JP3040912B2 - How to expand the fault tree - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for automatically creating data of a fault tree used for system analysis in consideration of a plurality of devices and a common system. This technique is useful when a plurality of devices having exactly the same specifications are installed, such as a nuclear power plant, so that the fault tree has the same shape.

[0002]

2. Description of the Related Art In a probabilistic safety assessment (PSA) of a nuclear power plant or the like, a fault tree analysis is used as an analysis method of system reliability. A fault tree is used to identify specific faults or accidents in the system, called top events,
The relation between a basic failure or an accident, which is considered to be statistically independent from each other, called a basic event, is graphically represented by an OR condition and an AND condition. by this,
A combination of abnormalities and failures that cause an undesired failure of the entire system, that is, a scenario and its occurrence frequency are analyzed, and a system analysis model is used to estimate a result caused by the scenario. A safety evaluation that uses such an approach is called stochastic safety evaluation.

[0003]

However, creating this fault tree for a huge system is difficult.
Very labor intensive. In particular, in a nuclear power plant or the like, in order to ensure safety, a plurality of devices having the same specifications are installed, the systems are multiplexed, and the fault tree has the same shape. However, when analyzing the fault tree, it is necessary to distinguish between devices with the same specifications.
Therefore, in the prior art, many similar fault tree data are separately created. Therefore, correcting and maintaining the fault tree requires correcting a plurality of data, which requires a great deal of effort.

[0004] Further, a common support system such as a common cooling facility and a power supply is generally connected to a plurality of devices having the same specifications. However, in order to handle them, it is necessary to specially handle the data of the fault tree as common, but there is a problem that the handling is difficult.

An object of the present invention is to develop a fault tree by taking into account a plurality of devices of the same type and a common support system in order to expand the fault tree data. It is to provide a method that can be created accurately and easily.

[0006]

SUMMARY OF THE INVENTION The present invention is a method for developing a fault tree for a multiplexed system such as a nuclear power plant. That is, in a fault tree analysis of a multiplexed system in which a plurality of subsystems forming a fault tree of the same shape exist and a common support system is connected to those subsystems, a fault of the subsystem is detected on a computer. A tree is constructed using the same data, and an identification code is assigned to each subsystem for each subsystem, and data is converted so as to indicate commonality for a common support system. How to expand.

Specifically, for example, a fault tree is expanded as follows. On the computer, the fault tree of the system is described with the same data assuming that only one subsystem exists for the same subsystem, and the top event number of the subtree is first added to the input event of the event-related data. At the time of input, an identification code for identifying each subsystem is given. For a common support system, a special code indicating a common event is given when an event number is first input to an input event of event-related data. Then, by the code conversion processing of the computer, all event numbers used in the subtree are assigned identification codes corresponding to the respective subsystems, and common events are subjected to data conversion according to the special codes, and the faults of the multiplex system are converted. Expand the tree. FIG. 1 shows a simple schematic flowchart.

[0008]

On a computer, data is common in the same subsystem, and a fault tree of the same kind of device is developed only by additional data called a code. For this reason, it is not necessary to create a plurality of pieces of the same data for subsystems having the same fault tree. For a common support system, a fault tree is developed by freely defining a special code representing commonality and performing data conversion processing. This makes it possible to automate the expansion of the fault tree in a multiplex system in which a plurality of subsystems forming a fault tree of the same shape exist and a common support system is connected to these subsystems.

[0009]

FIG. 2 is a system diagram showing an example of a simple multiplexing system. This is a system for injecting water into a nuclear reactor at a nuclear power plant. At nuclear power plants,
In order to improve the reliability of the system, equipment is multiplexed, and when expanding the fault tree, many fault trees with exactly the same shape are generated, and it is necessary to consider events that are common factors. is there. The system shown in FIG. 2 is for injecting water from the common tank 12 into the nuclear reactor 10, and is performed through the A pump 14 and the A valve 15 in the A system, and through the B pump 16 and the B valve 17 in the B system. The pumps 14 and 16 are connected to the common power supply 1
8 is driven. As described above, although the A system and the B system have almost the same shape, the common tank 12 and the common power supply 18 have the same shape.
Is common.

(1) Development of Fault Tree of Same Shape FIG. 3 shows a schematic fault tree developed for the system shown in FIG. In the fault tree of FIG. 3, the shape of the fault tree of system B is the same as that of system A. Therefore, exactly the same data may be used as the data of the fault tree. The fault tree below the event number E2 is a subtree. Therefore, in the method of the present invention, data is input as shown in FIG. 4 in order to expand the fault tree.

That is, since the fault trees having the event number E2 or less have exactly the same shape, the same subsystem is described on the computer with the same data as if only one subsystem exists, and the event relation When the event number E2 at the top of the subtree is first input as a data input event, an identification code for identifying each subsystem is given. Then, by the code conversion process of the computer, the identification numbers corresponding to the respective subsystems are assigned to all the event numbers used in the subtree.

For example, in FIG. 4, A and B are used as the identification codes, and E is used as the highest event number in the subtree.
A2 and A2 are added to 2 to obtain E2A and E2B. When this input data is expanded, the fault trees below the event number E2 are expanded into two in a signed form. For example, SETS, a well-known fault tree analysis code
Expanding to the format of FIG. Event number E2
The following events are denoted by A and B, and the same code is propagated to events E3, E4,... Located below the events, and a fault tree having a similar shape is created. As described above, the fault tree data can be shared by the same kind of devices, and the analysis data can be expanded so that a plurality of devices can be considered.

