JPH0217511A - Plant monitoring device - Google Patents

Abstract

PURPOSE:To safety and correctly operate a plant by using a knowledge base and an inference processing means, executing the ranking of an abnormality at the time of discovering the plant abnormality and executing the alarm output, etc., in accordance with the rank. CONSTITUTION:A plant monitoring device provides a detecting part 11 of plant data, an input processing part 12 to collect the plant data from the detector 11 at a constant period and execute the processing such as the engineering unit conversion, an average value operation, and data preservation and plant data base 13 to record the data. Further, a knowledge base 14, an inference processing part 15 to execute the inference processing, an inference result output part 16, a displaying output device 17 and an abnormality generating hysteresis data base 18 are provided. The above-mentioned knowledge base 14 is composed of an abnormality deciding knowledge 141, a ranking knowledge 142, a cause deciding knowledge 143, an abnormality countermeasure knowledge 144 and a non-experience abnormality deciding knowledge 145. Thus, the time series change, etc., of the data read periodically are always collated with the abnormality deciding knowledge, the abnormality is judged and forecasted and the ranking of the abnormality is executed.

Description

Detailed Description of the Invention [Object of the Invention] (Industrial Application Field) The present invention relates to a plant monitoring device including an instrumentation system in a large-scale plant such as a power generation plant or a chemical plant.

(Prior Art) In a large-scale plant, a large number of facilities and equipment are organically connected and arranged in a line, and a large number of detectors are also provided for monitoring the plant. Therefore, in order to monitor the plant, it is necessary to check the detection values of these many detectors, but conventional plant monitoring equipment alerts plant operators when the plant-like Ba exceeds a set limit value. The plant operator was responsible for identifying the cause of the abnormality and taking countermeasures.

(Problem to be solved by the invention) However, in such conventional plant monitoring devices, if the tolerance range provided for each detector is too narrow, alarms will be issued frequently, and conversely, if the tolerance range is too wide, abnormalities may occur. There was a problem in that detection was delayed and it was difficult to set an acceptable range.

In addition, even if the data of a certain detector exceeds the limit value and an alarm is issued, it does not necessarily mean that it is an abnormality that requires immediate treatment. The response was not determined, and identifying the cause of the abnormality was a difficult task even for experienced plant operators, placing an excessive burden on the operators. Particularly in plants where changes occur rapidly, by the time a detector exceeds its limit and an alarm is issued, the problem has progressed considerably, and unless countermeasures are taken within a short period of time, a major accident will result. There may be cases where this could lead to a situation. If such a situation is relied on by the operator's judgment, there is a risk that the operator may make a mistake in judgment due to the emergency situation, and there is a risk that the situation will worsen due to erroneous operation.

This invention solves the conventional problems as described above, and without issuing unnecessary alarms, it can quickly and accurately determine the true cause of abnormalities that require countermeasures, and take specific countermeasures. It is an object of the present invention to provide Plan 1 to a monitoring device that can safely and accurately monitor a plant by instructing even measures to reduce the work burden on plant operators.

[Structure of the Invention] (Means for Solving the Problems) A plant monitoring device of the present invention includes a plant data input section for inputting plant status data, and past and present plant data inputted by the plant data input section. A knowledge base that stores knowledge for determining current and future plant conditions based on the system, including abnormality judgment knowledge for determining abnormalities in plant conditions, and ranks for ranking abnormalities according to their urgency and importance. The system stores the following knowledge, cause determination knowledge for investigating the true cause of an abnormality, knowledge for countermeasures against the cause of an abnormality, and inexperienced abnormality determination knowledge for identifying and recording an abnormality that appears for the first time that is not in the knowledge base. an inference processing unit that performs inference processing by selecting and combining knowledge from this knowledge base; and an output unit that automatically distinguishes and outputs measures such as warnings for abnormalities based on the inference processing and results. It is equipped with the following.

(Operation) The plant monitoring device of the present invention constantly compares the time-series changes in periodically read data and the relationship between data with abnormality determination knowledge obtained from design conditions and the experience of plant operators. , to judge or predict abnormalities.

When an abnormality is detected, it may be an abnormality that does not need to be notified to plant operators in detail, such as a temporal data abnormality, or an abnormality that is important but does not require an emergency, or an abnormality that requires immediate countermeasures. Anomalies are ranked, and based on the rank of the anomaly, causes are investigated if necessary, and countermeasures to be instructed to plant operators are determined.

Warnings and display outputs are also made to be differentiated so that the plant operator can recognize the rank of the abnormality and take measures according to each rank.

(Example) Hereinafter, embodiments of the present invention will be explained in detail with reference to the drawings.

A plant monitoring device according to an embodiment of the present invention shown in FIG.
1.11. ... and a plant database 13 that records the plant data inputted from the input processing section 12. ing.

Additionally, there is a knowledge base 14 that stores knowledge about plant status determination and equipment, and a plant database 13.
Past and present data and knowledge base recorded in 14
an inference processing unit 15 that performs inference processing by comparing the information with the knowledge of
An inference result output unit 16 that receives the processing results of the inference processing unit 15 and outputs them according to the rank of each abnormality, and a display output device 17 such as a bell, annunciator, or CRT that outputs alarms or abnormality determination results. and a database 18 of abnormality occurrence history.

