JPS6155777A - Equipment diagnosis support system - Google Patents

Abstract

PURPOSE:To carry out an abnormal diagnosis of a device simply and efficiently by successively deducing and processing causes produced by an abnormality generating phenomenon according to rules stored based on an initially input information in accordance with the phenomenon. CONSTITUTION:A deductive processing division 3 urges an operator to carry out an input of an abnormal condition of a device through a work station 4 in examining the abnormal causes of the device. When the operator carries out the input of the condition of the device through the work station 4, the deductive processing division 3 internally represents the input information as a rule type COND. The deductive processing division 3 searches whether the same diagnostic rule as the obtained internal representing information exists in a rule data base or not and reads a serial diagnostic rule. When a COND section in which a rule type is defined exists, in order to obtain an information of Z described in the rule, an information for urging to enter the information is given to the operator through the work station 4. When these informations enter, a response to an initial input is given.

Description

DETAILED DESCRIPTION OF THE INVENTION [Technical Field of the Invention] The present invention relates to an equipment diagnosis support device that sequentially presents work procedures for abnormality diagnosis of various equipment in an atomic couplant, etc., and can improve the efficiency of the diagnostic work.

[Technical background of the invention and its problems]

Diagnosis work such as maintenance, repair, and inspection of various types of lines in nuclear couplants and the like is performed by checking a huge number of check items according to predetermined work procedures.

However, performing these checks requires numerous measurements and evaluation of the measurement results, and diagnosing all the equipment that makes up the plant requires a huge amount of work and time. .

Specifically, nuclear power plants undergo periodic inspections once a year, approximately every three months.

By the way, if an abnormality is found in the equipment during such periodic inspections, it is necessary to investigate the cause and take appropriate measures. For this purpose, it is generally necessary to conduct various investigations by experts who are familiar with the equipment.

However, this type of expert is not always available when an abnormality is discovered in a device, and even if an investigation is started by another expert, other abnormal phenomena may be discovered as the investigation progresses. Other experts are often needed. For this reason, it has been extremely difficult to appropriately and efficiently diagnose abnormal equipment when the abnormality is discovered.

[Object of the Invention] The present invention has been made in consideration of the above circumstances, and its purpose is to detect abnormalities in sieving utensils without having special knowledge regarding various types of ware. An object of the present invention is to provide a device diagnosis support device that can easily and efficiently perform abnormality diagnosis of the device.

[Summary of the invention]

The present invention stores, for example, a diagnostic inference process for diagnosing various types of equipment constituting a plant and its diagnostic work procedures as diagnostic rules in a rule database, and the diagnostic rules stored in this rule database. The configuration allows editing such as registration, addition, modification, deletion, etc. by a rule editing processing unit, and refers to the diagnostic rules stored in the rule database according to the status information of the device inputted via the work terminal, for example. An inference processing section is provided that sequentially infers the cause of the abnormality and unusual work procedures for investigating the cause, and the work procedures etc. inferred by this inference processing section are sequentially presented via the work terminal. It is designed to support abnormality diagnosis work.

[Embodiments of the invention]

Hereinafter, one embodiment of the present invention will be described with reference to the drawings.

The figure is a schematic configuration diagram of the embodiment apparatus, in which 1 is a rule database, 2 is a rule editing processing section, 3 is an inference processing section, and 4 is a work terminal consisting of, for example, a display and a keyboard.

The rule a result processing unit 2 creates diagnostic rules, such as information and work procedures necessary for diagnosing predetermined equipment (for example, various equipment constituting a plant), according to expert knowledge input from the work terminal 4, etc. This is then newly registered in the rule database 1, or additionally registered in the diagnostic rules already stored in the rule database 1.

The rule editing processing unit 2 also performs correction of diagnostic rules stored in the rule database 1 and deletion of diagnostic rules that are no longer needed. In other words, the rule editing processing section 2 edits and manages the diagnostic rules stored in the rule database 1. Then, such rule editing processing 8I12
Under the management of the rule database 1, various diagnostic rules for the device to be diagnosed are stored.

