JPH02166596A - Plant accident analyzing device - Google Patents

Abstract

PURPOSE:To exactly execute a countermeasure for an accident by converting an analog signal which is varied at the time of accident to a pattern by putting ranking in accordance with whether its variation rate is large or small, comparing this variation pattern with a variation pattern of an accident cause analysis file, and displaying its accident cause on an output device, when they are approximate. CONSTITUTION:With respect to an analog signal which is varied at the time of accident, an accident cause retrieving means 1 gives successively rankings 1, 2, 3..., for instance, from a large variation rate to a small variation rate in accordance with whether its variation rate is large or small, and converts a variation of the analog signal to a pattern. The accident cause retrieving means 1 reads in successively accident cause analysis files 2a, 2a... from a memory device 2, and compares successively a variation pattern of the analog signal of this accident cause analysis file and a variation pattern of accident data. When the accident cause analysis file 2a having the variation pattern being approximate to the variation pattern of this accident data is retrieved, the accident cause is outputted to output devices 3a, 3b, respectively, and displayed as the cause of the accident. In such a way, it is possible to cope exactly with the accident.

Description

DETAILED DESCRIPTION OF THE INVENTION [Object of the Invention] (Industrial Application Field) The present invention relates to a plant accident analysis device that automatically searches and displays the causes of accidents that occur in industrial plants such as thermal power plants.

(Prior Art) In conventional thermal power plants, unit computers are merely used as trip sequence loggers for analyzing the causes of plant accidents.

This trip sequence logger collects various contact changes in chronological order when an accident occurs, which is necessary for analyzing the cause of the accident, and cannot automatically search for the cause of the accident.

That is, in conventional thermal power plants, when an accident occurs, plant accident data collected in a trip sequence logger is manually analyzed by an operator to determine the cause of the accident.

(Problem to be solved by the invention) As mentioned above, in conventional industrial plants such as thermal power plants, plant accident analysis is performed manually by operators, which requires a lot of time and money. Moreover, it was difficult to make quick and accurate decisions.

This causes delays in responding to plant accidents and recovery, which in turn leads to a decline in plant operating rates.

The present invention has been made in consideration of the above circumstances, and its purpose is to provide a plant accident analysis device that can automatically search and display the cause of an accident when an accident occurs in a plant. .

[Structure of the invention]

(Means for Solving the Problems) When a plant accident occurs, a series of analog signals change in a unique pattern according to the cause of the accident, so by analyzing the change pattern of the analog signals, the cause of the accident can be determined. This was done with a focus on the ability to make inferences.

In other words, the present invention provides a plurality of accident cause analysis files in which analog signals in plant data that change at the time of the accident are ranked and patterned according to the magnitude of the rate of change, for each of the causes of various accidents assumed in the plant. When an accident occurs in the memory device to store the plan l-, the analog signals in the plant data that change at the time of the accident are ranked and patterned according to the magnitude of the rate of change, and this change pattern is stored in the memory device. The present invention is characterized by comprising an accident cause search means for comparing the change pattern of the accident cause analysis file read from the above and displaying at least the cause of the accident in the accident cause analysis file on an output device when an approximation is obtained.

(Function) When an accident occurs in a plant, the required analog signal changes depending on the cause of the accident.

Therefore, the accident cause retrieval means reads this analog signal as accident data, and ranks the analog signal according to the magnitude of the rate of change, for example, from large to small, to form a pattern.

The accident cause search means sequentially reads the accident cause analysis files from the memory device, sequentially compares the change pattern of the analog signal in the accident cause analysis file with the change pattern of the accident data, and When an accident cause analysis file having a change pattern similar to the change pattern is retrieved, at least the cause of the accident in the file is output to the output device and displayed.

Therefore, according to the present invention, when an accident occurs in a plant, the cause of the accident is displayed at least by the output device, and the cause of the accident is displayed quickly and accurately.

For this reason, it is possible to quickly respond to and recover from plant accidents, and it is possible to improve the operating rate of the plant.

(Example) Hereinafter, an example of the present invention will be described based on FIGS. 1 and 2.

FIG. 1 is an overall configuration diagram showing the overall configuration of an embodiment of the present invention. In the figure, an accident cause search means 1 is composed of a computer (not shown) that receives plant data P at the time of an accident in a plant, and a memory device 2; It has a CRT and a printer output device 3a, 3b, respectively.

The accident cause search means 1 receives various analog signals (not shown) that change at the time of an accident in a plant from the plant as accident data, and further sorts all of these analog signals according to the magnitude of the rate of change, for example, from large to small rate of change. Rankings 1, 2, 3, etc. are attached to each pattern to create a change pattern of the accident data, and this change pattern is sequentially compared with the change pattern of the accident cause analysis file read by the memory device. When a change pattern in the accident cause analysis file that is similar to the change pattern in the accident data is retrieved, at least the cause of the accident in the accident cause analysis file is output to the output devices 3a and 3b for display.

That is, the memory device 2 is an accident cause analysis file 2a which shows, as shown in FIG.

2a... are configured and stored respectively.

Each accident cause analysis file 2a is, for example, FT, P.

A large number of analog signals are created for each accident cause, such as "0 abnormality", and a series of analog signals that change during an accident that occurs based on the cause of the accident are ranked and patterned according to the magnitude of the rate of change.

Next, the operation of this embodiment will be explained.

When an accident occurs in a plant (not shown), a required analog signal of the blunt data output from the plant changes.

Therefore, the accident cause search means 1 reads the analog signals that change at the time of the accident as accident data, and ranks these analog signals according to the magnitude of the rate of change, for example, from the most serious to the least, 1, 2, 3, . . . are applied sequentially to pattern the changes in the analog signal.

Further, the accident cause search means 1 sequentially reads accident cause analysis files 2a, 2a, etc. from the memory device 2, and sequentially compares the change pattern of the analog signal of this accident cause analysis file with the change pattern of the accident data. Then, when an accident cause analysis file 2a having a change pattern approximating the change pattern of this accident data is retrieved, at least the cause of the accident is output from the file.

3b, and display them as the cause of the currently occurring accident.

Therefore, according to this embodiment, when an accident occurs in the plant, the cause of the accident is automatically and quickly and accurately displayed on the output devices 3a and 3b, respectively.

For this reason, plant accident response and recovery can be carried out quickly and accurately, and the availability rate and safety of plant operation can be improved.

〔Effect of the invention〕

As explained above, according to the present invention, when an accident occurs in a plant, at least the cause of the accident can be displayed quickly and accurately on the output device, so that countermeasures and recovery from the plant accident can be quickly and accurately carried out. It is possible to improve the availability and safety of plant operation.

[Brief explanation of the drawing]

FIG. 1 is an overall configuration diagram showing the overall configuration of an embodiment of the present invention, and FIG. 2 is a diagram showing the configuration of an accident cause analysis file stored in the memory device shown in FIG. 1. 1... Accident cause search means, 2... Memory device, 3a
...CRT output device, 3b... printer output device. Figure 1

Claims (1)

[Claims] A memory device that stores a plurality of accident cause analysis files in which analog signals in plant data that change at the time of each accident are ranked and patterned according to the magnitude of the rate of change, for each of the causes of various accidents assumed in the plant. , when an accident occurs in the plant, the analog signals in the plant data that change at the time of the accident are ranked and patterned according to the magnitude of the rate of change, and this change pattern is read from the memory device to analyze the cause of the accident. 1. A plant accident analysis device comprising an accident cause search means for displaying at least the cause of the accident in the accident cause analysis file on an output device when the change pattern of the file is compared and approximated.