JPH05236641A - Malfunction diagnosing apparatus for plant - Google Patents

Abstract

PURPOSE:To sense a malfunction in an operation manipulating state including starting and stopping operation of a plant by dynamically setting a monitoring reference of a process amount in response to a variation in the manipulating state of the plant. CONSTITUTION:A monitoring reference setter 10 retrieves a process amount to be affected by an operation from a monitoring reference setting knowledge base 11 based on an operation input by an operation input unit 9. Then, the setter 10 retrieves a malfunction criterion at the time of manipulating the process amount from the base 11. As a result of the retrieval, if the process amount to be affected by the influence of the operation is already an object to be monitored, the criterion is replaced with that of the process amount for constituting a monitoring reference immediately before it. If the process amount to be affected by the influence of the operation is not the object to be monitored, the process amount and the criterion are added to the reference immediately before it. A malfunction retrieving processor 4 detects a malfunction based on the dynamically set reference.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to detection of abnormal events occurring in large-scale plants such as power plants.

[0002]

2. Description of the Related Art FIG. 6 is a block diagram showing a conventional plant abnormality diagnosis apparatus. In the figure, 1 is an electronic computer, a process input unit 3 for reading a process amount from a power plant 2, an abnormality detecting unit 4 for detecting an abnormality from the process amount, and an abnormal event detected by the abnormality detecting unit 4, An abnormality cause inference unit 5 that infers an abnormal place and a cause from the data input in the process input unit 3, and knowledge and a cause of failure that qualitatively and quantitatively represent the characteristics at normal time created for each component device of the plant. And a knowledge base 6 composed of knowledge of causal relationships between events, and a display processing unit 7 for performing processing for outputting a generated event, failure location, failure cause, countermeasure, etc. to a CRT display device 8. It

Next, the operation will be described. The process input unit 3 inputs data from the power generation plant 2 and converts it into a data format used by the abnormality cause inference unit 5. The abnormality detection processing unit 4 monitors the amount of processes input and processed by the process input processing unit 3 and detects that an abnormality has occurred when the process amount deviates from a preset normal range. The anomaly cause inference unit 5 operates when an anomaly is detected by the anomaly detection processing unit 4, identifies a failure location / cause based on the detected anomalous event and process amount, and the inference result is displayed on the CRT by the display processing unit 7. It is displayed on the device 8.
The reason for the abnormality is inferred by the knowledge accumulated in the knowledge base 6 for each constituent device of the plant, which is qualitatively and quantitatively expressed as the characteristic in the normal state, and the knowledge of the causal relation between the failure cause and the event.

[0004]

In the conventional plant abnormality detection processing, since the preset abnormality determination point is used for abnormality determination regardless of the plant operation state, the plant operation state There is a problem that the standard for the abnormality determination according to the change cannot be set, the abnormality is erroneously detected and the abnormality is not detected, and as a result, the abnormality cause inference unit 5 malfunctions. In the current plant operation, in the fluctuating operating state such as starting and stopping, the above-mentioned problems are clear and important because the alarm signal for detecting the abnormality is originally used as a status monitor. Problem.

The present invention has been made in order to solve the above problems, and an object thereof is to obtain an abnormality diagnosing apparatus for detecting an abnormality occurring in a plant and identifying the cause of the abnormality.

[0006]

A plant abnormality diagnosing apparatus according to the present invention includes a knowledge base relating to a plant operation and a process amount affected by each operation, and an operation information and a process amount abnormality at each operation time point. Based on the knowledge base that describes the knowledge related to the judgment, set the abnormality judgment criteria for the process amount and the process amount to be monitored according to the changes in the operating condition of the plant, and dynamically set the monitoring criteria for abnormality detection. A means for setting is provided.

[0007]

The plant abnormality diagnosing device according to the present invention utilizes the input information relating to the plant operation and the knowledge base relating to the process amount to be monitored and the abnormality judgment criterion to determine the monitoring criterion according to the operating condition of the plant. The plant is dynamically set, and the plant process amount is monitored based on the dynamically set monitoring standard to detect the plant abnormality, infer the cause, and output the result to the CRT display device.

[0008]

An embodiment of the present invention will be described below with reference to the drawings. In FIG. 1, reference numeral 1 is an electronic computer, which is a process input unit 3 for input-processing a process amount from a power generation plant 2 which is a target of a plant abnormality diagnosis apparatus according to the present invention.
And an operation input unit 9 that inputs and processes the operation of the power plant 2, and a monitoring reference setting unit 10 that sets a monitoring reference using the operation information read by the operation input unit 9 and the monitoring reference setting knowledge base 11. And an abnormality detection unit 4 that detects an abnormality from the process amount read by the process input unit 3 and the monitoring reference dynamically set by the monitoring reference setting unit 10.

The monitoring standard is composed of a set of a process amount that needs to be monitored at a certain point during plant operation and an abnormality determination standard for determining whether or not the process amount is abnormal. Match the number of processes that need to be monitored.

As shown in FIG. 2, the monitoring reference setting knowledge base 11 has a correspondence relation between "operation" and "process amount affected by the operation", and as shown in FIG. 3, "process amount" and "for each operation". "Abnormality judgment standard".

FIG. 2 shows an example of the correspondence between "operation" and "process amount affected by the operation". In this example, "operation a1 causes process amount p1, process amount p2 and process amount". p3 is affected. ",
It is described that “the operation amount a1, the process amount p2, and the process amount p4 are affected by performing the operation a2”, “the operation amount p3 is affected by performing the operation a3,” and the like.

