JPH04134597A - Monitoring device for abnormality of plant - Google Patents

Abstract

PURPOSE:To speedily grasp an abnormal phenomenon and to realize a running corresponding operation by identifying to which abnormal phenomenon a previ ously decided phenomenon is applied based on a condition and the instantaneous value of a processing signal when the abnormality detection condition which is previously decided is satisfied. CONSTITUTION:A real time processing means 22 judges the abnormality detection condition which is previously decided based on the inputted process signal S, identifies the previously decided abnormal phenomenon based on the abnormality detection condition and the instantaneous value of the related process signal S and obtains running operation guidance corresponding to the abnormal phenomeon. A cause identification means 23 tracks and monitors a present phenomenon from an abnormality cause which is previously decided coping with the abnormal phenomenon and identifies the specified abnormality cause. A display means 24 displays the running operation guidance obtained in a real time processing means 22 and the abnormality cause identified in the cause identification means 23. Thus, the content of the abnormal phenomenon can speedily be predicted and grasped and operation correspondence can be guided.

Description

DETAILED DESCRIPTION OF THE INVENTION [Object of the Invention] (Industrial Application Field) The present invention relates to a plant abnormality monitoring device used for monitoring large-scale plants such as nuclear power plants and thermal power plants.

(Prior Art) Generally, an abnormality during operation of a power plant appears as a change in a process signal. Therefore, if the value of the process signal or its rate of change deviates from a predetermined alarm value or automatic trip value, the operation is restricted by issuing a warning or stopping the operation. This allows process signals to be monitored during plant operation to prevent serious damage from occurring.

That is, continuous recording is performed using a recorder and interfaces with operators and technicians are performed using a display device to monitor not only current values but also trends in process signal changes.

Furthermore, there are cases in which the monitoring function is increased by increasing the sophistication of the abnormality detection function, such as frequency component analysis of shaft vibration changes in rotating machines.

(Problem to be solved by the invention) However, in the past, in order to respond to plant abnormalities that require alarms and trips, operators have used their experience and expertise to respond to plant abnormalities that require alarms and trips. Cause estimation and driving decisions are made based on this.

Therefore, damage may occur or expand due to a delay in taking measures when a plant abnormality occurs or due to misidentification of process signal contents. In other words, there are many types of abnormal phenomena, the values and trends of the process signals to be verified are also diverse, and the progress of abnormal phenomena is slow, so serious symptoms may be overlooked depending on the lack of skill of operators and technicians. . In addition, due to incomplete understanding of the situation, operational measures may be delayed, and incorrect driving operations may be performed. In such a case, there is a risk that the damage will expand and lead to an unsafe trip. In particular, damage often occurs or expands when restarting after a trip.

An object of the present invention is to obtain a plant abnormality monitoring device that enables quick grasp of abnormal phenomena and accurate operational response operations, and narrows down the cause of abnormalities based on the history of phenomenon changes after detection of abnormal phenomena. do.

[Structure of the Invention] (Means for Solving the Problem) The present invention includes a process input means for inputting a process signal of a plant, and a predetermined abnormality detection condition determined based on the input process signal to detect the abnormality. real-time processing means that identifies a predetermined abnormal phenomenon based on the abnormality detection condition and the instantaneous value of the process signal related thereto when the detection condition is satisfied, and obtains operation guidance corresponding to the abnormal phenomenon; A cause identification means that tracks and monitors the current phenomenon to identify a specific cause of the abnormality from among the causes of the abnormality set in advance in response to the problem, and an operation guidance obtained by the real-time processing means and the abnormality identified by the cause identification means. The present invention is characterized by comprising display means for displaying the cause.

(Function) As a result, when the predetermined abnormality detection conditions are met,
Based on the abnormality detection conditions and the instantaneous value of the process signal, it is determined which of the predetermined abnormal phenomena the abnormal phenomenon corresponds to, and the abnormal phenomenon is specified. Then, monitor and track the current phenomenon to narrow down the cause of the abnormality.

In other words, when an abnormality symptom is detected that satisfies the abnormality detection conditions set so that abnormal phenomena can be predicted before reaching an alarm or trip state, the abnormality sign is detected based on the abnormality detection conditions and the instantaneous value of the process signal at the time of detection. to identify abnormal phenomena0
Next, the cause of the abnormality is specifically narrowed down based on the process signal and abnormality detection conditions to which the candidate for the abnormality cause is related.

(Example) An example of the present invention will be described below. FIG. 1 is a block diagram in which the plant abnormality monitoring device 11 of the present invention is applied to shaft vibration monitoring of a power generation plant I2.

