夫
    1997 年 117 巻 9 号 p. 1187-1193
    発行日: 1997/08/20
    公開日: 2008/12/19
    ジャーナル フリー
    The material processing by the excimer laser ablation is useful for the reasons of its high peak power, large absorption coefficient and small thermal effect. The laser processing of the polymer material is popular in the drilling of ink jet nozzles and the wire stripping. The ceramic materials are trying to process by the excimer laser ablation. Various ceramics were processed under the different values of energy density in our experiment. In ceramics processing, there exist some problems to be solved such as the debris, micro-cracks and ridges. To solve these problems are important for industrial use of the laser ablation. The authors found the processing in vacuum was suitable to avoid debris and micro-cracks. The debris could be easily removed from the surface by a weak laser radiation after the processing. Ridges disappeared when high energy density laser beam was irradiated and the ablated ceramics formed very smooth surface. Database systems about the laser parameters and the processing parameters including visual data were discussed.
  • 小海 文夫
    1997 年 117 巻 9 号 p. 1194-1198
    発行日: 1997/08/20
    公開日: 2008/12/19
    ジャーナル フリー
    Laser ionization time-of-flight mass spectrometry has been used to probe neutral species produced by laser ablation of graphite at 266nm and 1064nm. Arrival time profiles of the neutral species were measured by changing delay times of postionization. C3 was the major species. The yields of C and C2 normalized to that of C3 at 266nm (0.4-0.75J/cm2) were 0.16-0.17 and 0.13-0.15, respectively. The yields of C and C2 at 1064nm (0.7-1.0J/cm2) were 0.03-0.05 and 0.02-0.03, respectively. Arrival velocity distributions with different most probable or center-of-mass velocities were observed depending on ablation laser wavelength and fluence. However, C, C2, and C3 showed similar arrival velocity distributions. We discuss the formation of C, C2, and C3 in terms of a direct ejection process from the graphite surface and a collisional process in the ablation plume.
  • 田畑 仁, 川合 知二
    1997 年 117 巻 9 号 p. 1199-1205
    発行日: 1997/08/20
    公開日: 2008/12/19
    ジャーナル フリー
    The unique advantage of the pulsed laser deposition is its ability to produce highly oriented stoichiometric films at a low substrate temperature. Ferroelectric PbTiO3 thin films have been formed using 2nd laser assisted laser ablaion technique at low temperature, i.e., 350°C, on SrTiO3 single-crystal substrates and Pt/MgO electrodes. The second laser irradiation at the substrate surface is quite effective for crystallization of the films at low substrate temperature below 400°C. The suitable energy density (fluence) of the irradiation laser is in the range of 30-100mJ/cm2. X-ray diffraction patterns of PbTiO3 thin films show c-axis orientation, with a rocking angle of 1.0-0.5°. These films exhibit ferroelectric hysteresis loop. The dielectric constant and remanent polalyzation of the PbTiO3 films are in the range of 120-150 and 60-80μC/cm2, respectively.
  • Shunichi Sato, Kouki Shimizu, Yoichi Takashima, Yoshindo Soman
    1997 年 117 巻 9 号 p. 1206-1212
    発行日: 1997/08/20
    公開日: 2008/12/19
    ジャーナル フリー
    Laser etching characteristics of the stainless steel corrosion (oxide) films prepared under the nuclear-reactor-water-simulated conditions have been revealed. A Q-switched Nd: YAG laser of 12 ns pulse width was used to induce ablation processes. The laser beam was line-focused with a cylindrical lens and irradiated the traveling sample at different laser fluences. The etched surfaces were analyzed with a surface profiler, a scanning electron microscope, and an X-ray photoelectron spectroscope. It has been shown that the oxide films can be efficiently removed with a relatively low energy, 1064-nm laser pulse of 50mJ (2.3-2.9J/cm2). No metallic-elements-enriched layers were observed on the laser irradiated surfaces. The effects of the laser wavelength on etching characteristics was also investigated with the 2nd harmonics (532nm) and the 3rd harmonics (355nm) laser beams. It has been found that for the unoxidized samples (the bulk surface of the stainless steel) the etching efficiency increases with decreasing the laser wavelength. By contrast, no clear wavelength dependence of the etching efficiency was observed for the oxidized samples, an effect which may be attributed to relatively flat response of the absorption coefficient to the laser wavelength for the oxide films surfaces.
