All issues
Volume 22, Issue 9
Displaying 1-14 of 14 articles from this issue
- |<
- <
- 1
- >
- >|
-
Sirou TAMAKI, Morikazu TAKEGAKI, Hiroyuki EGUTI, Kouji NAKANO, Suguru ...1986 Volume 22 Issue 9 Pages 921-927
Published: September 30, 1986
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper proposes a method of designing an optimal IFIR (Integrated Finite Impulse Response) controller which minimizes the weighted mean square value of both output error and control input based only on systems impulse response. Since the controller is constructed as a FIR filter, the computation for designing the optimal one is reduced to solving a discrete-time Wiener-Hopf equation. This equation can be solved recursively by increasing the order of the system via a type of Levinson's algorithm. Hence, the order of the controller can be estimated by evaluating the power of the output error and control input. At first the optimal design of controller is treated for an uncorrelated reference output. Next, the design method is extended to the case of colored reference output, which yields a model matching method in frequency domain. Finally, the effectiveness of the proposed method is demonstrated by computer simulations.View full abstractDownload PDF (779K) -
Tadashi KONDO1986 Volume 22 Issue 9 Pages 928-934
Published: September 30, 1986
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn this paper, a revised GMDH (Group Method of Data Handling) algorithm, which can estimate directly degree of the complete polynomial, is developed. In the previous GMDH algorithms, the optimal partial polinomials, which are generated by using the second degree polynomials of two variables in each selection layer, are accumulated in multilayered structure to construct the complete polynomial, and so, degrees of the variables which construct the complete polynomial become positive integral multiples of the input variables. Therefore, the system, which can be described by the complete polynomials with degrees of the non-integers, are very difficult to be identified by the previous GMDH algorithm. The revised GMDH algorithm in this paper generates optimal partial polynomials, which can estimate directly degree of the complete polynomial in each selection layer, and constructs the complete polynomial by the combinations of them. Therefore, over the many kinds of the nonlinear systems can be identified by the revised GMDH algorithm. The revised GMDH algorithm is applied to a simple illustrative example and compared with the result obtained by the previous GMDH algorithm.View full abstractDownload PDF (992K) -
Osamu YOSHIE, Shigeru FUJIMURA, Hiroki MIYACHI, Kageo AKIZUKI1986 Volume 22 Issue 9 Pages 935-941
Published: September 30, 1986
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSAmbiguous concepts play an important role in the reasoning that appears in our mental activity. Many expert systems have been developed as models of such kind of activities recently, but most of them cannot deal with ambiguous expression directly. This leads to difficulty of the knowledgebase construction and the input of ambiguous facts.
So we propose the knowledge representation dealing with ambiguous concepts and top-down reasoning method using it. We're able to treat ambiguous knowledge directly and get a satisfactory conclusion from the fewest facts by this proposed method.
The efficiency of this method is demonstrated via an examble.View full abstractDownload PDF (1007K) -
Hajime ASE, Mitsuo KOYANAGI1986 Volume 22 Issue 9 Pages 942-947
Published: September 30, 1986
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSAiming to reduce consumption amount of ammonia (NH3), which is blown into the catalytic denitrification process of the refuse incinerator plant, application of the periodic control method has been studied.
Through a step response examination in an actual plant, remarkable process dynamic characteristics have been observed: namely, that time-constant (Tu) of blowing in NH3 is smaller than (Td) of blowing off.
Applying the optimal periodic control theory to the process, the following analytical results are obtained: (i) Optimal bang-bang type periodic control of NH3 blowing is more effective in reducing NH3 consumption than optimal continuous steady state control, and (ii) The NH3 consumption reduction rate depends on the Td/Tu ratio and the ζ parameter (ζ=aimed denitrification rate/process capable denitrification rate)
Because the theoretical solution of the optimal control period is approximately zero, a practical finite periodic control has been examined on an actual catalytic denitrification process. Results of the examination have shown that the NH3 reduction level is acceptable but lower than the theoretical value.
In actual plants, because of wide fluctuation of nitrogenoxide (NOx) generation, simple periodic control cannot maintain exhaust NOx concentration under the required NOx concentration for a long time. The periodic control with feedback correction of control timing has been proposed and demonstrated to be effective by simulation.View full abstractDownload PDF (682K) -
Shigeyuki HOSOE, Kihachi TAKEICHI, Satoshi KUMAI, Masami ITO1986 Volume 22 Issue 9 Pages 948-954
Published: September 30, 1986
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn this paper we give some theoretical consideration to biped locomotion of a 3-link walking robot, which is controlled by high gain feedbacks around two joints at the hip.
