JPH08249046A - Alarm estimate station - Google Patents

Abstract

PURPOSE: To improve the reliability and safety of an alarm estimate station by estimating the occurrence of abnormality caused by the change of the process value and evading an unexpected situation. CONSTITUTION: An alarm estimate station includes a 1st data base 21 which stores a history and a variance pattern of the past process value data, and a 2nd data base 22 which stores an operating procedure to cope with the variance pattern. Then a variance pattern addition means 24 adds the present variance pattern to the process value data based on the variance contents of the present process value data. The data estimate means 26, 27 and 28 estimate the presence or absence of abnormality for the process value data of the near future based on the history of the past process value data which are retrieved out of the data base 21 based on the present variance pattern. When the presence of abnormality is estimated, a cope-with procedure output means retrieves the data base 22 and outputs an operating procedure to a network through a monitor control panel to cope with the abnormality.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a supervisory control system for supervising and controlling a process amount, and in particular, it can predict an abnormal occurrence due to a change in the process amount to avoid an emergency and improve reliability and safety. Alarm prediction station.

[0002]

2. Description of the Related Art Conventionally, in the field of monitoring and controlling the process amount of a plant, a monitoring control system for monitoring the process amount and controlling the plant based on this process amount has been widely used.

In this type of supervisory control system, a supervisory control board 1 having a CRT 1a and a keyboard 1b is connected to a control station 3 via a network 2, as shown in FIG.

The control station 3 collects the process amount of the plant and sends it to the supervisory control panel 1 via the network 2. On the monitor control panel 1, this process amount is recorded by a recorder or the like while being displayed on the CRT 1a. When the process amount shows an abnormal value, the monitoring control panel 1 issues an alarm to notify the operator of the abnormality. When the operator detects an abnormality by an alarm, in the monitoring control panel 1, a control command for controlling the plant is sent to the control station 3 via the network 2 by the operation of the keyboard 1b by the operator. The control station 3 controls the plant according to this control command.

[0005]

However, in the monitoring control system as described above, since the process amount is dealt with after the abnormal value is shown, the time for investigating the cause of the abnormality and the time until the normal operation is restored are taken. May become enormous and the production efficiency of the plant may be reduced to lower the reliability.

Further, depending on the process amount and the details of the abnormality, there is a possibility that an unexpected situation may occur in the plant. However, dealing with this type of situation requires experience,
If an unexpected situation is brought about, it will be difficult for an operator who has no experience to deal with it accurately, and reliability and safety will be reduced. Therefore, it is necessary to avoid entry into an unexpected situation due to the occurrence of an abnormality.

The present invention has been made in consideration of the above situation, and provides an alarm prediction station capable of predicting an abnormal occurrence due to a change in process amount, avoiding an unexpected situation, and improving reliability and safety. The purpose is to

[0008]

The invention according to claim 1 relates to a supervisory control panel for collecting process amount data from a controller via a network and controlling the controller via the network. An alarm prediction station that is connected and capable of collecting the process amount data, for storing a history of past process amount data and a variation pattern thereof, and a countermeasure when the variation pattern is shown. When the current process amount data on the network and the second database in which the operation procedure is stored show a change, a change pattern adding means for adding the current change pattern to the process amount data based on the contents of the change. According to the current fluctuation pattern added by this fluctuation pattern adding means. The first database is searched to obtain a history of past process amount data corresponding to the current variation pattern, and whether there is an abnormality in the near future process amount data is determined based on the history of the past process amount data. A data predicting means for predicting, and when the data predicting means predicts that there is an abnormality in the near future process amount data, searches the second database to take measures against the fluctuation of the present process amount data. The alarm predicting station is provided with a handling procedure output means for outputting the operating procedure of (1) to the network toward the monitoring control board.

The invention according to claim 2 is, in the alarm prediction station according to claim 1, common information for associating the stored contents of the first and second databases with each other in each of the databases. The coping procedure output means that is set is an alarm prediction station that searches the second database based on the common information.