The diagnostic rules stored in this rule database 1 are:
For example, it is expressed in the following two types - ■ IF C0ND [Xl, Yl] & GON
O[X2. Y2] rcOND [Xn, Yn] THEN 0OND [X, Yl ■ FORC0ND (X, Y) Do [2(X, Y)] Here, the above X is a sentence expressing a qualitative phenomenon of the device to be diagnosed, or It consists of a variable representing a numerical value, and Y indicates either [Yes/No] determined by the content of the above-mentioned X or [True/False] representing the relationship between the numerical values. In addition, for GONO (X, Y) consisting of 2, the above-mentioned
Or, it represents the 1r work procedure to be executed regarding Y.

The inference processing unit 3 infers the cause of the abnormality of the device according to the rule indicated by the above-mentioned (■) among such diagnostic rules. Further, the inference processing unit 3 uses the rule (2) above to prompt the work terminal 4 to input status information of the equipment to be diagnosed. In accordance with the information input from the work terminal 4 according to the rule (2), the inference processing unit 3 checks whether each C0ND [X, Y] constituting the rule shown in (2) holds true. In other words, the work procedure is presented to the worker according to the rule shown in ■ above in order to check whether the rule shown in ■ above is satisfied.
The system waits for input of status information of the equipment to be diagnosed obtained through the diagnostic work.

Therefore, the inference processing unit 3 follows the above-mentioned diagnostic rule,
In accordance with the abnormality occurrence information of the equipment, status information of the parts to be checked on the equipment is presented to the worker via the work terminal 4 as a work procedure, and the status information of the parts to be checked is inputted via the work terminal 4. have When that information is input, the diagnosis rule (2) above is confirmed based on that information, and the next location to be diagnosed is processed using an IF logic.

By sequentially repeating this procedure, the diagnostic locations for the device to be diagnosed, the diagnostic methods, etc. are presented one after another via the work terminal 4 as diagnostic work procedures.

Thus, when such a device is used, when diagnosing a device in which an abnormality has occurred, work procedure information consisting of the diagnosis location and information necessary for diagnosis, etc., is presented one after another as the diagnostic work progresses. Therefore, even if the operator does not have highly specialized knowledge, the diagnosis can be performed efficiently. Therefore, when an abnormality occurs in a device, it is possible to promptly, simply, and efficiently diagnose the abnormality when an abnormality occurs in the equipment, even if the expert is not waiting.

Moreover, since various expert knowledge can be accumulated as diagnostic rules, effects such as being able to effectively deal with various causes of abnormalities can be achieved. Furthermore, by adopting the above-mentioned diagnostic rule structure, it becomes possible to describe quantitative phenomena in the same way as qualitative phenomena, which increases the efficiency of inference processing for pursuing the causes of abnormalities. This makes it possible to proceed with the diagnosis work of the equipment to be diagnosed in a short time and efficiently, and its practicality is high.

Next, an example of the work flow of the inference processing unit 4 using the above-mentioned diagnostic rules will be explained.

(I) First, when searching for the cause of an abnormality in the shaker, the inference processing unit 3 prompts the worker via the work terminal 4 to input the abnormal phenomenon state of the device, which will be a clue.

(II) In response to this, when the worker inputs information about the status of the equipment via the work terminal 4, the inference processing unit 3 converts the input information into the rule format indicated by ■ above, C0ND (
Internally expressed as (X, Y). For example, when status information such as "the voltage between terminals a and b is low" is input, internal expression information such as C0ND[Va-b is low; Yes] is obtained. Here, in the rule type (2) mentioned above, X and Y are X; (2) a-b is low Y: Yes.

(■) After that, the inference processing unit 3 searches whether the same diagnostic rule as the internal representation information obtained as described above exists in the rule database 1, and determines the CoND.
A set of diagnostic rules containing clauses is read from database 1. And ° the other CoND clauses included in that rule are
Check whether each is established.