When the gains are high enough, the dynamics of the robot can be uniformly approximated, on any finite time interval, by the one of a perpendicular pendulum.
Using this result and analyzing rigorously the one cycle of walking, we reveal why stable periodic motion is realized through high gain feedback control, and give some gait conditions for the realization.
The method and results are easily extended to the biped robot which has 4 or more links.View full abstractDownload PDF (741K) -
Hideaki ARAI, Atsushi FUJIMORI, Tomiji HISAMURA1986 Volume 22 Issue 9 Pages 955-961
Published: September 30, 1986
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn industrial situation such as FMS, there arises a need to optimize a schedule for a repetitive process in which one process assigns n jobs to m machines according to the prescribed ordering, e.g. a flow-shop or a job-shop problem, and the process is repeated many times.
In this paper, a Time Petri Net (TPN) is applied to a job-shop type repetitive process to obtain the time-suboptimal (non-delayed) schedule. The method is based on a reachable tree including time and the fact that the natural firing rule for TPN corresponds to a generation of a non-delayed schedule.
Then, based on the above result, the optimal restoring strategies are investigated which can be applied when the process undergoes an emergency stop and make a fresh start to restore to the above repetitive schedule after repairing.
Two cases are considered for restoring conditions. The results are illustrated by computer simulations.View full abstractDownload PDF (950K) -
Masaru UCHIYAMA, Fan LIU, Kyojiro HAKOMORI1986 Volume 22 Issue 9 Pages 962-967
Published: September 30, 1986
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSTo measure velocity distribution in unsteady fluid flow is very difficult but very important to analysis and control of fluid devices and biological systems. As to this problem, we have proposed a model-based method by which transient flow velocity profiles in a pipe are computationally estimated.
In this paper, we apply this method to the measurement process of a hot film velocimeter and realize a Kalman filter which estimates unsteady laminar flow velocity profiles in a pipe using the velocimeter.
First, the process of measurement with the velocimeter is analyzed theoretically and experimentally and it is shown that the velocimeter can be substantially regarded as a point-measuring velocimeter which takes measurement of flow velocity at the center point of a hot film sensor without disturbing the flow around it, if the length of the sensor is short enough compared to the inside diameter of the pipe in which the sensor is inserted.
Based upon the above result, a Kalman filter to estimate the velocity profile is derived, implemented on a personal computer, and executed in real time, yielding successfull estimation of fluid velocity profiles on the CRT display of the computer.View full abstractDownload PDF (1771K) -
Toshifumi TSUKIYAMA, Thomas S. HUANG1986 Volume 22 Issue 9 Pages 968-975
Published: September 30, 1986
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn order to navigate, vehicles need to deduce their own movements or their positions in given environments. This paper presents a method of motion stereo for finding a vehicle's motion from the 2-d image data taken with a camera mounted on the vehicle.
In motion stereo methods, point pairs which correspond to the same object points must be found in the two images and the object points must then be located in the 3-d world at every imaging point. Conventional matching methods have dealt with the case where apparent disparities of point pairs corresponding to the same object points in the two images are relatively small. From the view point of application to navigation of vehicles, taking images at long intervals is desirable, because it requires much computation to process the image data of complex scenes. However, then the camera's direction changes by a large angle from one image to the next. This paper deals with motion stereo using a sparse sequence of single TV images. The approach uses vertices of objects observed in a scene for matching the two images. However, to do matching based on structural information such as vertices of objects derived from 2-d images has some ambiguity. Therefore, range information of vertices of objects is used to find candidate pairs for matching. In addition, the matching is done with the guidance of a base line in the 3-d world which can be seen in both images.
The candidate vertices for matching are obtained from analyzing the edge maps of input images using a junction type table and range information of the segments which form the vertices. Range information is obtained by inverse perspective transformation. Some results of experiments using real objects are shown.View full abstractDownload PDF (2280K) -
Sadao FUJIMURA, Atsuki MATSUMURA1986 Volume 22 Issue 9 Pages 976-981
Published: September 30, 1986
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThe thin film is widely used in many fields of industries. The accurate measurement of film thickness is quite necessary for controlling the properties of the film, and for saving energy and resources.
This paper describes a method for measuring simultaneously the film thickness and the refractive index of the material the film is made of by using the spectral measurement of the transmission of the film.
