Yoshiomi Munesawa

HAZOP analysis system compliant with equipment models based on SDG

It is important to assess the risk in chemical plants. HAZOP is widely used in the risk assessment to identify hazard. An automatic analysis system is developed to perform HAZOP effectively. In this study, semi-automatic analysis system was developed by using the Signed Directed Graph (SDG) as a deviation in the behavior of the propagation of equipment. Versatility of analysis is raised based on the propagation of deviation by adding the device in accordance with the rules. Our developed HAZOP analysis system is applied to one chemical process. And the future works for this study are explained.

An approach to the automation of the assembly of flexible belt-shaped objects

A robot system was developed for the assembly of flexible belt-shaped objects, and the assembly procedures employed by workers in using it were analysed. A multi-fingered hand was developed to hold the objects at several points, and to move them into the proper position for attachment to a part. In addition, an image processing method that can recognize the gripping positions on source objects and conditions after the source objects are attached is proposed.

Development of Fault Diagnosis System Using Principal Component Analysis for Intelligent Operation Support System

In this paper, it is proposed to develop the fault diagnosis system using the principal component analysis (PCA) for the intelligent operation support system that calculates the effect of fault propagation in abnormalities situation and gives appropriate information to plant operators. This proposed system using PCA discriminates a failure of equipment based on process variables. The proposed method deals with process variables in steady condition and only one type warning alarm condition that is occurred by several different failures. A set of process variables on each failure is shown as the points on 2dimensional data space by PCA. This system judges as a failure of the equipment when a set of current process variables is closed to the point of a failure of equipment on the data space. The proposed fault diagnosis system is applied to process on a simulator and is confirm its validity. Chemical plants had become very complex for a lot of instrumentation and control systems. It is difficult for operators to predict the effect of fault and to decide corrective actions. In addition, quick response from operator is demanded as any delay response for abnormalities may expand the damage of chemical plant.� Many sensors are required to monitor the chemical process for safety. It is difficult to detect the fault in the process from too many sensors. Therefore many sensors’ data are reduced to less parameter by a multivariate analysis (Kano et al. 2001, 2004, Wise and Gallagher 1996). These methods detect the fault in the process, but not diagnosis the failure of equipment. Our research is developing “Intelligent Operation Support System”. This system calculates the effect of fault propagation in abnormal situations and gives appropriate information to operators. It will help operators to make quick judgments for safety. This system predicts process variables in the abnormalities situation’s plant using a simulator. The states of all equipment are inputted to a simulator in order to calculate process variables correctly. In this paper, it is proposed to develop the fault diagnosis system using the principal component analysis. This proposed system discriminates a failure of equipment in its early stage. This system is important for the intelligent operation support system. The principal component analysis discriminates a failure of equipment based on process variables from the actual plant. A data set is process variables in steady conditions and in abnormalities condition. These data sets are obtained from a simulator when the warning alarm is turned on after the selected equipment is artificially changed to a failure mode. The proposed method deals with process variables in steady condition and only one type warning alarm condition at a time. One type warning alarm is occurred by several different failures, so process variables in one failure are similar to another failure in one type warning alarm. So it is difficult to distinguish one failure among failures in one type warning alarm when the principal component analysis deals with failures in all warning alarm at a time. And the proposed method deal with selected process variables based on value changes after a failure is occurred. A set of process variables on each failure is shown as the points on 2dimensional data space by the principal component analysis. This system judges as a failure of the

An Inspection Method for Three-Dimensional Surface and Volume of Minute Subject by Image Processing

In this paper, we propose an inspection system of subject with three dimensional surface and volume by image processing using the distribution of brightness and section curves. The distribution of brightness on image depends on the inclination of subjects surface from an optical axis of the camera unit. Therefore, it is possible to inspect the three dimensional surface based on it. The sample is classified into some patterns by the visual inspection of worker. Next, the brightness pattern of a sample is measured, and its standard index is calculated. The weighting coefficient is determined based on the standard deviation of width of bright area on the sample. The volume of inspection subject is measured using a line laser device, and a control chart method is applied to maintain the quality of shape and volume of subjects through the production line. The proposed method is applied to soldering on electric device, and weight of lead in soldering is estimated from the measured volume.

Recognition Method by Image Processing Using 3D CAD Figure Information.

We propose a method for recognizing a subject from the image taken by a CCD camera and comparing with the CAD figure. The three-dimensional CAD figure is transformed into a two-dimensional figure in which some color is added inside the contour. The transformed figure is used as the input of image processing. The two-dimensional CAD figure and an image of the recognized subject from one camera are represented by characteristics, that is, the center of gravity, contour, the distance between them, and FFT. The figure of the subject is estimated from these characteristics by comparing with the database of CAD figure.