[0010]

Therefore, the invention corresponding to claim 1 provides the first database in which the history of the process amount data in the past and its variation pattern are stored and the variation pattern by taking the above means. A second database storing operation procedures for coping with the situation is provided, and when the variation pattern adding means indicates a variation in the current process amount data on the network, the process amount data is converted into the process amount data based on this variation content. The current variation pattern is added, and the data predicting means searches the first database according to the current variation pattern added by this variation pattern adding means to obtain a history of past process amount data corresponding to the current variation pattern. Then, based on the history of the past process amount data, the presence / absence of abnormality in the process amount data in the near future is predicted, and corrective action is taken. When the output unit predicts that there is an abnormality in the near future process amount data by the data estimating unit, the output unit searches the second database and monitors the operation procedure for coping with the fluctuation in the current process amount data. Since the data is output to the network toward the control panel, it is possible to predict an abnormal occurrence due to a change in the process amount, avoid an unexpected situation, and improve reliability and safety.

In the invention according to claim 2, common information for associating the stored contents of the first and second databases with each other according to claim 1 is set in each database, and the coping procedure output means is provided. Since the second database is searched based on the common information, it is possible to facilitate the search of the database in addition to the effect corresponding to claim 1.

[0012]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing a configuration of a supervisory control system having an alarm prediction station according to an embodiment of the present invention. The same parts as those in FIG. 5 are designated by the same reference numerals and detailed description thereof will be omitted. Only different parts will be described.

That is, the system of the present embodiment is intended to avoid an unexpected situation by predicting the occurrence of an abnormality due to a change in the process amount, and more specifically, predicts the occurrence of an abnormality as compared with the apparatus shown in FIG. Alarm prediction station 10 for
Is connected to the network 2.

Here, the alarm prediction station 10
Has a hardware configuration including a power supply 11, a mass storage device 12, a CPU 13, a work memory 14, and an I / O unit 15. Specifically, as shown in FIG. 21), coping method DB 22, data collecting section 23, feature adding section 24, process DB building section 25, process DB searching section 26, alarm predicting section 27, display determining section 28, DB related section 29, coping method DB building section. A functional block configuration including an information display unit 31 and an information display unit 31 is provided. In the mass storage device 12, the process DB 21 and the coping method DB 22 are constructed, and the data collecting unit 23, the feature adding unit 24, the process DB constructing unit 25, the process DB searching unit 26, and the alarm predicting unit 27. The CPU 1 corresponding to the display determination unit 28, the DB related unit 29, the countermeasure DB construction unit 30, and the information display unit 31.
The operation program of No. 3 is stored.

As shown in FIG. 3A, the process DB 21 stores a serial number SE for identifying the process amount data, a data source SO, and other related process amounts for each process amount data indicating a variation. Related information CF indicating, data number NU, data string D indicating actual process amount data
A, characteristic information SD indicating the pattern of the actual process amount data
The process data record PR consisting of the items is stored.

As shown in FIG. 3B, the countermeasure DB 22 has a serial number SE and a keyboard 1 for identifying the process amount data for each process amount data indicating an abnormality.
A remedy data record OR including items of an operation record OP indicating a history (operation procedure) of control commands sent from the network b to the network 2 is stored. The serial number SE is common information for associating the contents with the process DB 21 and the countermeasure DB 22.

The data collection unit 23 has a function of collecting process amount data from the control station 3 via the network 2 and notifying the feature amount giving unit 24 of the process amount data.

When a change occurs in the process amount data notified from the data collection unit 23, the feature adding unit 24 adds feature information indicating the feature of this change and the process amount data to which the feature information is added. It has a function to give to the process DB construction unit 25 and the process DB search unit 26.
Here, the characteristics of the fluctuation include, for example, periodic vibration, random vibration, vibration indicating a fall, vibration indicating a rise, divergence, periodic pulsation, sudden rise, sudden fall, etc. It can be identified by examining the pattern.