Vibration detectors 17 for detecting shaft vibrations S of each rotor are installed in the bearings 16 that support the rotors of the high-intermediate pressure turbine 13, the low-pressure turbine 14, and the generator 15, respectively. In addition, there is a tachometer 18 that detects the rotational speed N of the rotor, and each bearing part 1.
A phase reference pulse detector 19 is installed to serve as a reference for detecting the phase of the shaft vibration at 6, and receives various plant signals P indicating plant operating conditions such as generator output, steam temperature, pressure, bearing temperature, oil supply pressure, etc. Various sensors are installed for detection.

These process signals are read into the process input means 20 at a constant scanning cycle and stored in the data table 21. The real-time processing means 22 is a data table 21
Among the process signals stored in , the process signal S related to shaft vibration is first read, and it is determined whether a predetermined abnormality detection condition is satisfied. In other words, the abnormality detection conditions that are set so that abnormal phenomena can be predicted before reaching an alarm or trip state are determined, and if an abnormality symptom that satisfies the conditions is detected, the abnormality detection conditions and the A candidate for an abnormal phenomenon (single or multiple) is identified based on the instantaneous value of the process signal at the time of detection related to the condition.

On the other hand, the cause identification means 23 is the real-time processing means 22.
This is to identify the cause of the abnormal phenomenon identified in the process, and after the abnormal phenomenon occurs.

Monitor and track abnormality detection conditions and the instantaneous values of related process signals to narrow down the cause of the abnormality. Then, the result is output to the display device 24. Note that the display device 24 also displays a driving operation guide to be taken in response to the abnormal phenomenon identified by the real-time processing means 22.

Next, as abnormality detection conditions in the case of shaft vibration, the following are used. That is, (1) amplitude (a□), (2
) Amplitude change rate (a2), (3) Frequency component (a3), (
4) Define abnormality detection as follows using phase (a4), etc. As a result, the real-time processing means 2 for abnormality detection
2 can be automated.

(1) Use the predicted value (predicted current value) obtained from the calculation of the amplitude change rate shown in the next section for the amplitude +11 (a,).

(2) Using the past N pieces of data regarding the amplitude change rate (a2), the change therebetween is approximated by a straight line using the method of least squares, and the slope thereof is defined as the vibration change rate. When the scan time is ΔT, the time average rate of change is N×ΔT. This is shown in FIG. The amplitude change rate α obtained from 58 data up to Δτ ago becomes the current value.

(3) Regarding frequency component (a3), a threshold value is set for each of the following frequency components according to the operating state, and if any component exceeds the threshold value, it is determined that the frequency component is abnormal. Each frequency component is (rotation synchronous component ω.), (rotation synchronous double component 2ω.), (fractional harmonic component 1/2ω., 1/3ω.), (eigenvalue ω old P, ωL
F'A, ωLPB, ωGEN).

However, if the synchronous component exceeds the threshold value, it is treated as an amplitude abnormality (1).

(4) Regarding the phase (a4), the amplitude and phase values of the rotation synchronous component from the phase angle meter are monitored as the amount of change from reference data (actual values under normal operating conditions in the past).

Next, an abnormality level is set for each of these abnormality detection conditions as shown in FIG. The condition is satisfied if the caution value shown in the figure is exceeded. Further, the condition is satisfied when at least one of the detection conditions is satisfied. Therefore, each level obtained by ranking the abnormality level at that time becomes a trigger condition for identification of an abnormal phenomenon by the real-time processing means 22. In this ranking, as shown in FIG. 3, a caution value is placed before the conventional alarm value, and a stop value is placed before the trip value. This would set up a conventional alarm or anomaly prediction opportunity to detect anomaly symptoms before they result in a trip. That is, a time margin of ΔT□ and ΔT2 is provided.

In the end, four abnormality levels are determined for each abnormality detection condition as shown in FIG. 3, and driving operation guidance is determined corresponding to the abnormality level as shown in FIG. Driving operation guidance when the abnormality level has been reached is displayed on the display device 24.

That is, FIG. 4 is a driving operation guidance table stored in advance in the real-time processing means 22, and the driving operation guidance can be displayed appropriately depending on the abnormality level of the abnormality detection condition for shaft vibration. It is something.