The transmission of a thin film is determined by the interference of the multibeam light waves produced by multiple reflection of the light at the boundary between the film and the air. When used nomal incident light, the film thickness d and the refractive index n determine the period of the transmission in wave number, and the refractive index gives the contrast (the ratio of the minimum to the maximum) of the spectral transmission.
The measured spectral transmission, however, does not agree well with that of an ideal film. A new formulae for the spectral transmission T=A+Bxcos(2D/λ) (where A=2R2-2R+1, D=2πnd and λ is the wavelength of the light) is derived by taking the inhomogeneity of the film thickness into consideration. The regression analysis using about one period of the spectral data gives us the parameter values of A, B and D, and from them the film thickness (d) and the refractive index (n) are determined.
The validity of this new method was confirmed by numerical simulation and by experiments also. The precision of the measurement was a few percents.View full abstractDownload PDF (709K) -
Yasuo NAKAGAWA, Yoshitada OSHIDA, Takanori NINOMIYA, Hideaki SASAKI1986 Volume 22 Issue 9 Pages 982-987
Published: September 30, 1986
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper describes a shape detection method for solder joints of flat-packaged devices loaded on a circuit board. Flat-packaged devices are widely used for high density loaded boards for application in large computers, etc. For automating the visual inspection of the solder joints, a laser-spot scanning light-section method has been newly developed.
In this method, a laser spotlight is projected from a vertical direction, and is detected from an oblique direction where a view-field is limited to a vertical line which includes the projected spotlight. The spotlight and the view-field are codeflected by a Galvano mirror which is located at a co-focuspoint of two Fourier transform lenses.
This method enables a stable shape detection without a detection error by mirror-like reflection at solder surfaces which are often caused by a conventional slit-projecting light-section method.
The structure of a detecting head and some examples of detected shapes are also described. Detected shape clearly indicates the deference between good and bad solder joints. Detection time is about 0.2s/35mm stroke.View full abstractDownload PDF (1815K) -
Yoshio KANO, Shinya HASEBE, Toshiki SHIMAMURA, Shu FUNADA, Barney K. H ...1986 Volume 22 Issue 9 Pages 988-993
Published: September 30, 1986
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIt is very important to measure water potential for analyzing its biological physiology and growth mechanism. There are some reports about measurement of it, but some of them are destructive method. One method of them is the pressure-chamber method that needs some sample and is difficult to measure automatically.
Then, as the dielectric constant of water is very large comparing other organism in a stem, it is tried to measure the water potential by capacitance method which detects capacitance in proportion to the amount of water in the stem.
Two capacitance methods are developed, and one of them reads water potential directly using ceramics and the other detects the amount of water in a stem by two electrodes surrounding a stem.
The result of measurement is followed that the response of water potential in a stem decreases with vibration to sudden lighting like step-function. The phenomena is simulated by electric circuit by the analogy of water flow to current and water potential to voltage.View full abstractDownload PDF (728K) -
Hisashi TSURUOKA, Masamoto TAKATSUJI1986 Volume 22 Issue 9 Pages 994-999
Published: September 30, 1986
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSA complex refractive index, or attenuation constant imaging apparatus is proposed for the measurement of quality of fruit. Plant electric impedance is measured in water by 2.6GHz microwave irradiation. This process resolves the problem of low resolution, multipass propagation and refraction or diffraction at air-fruit interface.
An iterative method for imaging is considered. This principle is as follows. A multilayer shell model is constructed from target object. Each shell is given an initial attenuation constant. The measured transmission wave intensity of the object is then compared with the calculated intensity from the model. Attenuation constant of the model is corrected in proportion to the difference between the two. By repeating this process, the calculated attenuation constant is expected to approach the true value.
By using the attenuation constant profile of fruit, the sugar content of fruit can be infered. 19 melons were selected as an experimental sample. After measurement, each melon is cut into round slices and sugar content is measured by a sugar meter ordinarilly used in the field. As a result, the absolute error of sugar content measurement by the iterative method was found to be ±1.5BX.View full abstractDownload PDF (1809K) -
Kazunobu YOSHIDA, Yukio NISHIMURA, Yoh YONEZAWA1986 Volume 22 Issue 9 Pages 1000-1001
Published: September 30, 1986
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSDownload PDF (193K) -
Sigeru HATANAKA, Susumu HATTORI1986 Volume 22 Issue 9 Pages 1002-1004
Published: September 30, 1986
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSDownload PDF (396K)
- |<
- <
- 1
- >
- >|