The process DB construction unit 25 includes a feature addition unit 2
When the process amount data and the characteristic information thereof are given from 4, the process data record PR is constructed based on the process amount data and the characteristic information, and the process data record PR is added / updated to the process DB. Since the related information CF and the data string DA are time-series history information, the process data record PR has a property of being sequentially constructed from the occurrence of variation to the avoidance of abnormality.

The process DB search unit 26 includes the feature adding unit 2
4, when the process amount data and its characteristic information are given, the process DB is searched to extract the process data record PR to which the same characteristic information as this characteristic information is added, and the extracted process data record PR is alarm predicted. It has a function of giving to the section 27.

The alarm predicting section 27 predicts a change in the process amount data in the future based on the process data record PR given from the process DB searching section 26 to obtain an abnormality occurrence probability, and the obtained abnormality occurrence probability. Is provided to the display determination unit 28. As the probability of occurrence of abnormality, for example, it may be preset that 100% abnormality will occur in the case of the same variation pattern, or if the same variation pattern is displayed and left as it is. It may be calculated by counting the occurrence of an abnormality.

The display determination unit 28 compares the abnormality occurrence probability given by the alarm prediction unit 27 with a predetermined threshold value,
When the probability of occurrence of abnormality exceeds a threshold value, the monitor control panel 1 is caused to display the predicted future change in the process amount data and the operation record OR corresponding to the process data record PR having the same past characteristic information. Therefore, it has a function of giving a prediction result display command including the serial number SE to the DB related unit 29 and the information display unit 31.

When the prediction result display command is received from the display determination unit 28, the DB relation unit 29, based on the prediction result display command, associates the process DB 21 and the coping method DB 22 with the current process amount data. Serial number SE and operation record OP for the current process
It has a function of giving the saving method and the coping method DB construction unit 30.

The coping DB construction unit 30 is a DB relation unit 29.
When the current serial number SE and the save command for the operation record OP are received from the serial number S, the serial number S
A remedy data record OR including items of E and the operation record OP is constructed, and the constructed remedy data record O is constructed.
It has a function to add / update R to the coping method DB 22. The remedy data record OR is the operation record OP.
Since is a time-series history information, it has the property of being constructed sequentially from the occurrence of fluctuations to the avoidance of abnormalities.

When the information display section 31 receives a prediction result display command from the display determination section 28, based on this prediction result display command, the coping method data record O in the coping method DB 22.
The search control panel 1 is searched for R, and the operation record OR corresponding to the process data record PR having the same past characteristic information is monitored and controlled.
It has a function of sending to the network 2 so as to be displayed on the CRT 1a.

The information display section 31 also includes a display determination section 28.
When the prediction result display command is received from, the process data record PR in the process DB 21 is searched based on the prediction result display command, and among the item contents of the process data record PR having the same past characteristic information, At least a change in the actual process amount data is monitored by the CRT of the control panel 1.
It has a function of sending it to the network 2 so that it can be displayed on 1a.

Next, the operation of the supervisory control system and the alarm prediction station configured as above will be described with reference to the flowchart of FIG. The data collection unit 23
The process amount data is collected from the control station 3 via the network 2 (ST1), and the process amount data is notified to the feature adding unit 24.

The characteristic imparting unit 24 determines whether or not the notified process amount data has a change (ST2), and when the process amount data has a change, adds characteristic information indicating the characteristic of this change ( In ST3), the process amount data to which the characteristic information is added is given to the process DB construction unit 25 and the process DB search unit 26.

The process DB construction unit 25 includes the feature addition unit 2
4. When the process amount data and its characteristic information are given from 4, the construction of the process data record PR is started based on the process amount data and its characteristic information (ST
4).