  • 西川 博昭, 官 文傑, 松本 卓也, 金井 真樹, 川合 知二
    1997 年 117 巻 9 号 p. 1213-1218
    発行日: 1997/08/20
    公開日: 2008/12/19
    ジャーナル フリー
    In order to understand the mechanism and dynamics of laser ablation, we have performed two kinds of experiments using alkaline earth metals as targets, i.e., Ca, Sr and Ba. The one is time- and space-resolved measurement of probe laser transmittance in the plume. The other is the measurement of time-of-flight distribution and the amount of desorbed monovalentions using quadrupole mass spectrometer. In the former experiment, the general dynamics of plume expansion produced by the laser ablation is investigated. It shows that the desorbed chemical species give two velocity distribution components. For the fast component, the initial expanding process of the plume is described as the adiabatic expansion. As the plume density becomes low, the temperature and velocity distribution is frozen because of no energy exchange among species. In the latter experiments, the relationship between amount of desorbed monovalent ion and laser fluence is obtained. On the results, it is found out that the ablation of alkaline earth metals is triggered by multiphoton photochemical reaction. On the basis of these results, we have proposed the schematic aspects of laser ablation phenomena.
  • 竹内 伸直, 中鉢 憲賢, 堀田 卓, 成田 憲一
    1997 年 117 巻 9 号 p. 1219-1224
    発行日: 1997/08/20
    公開日: 2008/12/19
    ジャーナル フリー
    To observe earth potential difference (EPD) signals induced by seismic waves a modified measuring method has been operated continuously at Aobayama in Sendai. This method is successfuly used to reduce the unnecessary signals induced by artificial and natural earth current sources. We have also obtained seismic waveform records from Earthquake Observatory of Tohoku University which is located apart 700m from us. The earth potential difference signals and three signals of seismic wave (accelaration, velocity and displacement) of the Kobe earthquake are displayed. To analyze these complicated signals the spectrum transform have been performed and compared each others. A streaming potential model associated with earth accelaration is adopted to explain the induced earth potential difference signal. We estimate the ratio of the earth electric field (potential difference) to the pressure difference inside the ground. This ratio is reasonable on the assumpution of the streaming potential model.
  • 岡 久雄, 中村 貴彦
    1997 年 117 巻 9 号 p. 1225-1231
    発行日: 1997/08/20
    公開日: 2008/12/19
    ジャーナル フリー
    The body structure influences the tactile sensation and stiffness evaluation. This suggests conversely an estimation of internal body structure and viscoelasticity from skin surface. This paper deals with an estimation of not only body structure but also viscoelasticity under skin surface, by using an acoustic random vibration (30-1000Hz). This method is characterized by morphological and functional measurement of inside of the body. The shape and viscoelasticity of internal object (tumor: silicone-gel) are estimated with a mechanical impedance, which is measured from the surface. The viscoelasticity and effective vibrating radius are calculated from the impedance spectrum and they give a relationship between a depth and viscoelasticity of internal object. The experimental silicone-gel model includes internally two hemisphere gels, whose viscoelasticities are different from that of peripheral silicone-gel. The 3-D image of internal hemisphere gel is reconstructed, based on the relation between a depth and the effective vibrating radius. The viscoelasticity of internal hemisphere gel is also estimated, based on the relation between the depth and the measured viscoelasticity of silicone-gel slope model.
  • 河野 正, 小南 昌信, 日下 浩次
    1997 年 117 巻 9 号 p. 1232-1237
    発行日: 1997/08/20
    公開日: 2008/12/19
    ジャーナル フリー
    A method for designing electromagnetically coupled (EMC) printed dipole arrays is presented. The advantages of EMC dipoles are greater bandwidth, higher efficiency and an easier match to the feed lines, when compared to classically fed printed antennas. The excitation mechanism is provided by a strip transmission line embedded inside the substrate which couples energy parasitically to the microstrip antenna. Electromagnetically coupled microstrip dipoles have been investigated by empirical or approximate analysis techniques, and an approximate model has been derived for the microstrip dipoles. The method is based on the moment method solution of integral equation derived from approximate Green's function for a grounded dielecric slab. The element lengths, offsets and a stub length of an EMC dipole array are determined by the direct search optimization method of Hooke and Jeeves. To show the feasibility of the method proposed, the mathematical formulation and numerical results are presented for Chebyshev arrays.