(2) Processing of Common Events As shown in the examples of FIGS. 2 and 3, the event numbers used for the subtrees are numbered. However, there are usually common events, such as the common tank and common power supply shown in FIG. When such a common event exists, the expansion result shown in FIG. 5 is insufficient.

Therefore, in the method of the present invention, these developments are performed by freely defining special codes for the event-related data. That is, a fault code of the system is given on the computer, and a special code indicating the common event is given to the input event of the event-related data for the first time when the event number is input for the common support system. Then, the common event is subjected to data conversion according to the special code, and the fault tree of the multiplex system is expanded.

The definition of the code is performed, for example, as follows. *: Indicates a common event for all subtrees. ＄: common to AC, common to BD #: common to AB, common to CD When such a special code exists in the input event of the event-related data, the original code is automatically converted to another code. For example, if the special code is #, A → A, B →
When defined as A, C → B, D → B, AB common, C
D common is realized. The definition of the code conversion can be arbitrarily defined as long as the character can be used in a computer.

Here, as an example, FIG. 6 shows input data in consideration of common parts in the fault tree of the system of FIG. Here, * indicates commonality, and code conversion is as follows:
It is defined that any code is converted to a blank. As a result, the fault tree data expanded in the SETS format is as shown in FIG. Thus, the common event E
6, E7 and E9 are also standardized in the structure of the fault tree, and the state of the system can be accurately represented in the fault tree. Moreover, the method is very simple, since it is only necessary to input the defined special code at the position of the common event.

(Flowchart of the Method of the Present Invention) FIGS. 8 to 10 express the method of the present invention as detailed flowcharts. 8 is an explanation of variables used in the flowchart, FIG. 9 is an explanation of a main flowchart, and FIG. 10 is an explanation of a flowchart of a function C called from the main. FIG. 9 and FIG. 10 also show the functional description in each step.

Here, a slightly more complicated fault tree is used for the explanation of the flowchart. FIG. 11 shows the input data. In FIG. 11, the input data actually used in the flowchart are only the event number of the fault tree, the input event and the input code, the number of events, and the number of input events. FIG. 12 shows this input data as a fault tree and the input variable data required in the flowchart. In this case, the code conversion is performed as shown in FIG.
Defined as 3. The fault tree expanded in this manner is expanded into the shape shown in FIG. Actually, according to this flowchart, Fortran (FORTRA)
As a result of creating and executing a program according to N), FIG.
The fault tree having the shape of No. 4 was successfully developed.

[0019]

As described above, according to the present invention, in developing fault tree data, a fault tree of a plurality of devices of the same type is expanded with the same data in consideration of a plurality of devices of the same type and a common support system. Since the support system performs conversion by adding a special code, the effort of data creation can be greatly reduced, correction and maintenance become easy, and the fault tree can be created accurately and easily. This makes it possible to automatically output, for example, input data of a fault tree analysis code “SETS” used in a nuclear power plant.

[Brief description of the drawings]

FIG. 1 is a conceptual diagram of a method for expanding a fault tree according to the present invention.

FIG. 2 is an explanatory diagram showing an example of a simple multiplexing system.

FIG. 3 is an explanatory diagram showing an example of a fault tree for the system.

FIG. 4 is a diagram showing an example of the fault tree input data.

FIG. 5 is a diagram showing an example of expanding the fault tree.

FIG. 6 is a diagram showing another example of the fault tree input data.

FIG. 7 is a diagram showing an example of expanding the fault tree.

FIG. 8 is an explanatory diagram of variables used in a development flowchart.

FIG. 9 is a diagram showing a main flowchart of the method of the present invention.

FIG. 10 is a diagram showing a flowchart of a function C called from the main.

FIG. 11 is a diagram showing another example of the fault tree input data.

FIG. 12 is a diagram showing an example of a fault tree.

FIG. 13 is a diagram showing an example of a code conversion definition.

FIG. 14 is a diagram showing an example of expanding a fault tree.

[Explanation of symbols]

 10 Reactor 12 Common tank 14 A pump 15 A valve 16 B pump 17 B valve 18 Common power supply

Continuation of the front page (58) Field surveyed (Int.Cl. ⁷ , DB name) G06F 11/16-11/26 G06F 9/44 G05B 23/00-23/02 G21C 17/00 G21D 3/04

Claims (2)

(57) [Claims] In a fault tree analysis of a multiplexing system in which a plurality of subsystems forming a fault tree having the same shape exist and a common support system is connected to the subsystems, a sub-tree is provided on a computer. A fault tree of the system is constructed using the same data, an identification code is assigned to each subsystem for each subsystem, and data is converted for the common support system to indicate the commonality, and the multiplex system is converted. A method for expanding a fault tree, comprising expanding a fault tree. 2. A multiplexing system in which a plurality of devices having the same specification are installed to form a fault tree having the same shape, and a plurality of the same subsystems exist, and a common support system is connected to these subsystems. In the fault tree analysis of the above, on the computer, the fault tree of the system is described with the same data as if only one subsystem exists for the same subsystem. When an event number is input for the first time, an identification code that identifies each subsystem is attached. For common support systems, a special code that indicates a common event when an event number is first input for an input event of event-related data Is added to each event number used in the subtree by the code conversion process of the computer. , And the common event is subjected to data conversion according to the special code.
A method for expanding a fault tree, comprising expanding a fault tree of a multiplex system.