The knowledge base 14 includes abnormality determination knowledge 141, ranking knowledge 142, cause determination knowledge 143, and abnormality countermeasure knowledge 14.
4. Consists of inexperienced abnormality determination knowledge 145.

The operation of the BLAN 1 monitoring device having the above configuration will be described next.

Based on the flowchart in FIG. 2, the input processing unit 12 searches each data detector 11 at regular intervals and collects plant data. The collected data undergoes engineering unit conversion, moving average calculation, etc., and is stored in the plant database 1.
3 (step 21).

The inference processing unit 15 uses the abnormality determination knowledge 141 to calculate past and current process values, average values, control setting values, control operation values, etc. stored in the plant database 13 at each data input cycle in the data input process. An abnormality is determined by comparing the information with the following (step 22).

When an anomaly is detected or predicted in this anomaly determination process (step 23), the ranking knowledge 142 is used to determine whether the anomaly can be ignored, a serious anomaly requiring countermeasures, or an emergency. The abnormalities are classified into abnormalities requiring countermeasures and other unregistered abnormalities (step 24).

If the abnormality is negligible, the plant operator will not be notified, and if the abnormality requires action, the system will automatically take action based on knowledge.
Step 25).

If it is determined in step 24 that there is an abnormality that requires countermeasures,
Using the cause determination knowledge 143, formulate a cause hypothesis for the detected abnormality, verify it with data in the plant database 13, and if the causal situation is not the final cause, use similar reasoning to determine the true cause. demand. When the true cause of the abnormality is determined, the abnormality countermeasure knowledge 144 is used to link the abnormality that has occurred to its cause and countermeasure (step 26).

When an anomaly is detected and the anomaly is analyzed, a display output device 17 such as an annunciator displays 1 depending on the rank of the anomaly.
The occurrence of a major abnormality, the occurrence of an emergency abnormality, the occurrence of an abnormality of unknown cause, etc. are displayed, and a bell is also emitted with a different sound so that these can be distinguished (step 27).

In addition, for cases where the cause of the abnormality has been determined, the abnormality occurrence location and the cause location are located on the display output device 1 such as a CRT.
7 is displayed in different colors on the plant flow display section (steps 28 to 30), and countermeasures for urgent abnormalities are also displayed at the same time (steps 31 and 32).

Incidentally, the abnormality that has occurred is recorded in the abnormality occurrence history database 18 regardless of its rank (step 33).
.

In this way, by performing plant monitoring based on the knowledge base 14, comprehensive judgments can be made based on subtle changes such as the rate of change and acceleration of individual plant data, the relationship between data, and comparison with design conditions. It evaluates the plant status and ranks the detected abnormalities, and automatically processes abnormalities that occur frequently and that require simple treatment, and automatically processes serious abnormalities, especially those that require urgent action. By displaying the causes and countermeasures for such abnormalities at the same time on the display output device 17, it is possible to quickly respond to such abnormalities.

Furthermore, even if there is no knowledge of cause determination and an abnormality appears for the first time, it is confirmed that it is an unexperienced abnormality and the plant operator is notified, so that a response can be taken and the knowledge base 14 can be updated.
Therefore, since the alarm method for detected abnormalities is differentiated depending on the rank, plant operators can operate the plant with ample time without putting an undue burden on them, and even non-skilled operators can safely and efficiently operate the plant. This will enable accurate plant operation.

[Effects of the Invention] As described above, according to the present invention, the knowledge base and inference processing means are used to rank the abnormality when a plant abnormality is discovered, and the system ranks the abnormality from mild to serious to urgent according to its rank. It is possible to output alarms and display the causes of abnormalities and countermeasures, allowing plant operators to immediately know the causes and countermeasures for abnormalities that require urgent countermeasures, which greatly reduces the burden of plant operation and monitoring. Ru.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of an embodiment of the plant monitoring device of the present invention, and FIG. 2 is a flowchart showing operations for determining an abnormality and displaying the rank of the abnormality in the above embodiment. 11...Detector 12...Input processing section 13
... Plant database 14 ... Knowledge base 141 ... Abnormality judgment knowledge 142 ... Ranking knowledge 143 ... Cause judgment knowledge 144 ... Abnormality countermeasure knowledge 145 ... Inexperienced abnormality judgment knowledge 15 ...・Inference processing unit 17...Display output device agent Patent attorney Noriyuki Chika

Claims (1)

[Claims] A plant data input unit for inputting plant status data, and a knowledge base storing knowledge for determining current and future plant status based on past and present plant data input by the plant data input unit. Anomaly determination knowledge for determining abnormalities in plant conditions, ranking knowledge for ranking abnormalities according to their urgency and importance, cause determination knowledge for investigating the true cause of the abnormality, and cause of the abnormality. Identify the abnormality countermeasure knowledge and the abnormality that appears for the first time that is not in the knowledge base,
The one that stores the inexperienced abnormality judgment knowledge to be recorded,
A plant comprising an inference processing section that performs inference processing by selecting and combining knowledge from this knowledge base, and an output section that automatically distinguishes and outputs measures such as warnings for abnormalities based on the results of this inference processing. monitoring equipment.