巻号一覧
30 巻, 10 号
選択された号の論文の20件中1～20を表示しています
- |<
- <
- 1
- >
- >|
-
吉田 篤正, 拝田 健, 松岡 啓充, 鷲尾 誠一1994 年 30 巻 10 号 p. 1133-1140
発行日: 1994/10/31
公開日: 2009/03/27
ジャーナル フリーContinuous measurements by the spectral bi-angular reflection method are carried out for frost layer growing at the atmospheric pressure and the room temperature in the surrounding atmosphere. A four-flux model of radiation transfer is introduced to analyze the measured data. The spectral absorption and scattering coefficients are evaluated in connection with the bulk density of frost layer; The absorption coefficients at the absorption band of ice prove to be proportional to the bulk density of frost layer. The results of the optical measurements show the necessity to consider the stratified structure of frost layer in the 'crystal growth period'.抄録全体を表示PDF形式でダウンロード (1238K) -
田中 正吾, 山根 健治1994 年 30 巻 10 号 p. 1141-1150
発行日: 1994/10/31
公開日: 2009/03/27
ジャーナル フリーUnknown curved-surface can be estimated by a linear combination of a set of adequate basis functions. We previously showed that when a sufficient number of observation points are available and the observation points are distributed uniformly in the region, a set of Gaussian type functions with a uniform dispersion, whose centers are allocated in a regular triangle lattice, can be efficiently used as the basis functions.
This paper proposes optimal basis functions with a nonuniform dispersion based on the distribution pattern of the observation points for the case where observation points are nonuniformly distributed and the number of observation points is not so many.抄録全体を表示PDF形式でダウンロード (2054K) -
長棟 章生, 手塚 浩一1994 年 30 巻 10 号 p. 1151-1157
発行日: 1994/10/31
公開日: 2009/03/27
ジャーナル フリーA method of coded pulse compression employing two maximal length sequences (M-seq.) which are the same in code but generated by slightly different clock frequencies is newly applied to subsurface radars. This method practically gains the sensitivity of the signal without any sacrifice of resolution.
The M-seq. pulse-compression underground survey radars are theoretically analyzed and the design criteria are obtained. The waveform made by phase modulation of the M-seq. over the carrier synchronized by the clock frequency generating the M-seq. is found to be more effective than the original M-seq. as transmitting wave.
Two prototypes, the surface scanning underground radar and the cross bore-hole radar, are tested and confirmed to be highly sensitive with enough resolution, showing their promising prospect such as inspection of buried pipes in the deeper underground and more precise survey of stratum structure than the conventional subsurface radars with mono-pulse.抄録全体を表示PDF形式でダウンロード (1770K) -
花泉 弘, 藤村 貞夫, 山岸 健太郎1994 年 30 巻 10 号 p. 1158-1164
発行日: 1994/10/31
公開日: 2009/03/27
ジャーナル フリーA method is proposed for measurement of 2-dimensional temperature distribution in the IFOV (Instantaneous Field Of View) of an thermal infrared radiometer. This method resolves the IFOV into many cells for getting 2-dimensional temperature distribution, though the former methods synthesize measurements obtained from each IFOV of array sensor. This method is useful to measure 2-dimensional temperature distribution of object whose size is almost the same as the IFOV without any scanning mechanism and/or optics.
In the proposed method we measure projection data in the IFOV of an thermal radiometer, and reconstruct 2-dimensional temperature distribution from the projection by using CT (Computer Tomography) algorithm. Temperature distribution is obtained by compensation of the measured distribution by sensitivity distribution in the IFOV of the radiometer.
In this paper, we describe the principle, procedures of the method and measurement system. Application result of the method to practical evaluation of performance of cavities is also shown.抄録全体を表示PDF形式でダウンロード (2671K) -
宮元 慎一, 有永 真司1994 年 30 巻 10 号 p. 1165-1171
発行日: 1994/10/31
公開日: 2009/03/27
ジャーナル フリーThis paper presents design of H∞-control for plants that have integrators which have been difficult to be handled in the H∞-mixed sensitivity problem.
In output feedback control based on the mixed sensitivity problem, the plant must not have poles at the origin of the complex plane because of the restrictions of the 2-Riccati algorithm. However, there are many plants that have integrators such as motors and vehicles. Some design methods for plants that have integrators have already been proposed, but in the methods the controllers have no integrators. Therefore if the stepwise disturbances are injected at the input or the states of the plant, the error between the output of the plant and the reference will not converge to zero.
The proposed design is based on partial state feedback control which uses some states of the plant. As in the partial feedback control proposed by Zhang and Hosoe, there is no limitation on the pole of the plant at the origin, it can be applied to the plants that have integrators. The proposed control system has the integrators in the controller, so even if the disturbances are injected at the input or the states of the plants the error will converge to zero. Furthermore in the proposed control system, the degree of the controller can be reduced by the number of the measurable states.