Next, identification of an abnormal phenomenon when the abnormality detection condition is satisfied is performed by the real-time processing means 22 as follows. That is, as shown in FIG.
The abnormality is identified based on this abnormal phenomenon table. The real-time processing means 22 selects an abnormal phenomenon P that holds true based on the abnormality detection condition a that holds true and the process data b. For example, abnormality detection condition axr 8
2 is established and the process data b1 exceeds the reference value, it is identified as an abnormal phenomenon P4. Here, in Figure 5, if it is positively affirmed, ○, if it is denied, ×,
If there is no involvement, it is matrixed as unmarked as it is not applicable and a positive or negative decision cannot be made.

Next, when an abnormal phenomenon is identified, the real-time processing means 5 triggers the cause identification means 23 to identify the specific cause for which countermeasures have been known from past operation, maintenance, and design experience. For example, Fig. 6 shows candidate causes α for the abnormal phenomenon P of shaft vibration.
Regarding U, cause candidate α is classified in advance.Here, cause candidate α,j is the J-th cause belonging to the abnormal phenomenon P, and the countermeasures have been materialized based on past operational experience. be.

That is, the cause identification means 23 stores the cause candidate table shown in FIG. 6 in advance, and the real-time processing means 22
The cause candidate α1 corresponding to the abnormality #P□ identified in is identified based on this cause candidate table. On the other hand, the cause identification means 23 also stores a tracking item table shown in FIG. 7, and based on this table, it tracks and monitors cause candidates α□ corresponding to abnormal phenomenon P ,
Identify and narrow down j. In other words, the anomaly detection condition α for cause candidate α, 2 and the characteristics of changes in process demonstration are organized in advance as shown in Figure 7 based on past cases and knowledge, and the cause candidate table shown in Figure 6 is When the cause candidate α, 4 identified in 4 satisfies the conditions of the tracking item table in FIG. 7, the cause candidate α, 4 is determined to be the cause of the abnormal phenomenon P.

Thereby, by applying the cause identification means 23 as a follow-up and monitoring period from the detection of an abnormality symptom by the real-time processing means 22 to the generation of an alarm, trip, or stoppage, it is possible to further narrow down the number of cause candidates.

The above explanation deals with shaft vibration as a target for plant abnormality monitoring, but abnormality detection conditions can also be defined for heater water level, turbine casing upper and lower temperature difference, condenser vacuum level, auxiliary steam header pressure, etc. Driving guidance can be set according to the abnormality level, and ultimately everything from classification of abnormality level to driving guidance and narrowing down of the cause of the abnormality can be done automatically.

[Effects of the Invention] As described above, according to the present invention, it is possible to quickly predict and grasp the content of abnormal phenomena and guide operational responses by detecting abnormality signs early in the monitoring of power plants. can. In other words, before the alarm is tripped or stopped, the number of cause candidates can be narrowed down based on the history of phenomenon changes after the abnormality symptom is detected.

Therefore, specific countermeasures can be further narrowed down, and it becomes possible to prevent the occurrence of alarms, trips, and stoppages that have been avoided in the past.

At the same time, it is possible to prevent delays and errors in driving operations, as well as the occurrence and spread of damage caused by them.

[Brief explanation of drawings]

Figure 1 is a block diagram in which the plant abnormality monitoring device of the present invention is applied to shaft vibration monitoring of a power generation plant, Figure 2 is an explanatory diagram for determining the rate of change and predicted value of a process signal, and Figure 3 is an illustration of abnormality detection. Fig. 4 is an explanatory diagram of the abnormality level of the conditions, Fig. 4 is an explanatory diagram of the driving operation guidance table, Fig. 5 is an explanatory diagram of the abnormal phenomenon table, Fig. 6 is an explanatory diagram of the cause candidate table, and Fig. 7 is an explanatory diagram of the tracking item table. It is an explanatory diagram. DESCRIPTION OF SYMBOLS 11... Plant abnormality monitoring device, 12... Power generation plant, 20... Process input means, 21... Data table, 22... Real-time processing means, 23...
- Cause identification means, 24... display device. Figure 1 Agent Patent Attorney Crest 1) Makoto Figure Figure

Claims (1)

[Claims] A process input means for inputting a process signal of the plant, and a predetermined abnormality detection condition is determined based on the input process signal, and when the abnormality detection condition is satisfied, the abnormality detection condition and the related process signal are determined. A real-time processing means that identifies a predetermined abnormal phenomenon based on the instantaneous value and obtains operational guidance corresponding to the abnormal phenomenon; The present invention is characterized by comprising cause identification means for tracking and monitoring to identify a specific cause of an abnormality, and a display device for displaying driving operation guidance obtained by the real-time processing means and the cause of the abnormality identified by the cause identification means. Plant abnormality monitoring device.