FIG. 3 shows an example of the correspondence relationship between the "process amount" and the "abnormality determination criterion for each operation". In this example, the "process amount p1 is abnormal when the operation a1 is performed. If the judgment criterion sp1a1 is deviated, it is regarded as abnormal, and if the operation a2 is performed, the abnormality judgment criterion sp1a2 is judged.
When the operation a1 is performed, the abnormality determination reference sp2
It is described as "abnormal when it deviates from a1 and abnormal when it deviates from the abnormality judgment reference sp2a2 when the operation a2 is performed".

Next, the operation of the monitoring standard setting unit 10 and the abnormality detecting unit 4 in the above embodiment will be described with reference to the flowchart of FIG. 4 and FIG. 5 showing changes in the monitoring standard. The monitoring reference setting unit 10 searches the monitoring reference setting knowledge base 11 for the process amount affected by the operation based on the operation input by the operation input unit 3. Then, the abnormality criterion of the process amount when the operation is performed is searched from the monitoring criterion setting knowledge base 11. The following four changes occur in the monitoring criteria depending on the search results.

Case 1 If the process amount affected by the operation is already monitored, the abnormality determination criterion of the process amount is replaced with the abnormality determination criterion of the process amount that constitutes the immediately preceding monitoring criterion. Case 2 If the process amount affected by the operation is not the target of monitoring, the set of the process amount and the abnormality determination criterion of the process amount is added to the immediately preceding monitoring criterion. Case 3 If the process amount that was the target of monitoring immediately before is no longer required to be monitored due to the end of the operation, the set of the process amount and the abnormality determination criterion of the process amount is deleted from the immediately preceding monitoring criterion. Case 4 If the process amount to be monitored and its abnormality determination criteria match those of the immediately preceding operation, the monitoring criteria are the same as the immediately preceding monitoring criteria. As described above, the abnormality detection unit (4) detects an abnormality based on the monitoring standard that is dynamically set.

Next, assuming that the operation a1 is performed at the time t1, the operation a2 is performed at the time t2, and the operation a3 is performed at the time t3, the change of the monitoring reference will be described over time. It is assumed that the process amount is monitored by the monitoring standard w0 before the time t1.

When the operation a1 is performed at time t1, the monitoring reference setting unit 10 causes the process amount p1, the process amount p2,
The process amount p3 is the monitoring target, and the abnormality determination reference sp1a1 and the process amount p2 in the operation a1 of the process amount p1
Abnormality criterion sp2a1 and process amount p in the operation a1 of
The abnormality determination standard sp3a1 in the operation a1 of 3 is the monitoring standard w
0 (case 2), the monitoring standard at time t1 is w
Set to 1. In the abnormality detection unit 10, the current monitoring standard w
According to 1, the process amount is monitored and an abnormality is detected.

When the operation a2 is performed at time t2, the process amount p4 is added to the object of monitoring by the monitoring reference setting unit 10, and the abnormality determination reference s for the operation a2 of the process amount p4 is added.
p4a2 is added to the monitoring reference w1 (case 2), the process amount p3 is removed from the monitoring target, and the process amount p3
The abnormality determination standard sp3a1 in the operation a1 of is the monitoring standard w1.
From the case (case 3), a new abnormality determination criterion sp1a2 is adopted for the process amount p1 (case 1), and a new abnormality determination criterion sp for the process amount p2 is adopted.
As a result of adopting 2a2 (case 1), the monitoring reference at time t2 is set to w2. The abnormality detection unit 10 monitors the process amount according to the monitoring standard w2 and detects an abnormality.

When the operation a3 is performed at time t3, the monitoring reference setting unit 10 sets the monitoring reference to w3, and the abnormality detection unit 10 detects an abnormality according to the monitoring reference w3.

[0019]

As described above, according to the present invention, since the monitoring standard is dynamically set according to the operating condition of the plant, even in various operating conditions including operations such as starting and stopping the plant. It is possible to detect an abnormality.

Although the present invention has been described as an abnormality diagnosis device,
By using the output of the abnormality detection unit 10 as an alarm,
Application to alarm devices is also possible.

[Brief description of drawings]

FIG. 1 is a configuration diagram showing a plant abnormality diagnosis apparatus according to an embodiment of the present invention.

FIG. 2 is an explanatory diagram showing the contents of a knowledge base.

FIG. 3 is an explanatory diagram showing the contents of a knowledge base.

FIG. 4 is a flowchart showing the operation of a monitoring standard setting unit.

FIG. 5 is an explanatory diagram showing changes in monitoring standards.

FIG. 6 is a configuration diagram of a conventional plant abnormality diagnosis device.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 electronic computer 2 power generation plant 3 process input unit 4 abnormality detection processing unit 5 abnormality cause inference unit 6 inference knowledge base 7 display processing unit 8 CRT display device 9 operation input unit 10 monitoring reference setting unit 11 monitoring reference setting knowledge base

Claims (1)

[Claims] 1. A process input processing unit that reads a process amount of a plant, an abnormality detection processing unit that detects an abnormality based on a monitoring standard, an abnormality cause inference unit that identifies an abnormality cause, and knowledge about abnormality cause identification. In a plant abnormality diagnosis device composed of a knowledge base, the operation input processing unit that reads the operation state of the plant and the knowledge base for setting the monitoring standard that describes the relationship between the operation and the process amount affected by the operation. By providing a monitoring standard setting unit that dynamically sets the monitoring standard by using the operation read by the operation input processing unit and the monitoring standard setting knowledge base, the monitoring standard of the process amount can be changed in the operating condition of the plant. An abnormality diagnosis device for a plant, which can be dynamically set according to the above.