This paper shows the difficulties of the output feedback control in the mixed sensitivity problem, the design method based on the partial feedback control and the application to motion control.抄録全体を表示PDF形式でダウンロード (1179K) -
石原 正1994 年 30 巻 10 号 p. 1172-1177
発行日: 1994/10/31
公開日: 2009/03/27
ジャーナル フリーFor discrete-time non-minimum phase plants, the feedback property achieved by enforcing the conventional loop transfer recovery technique is discussed in this paper. Considering digital implementation accounting the computation delay in the controller, we use a discrete-time LQG controller including a prediction type Kalman filter. For simplicity, we assume that a non-minimum phase plant has a single real unstable zero. First, we give an explicit representation of the sensitivity matrix achieved by enforcing the conventional loop transfer recovery procedure at the plant input. The directional property between the target and the achievable sensitivity matrices is clarified. In addition, we show that the sensitivity matrix obtained by enforcing the conventional procedure coincides with that obtained by the partial loop transfer recovery technique which attempts to recover the feedback property related to the minimum phase part of the plant. This fact provides a clear system-theoretic meaning to the enforcement of the loop transfer recovery technique for non-minimum phase plants.抄録全体を表示PDF形式でダウンロード (839K) -
西山 清, 佐藤 治1994 年 30 巻 10 号 p. 1178-1187
発行日: 1994/10/31
公開日: 2009/03/27
ジャーナル フリーA nonlinear filter based on complex Kalman filter is proposed for estimating a sinusoidal signal and its parameters from the observed signal corrupted by white noise. In the proposed filtering algorithm, the forgetting factor, local iteration and the other techniques are employed in order to upgrade the filter performance. Furthermore, the roles of such techniques above mentioned are also analyzed theoretically. From the results, it is derived that the forgetting factor keeps the mean square error used in the filter more than a certain value, and the iteration improves the accuracy of the estimates by reducing the effects of modeling error and observation noise. On the other hand, it is demonstrated by some simulations that the proposed nonlinear filter is effective as a method for estimating a sinusoidal signal and its frequency under the low SN ratio.抄録全体を表示PDF形式でダウンロード (1502K) -
山田 陽滋, 三田 賢志, 藤田 和秀, 土田 縫夫, 今井 孝二1994 年 30 巻 10 号 p. 1188-1194
発行日: 1994/10/31
公開日: 2009/03/27
ジャーナル フリーThere have been a number of slip sensors proposed and developed for controlling the grasping force of robot hands in order not to drop objects. However, detecting only slips is insufficient for control since it may be too late to increase the grasping force after detecting the object's slips and the friction system of the contact part changes to be dynamic. In the paper, we stress the necessity of detecting the static friction coefficient of the contact surface between an object and a robot hand, and describe the development of a sensor and an algorithm to detect the coefficient: The sensor mounted on the hand detects three-axis contact force components and acquires slip information due to stick-slip vibration at the contact surface. The algorithm to integrate the contact force and the slip information enables the robot to detect the static friction coefficient in the active motion of lifting the object with the hand. Further experiments verify that the proposed method in the study allows a two-fingered parallel hand to detect the static friction coefficient for various surfaces and demonstrate adaptive grasping force control of the hand in response to changes in the weight of the grasped object by using the detected static friction coefficient.抄録全体を表示PDF形式でダウンロード (1568K) -
田沼 正也1994 年 30 巻 10 号 p. 1195-1201
発行日: 1994/10/31
公開日: 2009/03/27
ジャーナル フリーFor stable and efficient transient analysis of complicated block diagram used in process simulation, quasi state variable method is newly proposed. For stiff system appeared in many fields, implicit formula is known to be robust for stabilizing the transient analysis. One of the problems occurred in using implicit formula is to solve simultaneous equations at each time step. The state variable model can produce the smallest computation model but it is difficult to develop state variable equation analytically from nonlinear block diagram.
The proposed method generates numerically the computation model equivalent to that derived from state variable model by applying an automatic partial derivatives computing algorithm to block diagram of a kind of directed graph. The proposed method has several advantages. (1) The model of small dimension permits the efficient analysis. (2) The computation model for nonlinear block diagram can be generated by regular procedure. (3) The proposed method can be applied to steady state analysis.