  • 斉 天, 川名 学, 平松 友康, 宮崎 道雄
    1997 年 117 巻 9 号 p. 1238-1244
    発行日: 1997/08/20
    公開日: 2008/12/19
    ジャーナル フリー
    The Genetic Algorithms (GA) which is often used as an optimization algorithms is based on simulation of nature genetic inheritance and Darwinian striving for survival. And it is well known because of its good performance for the optimization problems. However, when the GA is used as an optimization method, it has some problems. The most difficult one is that users must determine the GA's parameters which are used for genetic processing and often affect the results sensitively. So users need a great skill of GA. For the determination problem of GA's parameters, many methods have been proposed. But such methods seem to have a high interdependence with experience. In this paper, we propose a new method for dynamical control of the GA's parameters by using the intermittency chaos. This method was applied to the De Jong's 5 functions and the knapsack problem as GA's task problems. As the result, good performances of it were obtained. Algorithms of this method are also easy to be understood. So we can expect this method as a general method for the determination problems of the GA's parameters.
  • 横尾 裕規, 萩原 将文
    1997 年 117 巻 9 号 p. 1245-1252
    発行日: 1997/08/20
    公開日: 2008/12/19
    ジャーナル フリー
    To construct a face recognition system, facial regions should be detected before a face recognition process. A lot of studies have been made for the face recognition. However, the detection of facial region is still a difficult problem. In this paper, a novel method to detect human facial regions from an image with a complex background has been proposed. To detect human facial regions precisely, we employ the Genetic Algorithm (GA). First, the proposed method detects edges from an input image. Second, it searches ellipse regions using GA based on the idea that human faces can be approximated by ellipsoid. Finally, it judges whether the region is a human face or not one by one. We confirmed the validity and the effectiveness of the proposed method by computer simulations.
  • 小澤 誠一, 堤 一義, 馬場 則夫
    1997 年 117 巻 9 号 p. 1253-1258
    発行日: 1997/08/20
    公開日: 2008/12/19
    ジャーナル フリー
    In this paper, the association characteristics of Cross-Coupled Hopfield Nets (CCHN) proposed as a modular neural network model are discussed in an analytical way. In the CCHN, an arbitrary number of modules (Hopfield networks) can be mutually connected via feedforward networks called “internetworks”, whose outputs generate the interactions among module networks. To evaluate the CCHN as a modular neural network, it has been applied to associative memories so far. Although its excellent association performance is supported by many simulation results, it is still difficult to compute the memory capacity exactly and examine the dynamical properties rigorously, because the information processing of the CCHN includes strong nonlinearity. Hence, as the first step to the analytical approach, this paper focuses on a 1-module CCHN whose interaction is realized by a two-layered feedforward internetwork. In this case, the connection matrix of the CCHN degenerates into a single square-matrix like a conventional auto-association type of associative memory. Through the eigenvalue analysis for the connection matrix, we reveal that the essential differences between the association characteristics of the CCHN and a conventional auto-correlation associative memory originate from the dynamics in the noise-space which is the orthogonal complement of the subspace generated from memory patterns.
  • Min HAN, Kotaro HIRASAWA, Masanao OHBAYASHI, Hirofumi FUJITA
    1997 年 117 巻 9 号 p. 1259-1266
    発行日: 1997/08/20
    公開日: 2008/12/19
    ジャーナル フリー
    It has been already reported that the learning algorithm of Universal Learning Network (U.L.N.) by forward and backward propagation is useful for the modeling, managing, and controlling of the large scale complicated systems such as industrial plants, economic, social and life phenomena[1][2][7][8].
    Universal Learning Network is a network which can model and control naturally the large scale complicated systems and consists of nonlinearly operated nodes and multi-branches that may have arbitrary time delays including zero or minus ones. Therefore, U.L.N. can be applied to many kinds of systems which are difficult to be expressed as the ordinary first order difference equation with one sampling time delays.