The process DB search unit 26 includes the feature addition unit 2
4, when the process amount data and its characteristic information are given, the process DB 21 is searched based on the characteristic information to extract the process data record PR having the same characteristic information as this characteristic information (ST5). The extracted process data record PR is given to the alarm prediction unit 27.

The alarm predicting unit 27 predicts future changes in the process amount data based on the process data record PR given from the process DB searching unit 26 (S
T6) The probability of occurrence of abnormality is calculated (ST7), and the calculated probability of occurrence of abnormality is given to the display determination unit 28.

The display determination unit 28 compares the abnormality occurrence probability given by the alarm prediction unit 27 with a predetermined threshold value to determine whether there is a possibility of abnormality occurrence (ST8), and this abnormality occurrence probability is threshold. When the value exceeds the value, it is determined that an abnormality may occur, and the process data record PR has the same future predicted change in the process amount data and the same characteristic information in the past.
The operation record OR corresponding to the CRT 1a of the monitoring control panel 1
In order to display it on the screen, a prediction result display command including the serial number SE is given to the DB related unit 29 and the information display unit 31.

Further, the display judging section 31 carries out step ST8.
When there is no possibility of occurrence of abnormalities in the keyboard 1b
It is determined whether or not the display request from the keyboard 1b has been received (ST9). When the display request is not received from the keyboard 1b, the determination of step ST9 is repeated. DB
It is given to the related part 29 and the information display part 31.

When the information display unit 31 receives the prediction result display command from the display determination unit 28, the countermeasure DB 22 is based on the serial number SE included in the prediction result display command.
In the process DB 21 based on the prediction result display command while obtaining the operation record OR corresponding to the past process data record PR having the same characteristic information.
Of the process data record PR having the same characteristic information in the past, the change in the actual process amount data is obtained, and the display data including the change in the past operation record OR and the actual process amount data is obtained. (ST1
0), send this display data to the network 2 (S
T11) Display on the CRT 1a of the monitor control panel 1.

By displaying such past operation records,
The key input operation of the operator for coping is recorded (ST12). That is, when the prediction result display command is received from the display determination unit 28, the DB related unit 29 associates the process DB 21 with the coping method DB 22 based on the prediction result display command (ST13), and the current process. Remedy D for the serial number SE in the quantity data and the save command of the operation record OP related to the current process
It is given to the B construction unit 30.

The coping method DB construction unit 30 is a DB related unit 29.
When the current serial number SE and the save command for the operation record OP are received from the serial number S, the serial number S
The construction of a coping method data record OR including E and the item of the operation record OP is started (ST14).

Next, by the operation of the operator for coping,
It is assumed that the abnormality is avoided. The coping method DB construction unit 30 finishes the construction of the coping method data record OR whose construction is started from step ST14 by the operation of the operator, and writes the coping method data record OR to the coping method DB 22 (ST1.
5).

Similarly, the process DB construction unit 30 finishes the construction of the process data record PR whose construction has started from step ST4 by the operation of the operator, and writes the process data record PR in the process DB 21 (S).
T16).

As described above, according to the present embodiment, when the current process amount data on the network 2 indicates a change, the feature adding section 24 adds the feature information to the process amount data based on the contents of the change. , Process DB search unit 26
However, the process DB 21 is searched based on this characteristic information to obtain the history of the past process amount data corresponding to the current variation pattern, and the alarm prediction unit 27 determines the future based on the history of the past process amount data. The change probability of the process amount data is predicted to calculate the probability of occurrence of abnormality, and the display determination unit 28 predicts whether or not there is a possibility of occurrence of abnormality in the process amount data in the near future based on the probability of abnormality occurrence. Remedy D when it is predicted that an abnormality may occur
B22 is searched and the operation procedure for coping with the current fluctuation of the process amount data is output to the network 2 toward the monitor control panel 1, so that an unexpected occurrence due to a change in the process amount is predicted and an unexpected situation. Can be avoided, and reliability and safety can be improved.