Simulations for steam in super heater of boiler show that the quasi state variable method is effective for realizing stable and efficient analysis of stiff systems.抄録全体を表示PDF形式でダウンロード (1104K) -
杉江 俊治, 田内 豊1994 年 30 巻 10 号 p. 1202-1208
発行日: 1994/10/31
公開日: 2009/03/27
ジャーナル フリーIn this paper, we propose a robust controller design method for magnetic levitation systems based on H2/H∞ control theory. As a feedback controller, an H∞ suboptimal controller is designed via the loop shaping procedure to obtain the satisfactory feedback properties. And in order to obtain the desired command response, an H2 optimal controller is adopted. The obtained controller also achieves the robust tracking for step command signals. The effectiveness of the controller is evaluated by experiments of magnetic leviation.抄録全体を表示PDF形式でダウンロード (1736K) -
対称なシステムに対する対称なコントローラの設計池田 雅夫, 竹内 和啓1994 年 30 巻 10 号 p. 1209-1215
発行日: 1994/10/31
公開日: 2009/03/27
ジャーナル フリーTransfer matrices of structures with collocated sensors and actuators are symmetric. The symmetry is independent of the entities of mass, damping, or stiffness matrices, and is a nonparametric nature. This paper first shows that for such systems, symmetric controllers are more robust than nonsymmetric ones in stabilization. Then, to utilize the state-space methods in designing symmetric controllers, the class of state equations for symmetric systems is given. The property of state equations of symmetric systems suggests us how to choose the state-feedback gains and observer gains to obtain symmetric observer-based-state-feedback controllers. Finally, symmetric LQG, H2, and H∞ controllers are designed under proper assumptions.抄録全体を表示PDF形式でダウンロード (1398K) -
最適ディジタル制御による定速昇温制御玉城 史朗, 大城 尚紀, 山本 哲彦, 金城 寛, 平良 初男, 棚原 朗1994 年 30 巻 10 号 p. 1216-1224
発行日: 1994/10/31
公開日: 2009/03/27
ジャーナル フリーThermoluminescence (TL) is a calorescence phenomenon observed while heating minerals. TL intensity depends on the total quantity of radiation dose in the past. Due to this characteristic, measuring of TL intensity of natural minerals has been paid much attention as a useful method of age-estimation in archaeological and geological fields. Since TL intensity depends on the heating rate of the specimen, a constant rising temperature control system is necessary. But, even a commercial apparatus for TL measurement usually fails the rising temperature characteristic.
This paper progress a new TL age-estimating method. To establish for this, a temperature control system is designed and a measurement system for TL intensity is developed. The optimal digital feedback control algorithm based on impulse response of the system is applied to this temperature control system. The control result shows that the system can precisely control the constant rising temperature within permissible error of 1 degree for the temperature range of 150-400 degree centigrade. Then the system is introduced to TL age-estimation system of fossil-oyster.
The age is estimated by calculating the proportion of the annual cumulative dose at sampling area to the sum of natural cumulative dose that obtained by integrated TL intensity measured at a constant rising temperature. This result is reasonable.抄録全体を表示PDF形式でダウンロード (3094K) -
山田 克彦, 吉河 章二, 吉田 憲正, 小山 浩1994 年 30 巻 10 号 p. 1225-1233
発行日: 1994/10/31
公開日: 2009/03/27
ジャーナル フリーThis paper deals with control schemes of relative motion between two spacecraft (chaser and target). Two cases are considered: 1) the case where relative 6 d. o. f. (degrees of freedom) are all controlled and 2) the case where relative LOS (Line Of Sight) angles are controlled. The basic idea of the control schemes is the input-output linearization. The control schemes are derived systematically by using the equations of motion about the controlled variables. In the relative 6 d. o. f. control, linear parameters are used to express the relative attitude of the chaser to the target. The relation between their time derivatives and the angular velocities of both spacecraft is derived. Based on this relation, we obtain the control scheme of the relative 6 d. o. f. control in a simple form. In the LOS angles control, we propose to control the LOS angles of the chaser to the target and the chaser attitude in the inertial space separately. This control law easily realizes the fly-around motion of the chaser to the target without sensing the inertial position of the chaser. Numerical simulations have been executed to verify the proposed control schemes.抄録全体を表示PDF形式でダウンロード (1614K) -
山田 孝行, 藪田 哲郎1994 年 30 巻 10 号 p. 1234-1241
発行日: 1994/10/31
公開日: 2009/03/27
ジャーナル フリーThere have been many studies on the application of neural networks to servo controllers in order to design flexible controllers which have nonlinear mapping capabilities. Most of them are studies on series controllers which use the neural network output as the object plant input. Only M. Kawato has proposed a parallel controller which uses both the neural network output and the conventional control output as the object plant input. However, his proposed controller is analog rather than digital. The stability of a digital controller is different from that of an analog controller. We should investigate digital controller stability in order to construct a neural network servo controller which uses software.