    In this paper, a new method is presented in order to make an optimal modeling of the dynamic system using U.L.N., where the word “optimal” means the establishment of the compact model as much as possible and the building of the model that has better performances. For the compactness of the modeling, a special filtering mechanism on all of the branches that cuts unnecessary branches are introduced. From simulation results, it has been clarified that selecting appropriate parameter variables in U.L.N. can make the compromised modeling in terms of modeling error and compactness of the model.
  • 荒木 獻次, 陳 星橋
    1997 年 117 巻 9 号 p. 1267-1273
    発行日: 1997/08/20
    公開日: 2008/12/19
    ジャーナル フリー
    This paper describes a control scheme for on-line control of a spot welding machine. From the simulation of the soft-steel welding process by constant current, constant power and variable power methods with numerical analysis by FEM, nugget sizes etc. obtained are compared for these three methods. Variable power method is the best one for obtaining the biggest nugget and is used here as a reference (welding energy) model. And a fuzzy adaptive controller (FAC) instead of an usual fuzzy controller (FC) is developed and used for the improvement of the control accuracy of the spot welding machine. The membership functions of the fuzzy adaptive controller will be modified in real time with the adaptive algorithms according to distinct operating stages and working environment. Combining the reference energy model and FAC, a model reference fuzzy adaptive control (MRFAC) scheme is developed. The results of simulation and experiments with a dummy circuit of the spot welding system show that it is possible to control welding energy with on-line. Simulations also show that the control performance of the welding system with FAC becomes better than that with FC. And the experimental result indicates that the MRFAC system has also robustness for parameters changes, welding stage changes and disturbance.
  • 瀧森 徹, 小川 哲朗, 西田 眞
    1997 年 117 巻 9 号 p. 1274-1280
    発行日: 1997/08/20
    公開日: 2008/12/19
    ジャーナル フリー
    In order to examine the confines of electrical stimulation in layer 2/3 of visual cortex in the brain slice preperation, we estimated the effective range of the stimulation based on the excitatory postsynaptic potential (EPSP) evoked in layer V neuron which receives input from layer 2/3. For this purpose, we recorded and compared EPSPs amplitudes evoked by stimulations at directly over site of recording electrode and lateral site in layer 2/3. Since the EPSP increased linearly with stimulus intensity before the saturation, it was considered that the EPSP correlates with the number of projecting neurons in area directly excited with the stimulation. Then we formed the region model by which we can get the ratios between the neuron numbers in areas excited by different sites stimulation against the stimulus effective ranges. And in the stimulus intensity for action potential threshold of layer 5 neuron, we evaluated the effective range for the relative values of EPSPs to be produced with the stimulations of 250μm lateral site and directly over site. In the model, the ratio increased monotonically with the effective range and in the case of 250μm for the effective range, the ratio bewteen those EPSPs was less than the value in the model. These results led the conclusion that the effective range of the intensity for layer 5 neuron to generate the output is confined within 250μm from directly over site, that is, within layer 2/3.
  • 小原 和博, 中村 行宏
    1997 年 117 巻 9 号 p. 1281-1290
    発行日: 1997/08/20
    公開日: 2008/12/19
    ジャーナル フリー
    We investigate how training patterns should be presented to a back-propagation neural network (BPNN) so as to train the BPNN with a small deviation of training patterns and to improve the BPNN's learning speed. First, we explain the problem with a conventional learning technique, in which all training patterns are presented to a BPNN equally. Then, we propose a selective presentation of training set to a BPNN. In a proposed technique, using several criterion values for both the mean summed squared error and individual summed squared errors, we detect poorly-trained patterns and present them more often. The effectiveness of the proposed technique is confirmed by evaluation experiments using mesh patterns extracted from handwritten digits and two-dimensional Gaussian distribution data.
  • 柴田 克成, 岡部 洋一
    1997 年 117 巻 9 号 p. 1291-1299
    発行日: 1997/08/20
    公開日: 2008/12/19
    ジャーナル フリー
    In order to realize the mapping from spatial information to temporal information, Temporal Smoothing (TS) Learning is proposed. In this learning, the output of a learning unit, to which sensory signals are given as input, is trained to be smooth along time. In other words, the learning unit is trained so as that the second time derivative of the output itself becomes 0.