Further, according to the present embodiment, the process DB2
1 and the serial number SE for associating the stored contents of the countermeasure DB 22 with each other are set in the DBs 21 and 22, respectively, and the information display unit 31 searches the countermeasure DB 22 based on the serial number SE. The search can be facilitated.

Further, the operation record OP may be recorded by executing a procedure different from the displayed past operation procedure depending on the judgment of the operator. That is, since the operation record OP is accurately specified by the learning effect, when the same abnormal situation occurs frequently, the accuracy of the prediction and the example of the coping method become excellent, and the abnormality investigation and the automatic operation are performed. You can expect it to be a great help in driving. In addition, the present invention can be modified in various ways without departing from the scope of the invention.

[0042]

As described above, according to the first aspect of the invention, the first database in which the history of process amount data in the past and its variation pattern are stored, and the countermeasure for dealing with the variation pattern, are provided. A second database in which operating procedures are stored is provided, and when the fluctuation pattern adding means indicates fluctuations in the current process amount data on the network, the current fluctuation pattern is added to the process amount data based on the fluctuation contents. Then, the data predicting means searches the first database according to the current variation pattern added by the variation pattern adding means, obtains a history of past process amount data corresponding to the current variation pattern, and calculates the past process amount. Based on the history of the data, it is predicted whether there is an abnormality in the near future process amount data, and the coping procedure output means uses this data. When the measuring means predicts that there is an abnormality in the process amount data in the near future, the second database is searched, and the operation procedure for coping with the fluctuation of the current process amount data is directed to the monitoring control panel through the network. Therefore, it is possible to provide an alarm prediction station capable of predicting an abnormal occurrence due to a change in the process amount, avoiding an unexpected situation, and improving reliability and safety.

According to the invention of claim 2, claim 1
Common information for associating the stored contents of the first and second databases with each other is set in each database,
Since the coping procedure output means searches the second database on the basis of the common information, it is possible to provide an alarm prediction station capable of facilitating the search of the database in addition to the effect of claim 1.

[Brief description of drawings]

FIG. 1 is a block diagram showing the configuration of a supervisory control system having an alarm prediction station according to an embodiment of the present invention,

FIG. 2 is a functional block diagram showing a configuration of an alarm prediction station in the embodiment.

FIG. 3 is a schematic diagram showing stored contents of a database in the embodiment.

FIG. 4 is a flow chart for explaining the operation of the embodiment.

FIG. 5 is a block diagram showing the configuration of a general supervisory control system.

[Explanation of symbols]

21 ... Process DB, 22 ... Remedy DB, 23 ... Data collection unit, 24 ... Feature giving unit, 25 ... Process DB construction unit, 26 ... Process DB search unit, 27 ... Alarm prediction unit, 28 ... Display determination unit, 29 ... DB related section, 30 ... Remedy DB construction section, 31 ... Information display section.

Claims (2)

[Claims] 1. An alarm predictor capable of collecting the process amount data from a controller, which collects the process amount data via the network and controls the controller via the network. A station, a first database in which a history of past process amount data and its variation pattern are stored, and a second database in which an operation procedure for coping with the variation pattern is stored, When the current process amount data on the network shows a fluctuation, a fluctuation pattern adding means for adding the current fluctuation pattern to the process amount data based on the contents of the fluctuation, and a current pattern added by the fluctuation pattern adding means. Searching the first database according to a variation pattern, A data prediction unit that obtains a history of past process amount data corresponding to the present variation pattern and predicts whether there is an abnormality in the near future process amount data based on the history of the past process amount data; When it is predicted by the means that there is an abnormality in the near future process amount data, the second database is searched, and an operation procedure for coping with a change in the current process amount data is directed to the monitoring control panel. And a countermeasure output means for outputting to the network. 2. The alarm prediction station according to claim 1, wherein common information for associating the stored contents of the first and second databases with each other is set in each of the databases, and the coping procedure output means is provided. An alarm prediction station, wherein the second database is searched based on the common information.