Therefore, this paper proposes an adaptive type feedforward feedback controller which uses a neural network. The proposed controller is a parallel type controller. It is designed for digital control systems and its stability condition is investigated. Simulated results for a second order plant confirm both its effectiveness and characteristics.抄録全体を表示PDF形式でダウンロード (1229K) -
市川 芳明, 石井 良和1994 年 30 巻 10 号 p. 1242-1250
発行日: 1994/10/31
公開日: 2009/03/27
ジャーナル フリーThis paper presents methods to retain diversity of the allele distribution in the search of Genetic Algorithms. The purpose of this study is to refine the convergence behaviors of Genetic Algorithms for use in multi-variable optimization. We first evaluate“rareness”of a chromosome using the distribution of alleles in a population. Then the fitness of a chromosome is adjusted with the rareness so that rare chromosomes will be likely to survive. In addition, “mutation width”is introduced to tune the effect of the mutation which can generate rare chromosomes while conserving effective convergence.抄録全体を表示PDF形式でダウンロード (1656K) -
田上 隆徳, 宮本 定明, 最上 義夫1994 年 30 巻 10 号 p. 1251-1259
発行日: 1994/10/31
公開日: 2009/03/27
ジャーナル フリーThe aim of the present paper is to show effectiveness of Genetic Algorithm (GA) for cluster analysis problems in which the classification criteria are not the Euclidean distance.
GA is known as a most effective method to solve combinatorial optimization problems by simulating the process of natural evolution and natural genetics. In a traditional GA, its dispersion of the evaluation function will be large, depending on the operators using random numbers or the control parameters. In this paper, for the purpose of reducing the dispersion of the evaluation function, a concept of the degree of population diversity is introduced as an index for an internal state of the whole population, and we use this index as a control parameter of the genetic operators such as crossover, mutation and selection.
From numerical simulations, the following results are obtained: 1) The GA using the concept of the degree of population diversity is superior to the traditional GA with respect to the convergence of the evaluation function and the reduction of the variance of the evaluation function. 2) When data have observation noises, the GA using the proposed method shows robustness for data classification, while the classical method of clustering shows sensitive results.抄録全体を表示PDF形式でダウンロード (1735K) -
西村 金太郎, 宮内 秀和, 藤瀬 健領, 大谷 敏昭, 岡田 三郎, 横川 洪1994 年 30 巻 10 号 p. 1260-1262
発行日: 1994/10/31
公開日: 2009/03/27
ジャーナル フリーPolished curved surfaces were measured with a QUESTAR type tele-microscope. Introduction of image processing techniques enabled us to detect scratches on the polished surface on-line.抄録全体を表示PDF形式でダウンロード (1775K) -
花泉 弘, 藤村 貞夫, 豊田 弘道1994 年 30 巻 10 号 p. 1263-1265
発行日: 1994/10/31
公開日: 2009/03/27
ジャーナル フリーIn the wide band radiation thermometry, variation of spectral emissivity in an observation band causes systematic error in measured temperature. Usually, we measure reflectance, derive emissivity from the reflectance and estimate object temperature by solving radiation equation. In these procedures, we assume that the reflectance and the emissivity do not depend on object temperature and that sum of them is unity. However, the variation of spectral emissivity yields temperature dependency of emissivity and reflectance.
In this paper, we propose a method for obtaining object temperature without the error. In this method, we use a variable temperature radiator. Reflectances (we call them differential refrectance) measured at various temperatures using the radiator are transformed to (interval) emissivities at the same temperature, respectively. By solving radiation equation with interpolation of the neighboring emissivities, we get object temperature with high accuracy. Here, we describe the principle and the procedures of the proposed method. The validity of the method is confirmed by a numerical simulation.抄録全体を表示PDF形式でダウンロード (353K) -
長堂 勤, 示村 悦二郎, 石田 力, 内田 健康1994 年 30 巻 10 号 p. 1266-1268
発行日: 1994/10/31
公開日: 2009/03/27
ジャーナル フリーIn this paper, a new controller reduction algorithm by indirect method with riccati equations is proposed. Low order controller is obtained by transforming the state-space description of a controller to balanced coordinates using two riccati equations and truncating some state-spaces.抄録全体を表示PDF形式でダウンロード (379K) -
太田 淳, 久村 富持1994 年 30 巻 10 号 p. 1269-1270
発行日: 1994/10/31
公開日: 2009/03/27
ジャーナル フリーPetri net is a model for discrete event systems which has modeling power more than finite automaton and less than Turing machine. We show that Petri net with single-weighted arcs can simulate Turing machine under free firing rule.抄録全体を表示PDF形式でダウンロード (199K)
- |<
- <
- 1
- >
- >|