(■) However, in this process, if there is a CoND clause in which the rule type described in (■) is defined, information that prompts you to input the information in order to obtain the information in 2 described in the corresponding rule. is presented to the worker via the work terminal 4 as the next work procedure. For example, C0ND [Va-b is low; YesS] and C0ND [Va-b is low; YesS]
(Resistance C is defective: Y13S) forcOND [Resistance C]
Resistance value: 1 is Ω or smaller Do [Measure the resistance value of resistor C] If a diagnostic rule is written, follow the second CoND clause of this rule and write ``Measure the resistance value of resistor C.'' Please” message is output as the diagnostic work procedure at that point. According to this message, the resistance value of the resistor C of the equipment to be diagnosed is measured, and the constant resistance value thereof is inputted via the work terminal 4.

(v) After that, this input information is processed as the variable 2, and it is checked whether the corresponding CoND clause is established. In this checking process, if the condition is not satisfied, the inference processing unit 3 stops checking the rule. Then, other rules containing the CoND clause present in the initial input are searched and similar processing is executed.

In addition, if all CoND clauses in a rule are established, all CoND clauses described in the THEN clause of that rule
When the CoND clause is established, other rules including that CoND clause are searched and the same process is performed.

(Vl) As described above, the inference processing unit 3 checks whether CONDw included in the rule holds,
When the CoND clause included in the THEN clause is established and a rule indicating the end of inference is written for the CoND clause, the worker uses the information of X and Y of the CoND clause as an answer to the initial input. The above-mentioned process is completed.

(■) Note that if there is no rule in the rule database 1 that includes an internal representation for the initial input as a CoND clause, or if a rule indicating the end of inference is not defined, the answer to the initial input is unknown. Outputs the message as a message and ends the process.

(■) Conversely, if multiple answers are found for the initial input, all of them are presented. In addition, if a final answer to the initial input is obtained, along with that answer,
The work procedure that led to the answer (diagnosis rules and CoN)
It would be convenient if the sections D) were presented and output at the same time.

〔Effect of the invention〕

In this way, according to this device, based on the information that is initially input in response to an abnormal phenomenon occurring in a device to be diagnosed, the cause of the phenomenon is sequentially inferred according to the rules stored in the rule database 1, and the diagnosis is performed. While sequentially presenting the work procedure, we will prompt you to input the status information of each part of the equipment mentioned above, and based on this information, we will pursue the cause of the abnormality.
Even if the worker does not have special specialized knowledge,
The operator can be appropriately supported to effectively diagnose the equipment to be diagnosed. Moreover, since the work procedure is accurately instructed, great practical effects such as the ability to easily improve diagnostic work efficiency can be achieved.

Note that the present invention is not limited to the embodiments described above. For example, if countermeasures against the investigated cause of an abnormality are incorporated as part of the rules, the countermeasures can be taken effectively. Although the work terminal 4 is shown here as having a display and a keyboard, it is also possible to use a graphic display, a data tablet, or even a voice input/output device. Furthermore, various sensors may be used in combination to automatically input the status information of the equipment to be diagnosed. In short, the present invention can be implemented with various modifications without departing from the gist thereof.

[Brief explanation of the drawing]

The figure is a schematic configuration diagram of an apparatus according to an embodiment of the present invention. 1... Rule database, 2... Rule IPA collection processing section, 3... Inference processing section, 4... Work terminal.

Claims (4)

[Claims] (1) A rule database that stores the diagnostic reasoning process and its work procedure for diagnosing the device to be diagnosed as diagnostic rules, a rule editing processing unit that edits the diagnostic rules stored in this rule database, and an inference processing unit that references the diagnostic rules stored in the rule database according to information input via the terminal, infers a work procedure necessary for diagnosing the device to be diagnosed, and presents it via the work terminal; An equipment diagnosis support device characterized by comprising: (2) The equipment diagnosis support device according to claim 1, wherein the equipment to be diagnosed is comprised of equipment that constitutes a plant. (3) The device diagnosis support device according to claim 1, wherein the diagnostic rule editing process consists of registering and adding new rules, and modifying and deleting existing rules. (4) The inference processing unit stores the cause of the abnormality and the diagnostic work procedure for the device to be diagnosed to search for the cause in a rule database based on the status information of the device to be diagnosed that is input for the abnormality of the device to be diagnosed. 2. The device diagnosis support device according to claim 1, wherein the device makes inferences sequentially according to diagnostic rules stored in the device and presents the next diagnostic work procedure for the device to be diagnosed.