    This learning can be applied to integrate local sensory signals into an analog spatial signal. It also can be used that an agent learn evaluation function in delayed reinforcement learning on behalf of TD Learning(8)(7) when only one target state is chosen. When a neural network was employed as a learning unit and visual signals were given as inputs directly, the hidden neurons in the neural network represented spatial information and had a adaptability of changing the representation according to the agent's motion characteristics.
  • 兪 文偉, 横井 浩史, 嘉数 侑昇
    1997 年 117 巻 9 号 p. 1300-1307
    発行日: 1997/08/20
    公開日: 2008/12/19
    ジャーナル フリー
    In this paper, we propose the State Grouping scheme for coping with the problem of scaling up the Reinforcement Learning Algorithm to real, large size application. The grouping scheme is based on geographical and trial-error information, and is made up with state generating, state combining, state splitting, state forgetting procedures, with corresponding action selecting module and learning module. Also, we discuss the Labeling Based Evaluation scheme which can evaluate the opportunity of the state-action pair, therefore, use better experience to guide the exploration of the state-space effectively. Incorporating the Labeling Based Evaluation and State Grouping scheme into the Reinforcement Learning Algorithm, we get the approach that can generate organized state space for Reinforcement Learning, and do problem solving as well. We argue that the approach with this kind of ability is necessary for autonomous agent, namely, autonomous agent can not act depending on any pre-defined map, instead, it should search the environment as well as find the optimal problem solution autonomously and simultaneously. By solving the large state-size 3-DOF and 4-link manipulator problem, we show the efficiency of the proposed approach, i.e., the agent can achieve the optimal or sub-optimal path with less memory and less time.
  • 阿部 紘士, 石渡 厚
    1997 年 117 巻 9 号 p. 1308-1316
    発行日: 1997/08/20
    公開日: 2008/12/19
    ジャーナル フリー
    In current railways, many railway information systems have been individually introduced for safety operation of trains and certainly practice of operation plan. And, it has not been executed positively for even long-distance passengers of the ordinary trunk lines to serve amusements like music and pictures and railway information.
    This paper presents a concept of multimedia railway information and communication system (MURICS) for train operation and passengers, and proposes a new radio communication system between trains and ground facilities which adopts subcarrier multiplexed lightwave system in feeder links and 45GHz band millimeter-wave radio communication system in radio links.
    Also, this paper describes performance estimation of this proposed system by computer simulation, and shows that this system is able to put to practical use by results of estimation.
  • 朴 炳植, 有本 慎司, 鈴木 胖
    1997 年 117 巻 9 号 p. 1317-1324
    発行日: 1997/08/20
    公開日: 2008/12/19
    ジャーナル フリー
    A model has been developed to estimate optimal mix of traffic types under environmental constraints in the future. The model constructed is a linear-programming model, where the share of each traffic means among zones is determined so as to minimize the generalized cost obtained as a sum of the trip cost and travel time converted into a money term under various constraints, such as the capacity of each transportation facility and the permissible amount of pollutant emission. By applying to the actual data of the OD survey in 1990 in Osaka, it has been confirmed that the model can explain fairly well the actual situations of modal split among different traffic means. The impacts of policies of imposing control on nitrogen oxide (NOx) emission as well as carbon dioxide (CO2) emission were investigated for Osaka city in the year 2010. It has been shown from scenario analyses using the model that it would be impossible to satisfy the environmental constraints only by reducing unit NOx exhaust coefficient of private car and taxi, and that electric car and natural-gas firing bus should be introduced to the rate of 3.0-3.1% and 0.97-1.2% of the total traffic means, respectively.
  • 豊田 計時, 塚元 康輔, 宮田 武雄
    1997 年 117 巻 9 号 p. 1325-1326
    発行日: 1997/08/20
    公開日: 2008/12/19
    ジャーナル フリー
  • 崔 賢洛, 徐 粒, 斉藤 制海
    1997 年 117 巻 9 号 p. 1327-1328
    発行日: 1997/08/20
    公開日: 2008/12/19
    ジャーナル フリー
  • 景山 陽一, 西田 眞
    1997 年 117 巻 9 号 p. 1329-1330
    発行日: 1997/08/20
    公開日: 2008/12/19
    ジャーナル フリー
  • 辻 三郎
    1997 年 117 巻 9 号 p. 1331-1337
    発行日: 1997/08/20
    公開日: 2008/12/19
    ジャーナル フリー
feedback
Top