JP3550060B2 - Plant operation support system - Google Patents

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a plant operation support system using a computer, and particularly, when a displacement typified by an abnormality occurs in a plant, evaluates the state of a process device such as an absorption tower, a reactor, causes of an event, and an event. The present invention relates to an operation support system for a plant that predicts a change of a plant and outputs a coping method.
[0002]
[Prior art]
In the process industries such as chemistry, energy, food, and medicine, a computer system (computer) is generally introduced to control the operation of the plant and improve the operation efficiency of the plant. In recent years, in operation control of this type of plant, higher control and higher functionality have been required.
[0003]
On the other hand, a reduction in the number of plant operators is required in order to streamline plant operation and save labor. For this reason, plant operation monitoring from a conventional panel display operation monitoring to a computer equipped with a distributed control system using a CRT console has become the mainstream.
[0004]
The distributed control system CRT display not only displays process measurement data such as flow rate, temperature, pressure, etc., but also displays the measurement data or the rate of change of the measurement data at a preset upper or lower alarm level. A function is provided for outputting a warning message, a guidance message to the operator, and the like at a stage beyond the limit. In response to these messages, the operator operates various control devices of the plant, cancels the state displacement that has occurred in the plant, and returns to the standard state.
[0005]
[Problems to be solved by the invention]
However, in the conventional distributed control system, when the measured data or the rate of change of the measured data exceeds the set upper and lower limit values, only an alarm indicating that a state displacement has occurred in the operation state of the plant is displayed. It is. Therefore, there is a problem that the cause of occurrence of the state displacement or a measure for eliminating the state displacement and returning to the standard state largely depends on the judgment of the operator, and it requires time and experience.
[0006]
An object of the present invention is to evaluate a state of a plant and predict a change in an event when a state displacement representing the abnormal situation occurs during a plant operation of the plant including start-up and shutdown, and cope with the situation. The present invention provides a plant operation support system that outputs the guidance of (1).
[0007]
In order to achieve the above object, a plant operation support system according to the present invention evaluates the state of a plant when a state displacement occurs in an operating plant including start and stop, and evaluates the state of the event. In a plant operation support system including a computer system that predicts a change and outputs a coping method,
And plant database for storing process data collected during plant operation, assumed in advance the contents of the case where the state displaced occurs in the plant, target site, event contents, factors, knowledge diagnosis contents of such spillover events and necessary actions evaluated and knowledge database for storing the sheets, and current data from the plant data accumulated the diagnostic contents knowledge database having stored in said plant database, searches the state compatible diagnostic for current plant operation A processing device, and an inference result database for storing a diagnosis result of the arithmetic processing device.
The upper and lower limits of the analog data of the plant are divided into three equal parts, the analog data is converted into internal data from -3 to 3, and a vector sequence of the internal data is obtained,
An evaluation point is calculated based on an inner product of the vector sequence of the acquired internal data and the vector sequence of the transmission degree defined at the time of inputting the alarm signal information of the knowledge aggregation sheet, and the alarm signal information of the knowledge aggregation sheet having the largest evaluation point is calculated. It is characterized in that priority processing is performed .
[0008]
The arithmetic processing unit can have a function of determining a priority level of the state displacement to be eliminated from the diagnosis result. Further, the arithmetic processing unit may have a function of comparing the current diagnosis result with the previous diagnosis result and detecting a change in the priority level of the state displacement.
[0009]
Furthermore, the arithmetic processing unit may have a function of automatically selecting process data related to the occurrence of a state displacement from a diagnosis result based on a knowledge database.
[0010]
The plant operation support system includes a man-machine device having a display unit and an operation unit, and the man-machine device includes a display unit that acquires a calculation result from an inference result database and displays data for plant operation control. be able to.
[0011]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, embodiments of the present invention will be described with reference to the drawings.
[0012]
FIG. 1 is a block diagram showing the configuration of one embodiment of the plant operation support system of the present invention. The arrows between the components indicate the flow of information or signals.
[0013]
As shown in FIG. 1, a plant operation support system 1 of the present embodiment includes a computer system 2, a man-machine device 3 connected to the computer system 2, an instrumentation control system 5 of a plant 4, and another operation support system. 6 has a communication device 10 for taking in plant data.
[0014]
The computer system 2 includes a plant database 11 for storing plant data captured via the communication device 10, a knowledge database 12 for storing diagnosis contents of the plant, and a B & M (builder & maintenance) for generating the knowledge database 12. And 13) a function 13. Further, the computer system 2 has an arithmetic processing unit (collation engine) 14 for diagnosing the state of the plant using the plant database 11 and the knowledge database 12, and an inference result database 15 for storing the diagnosis result.
[0015]
In this embodiment, the B & M function 13 is mounted on the computer system 2 that constitutes the plant operation support system 1. However, the B & M function 13 is mounted on a different computer system, and the knowledge database generated there is stored in the computer system 2. It may be configured to load the data into the database 12.
[0016]
The plant operation support system 1 can also send the diagnosis result stored in the inference result database 15 to the instrumentation control system 5 or another computer system via the communication device 10.
[0017]
The man-machine device 3 includes a display unit 20 and an operation unit 21 for monitoring operation of a plant.
[0018]
The display unit 20 has, for example, a CRT display, and can display various calculation results of the computer system 2, contents of various databases, and the like. The operation unit 21 includes, for example, all or a part of input operation devices such as a keyboard, a mouse, and a touch panel, and receives various instructions to the computer system 2.
[0019]
Using this man-machine device 3, measurement, analysis, vibration diagnosis, and the like performed offline can be taken into the plant database 11 by the operation of the operator.
[0020]
Next, details of the function of the plant operation support system of the present embodiment will be described, but the present invention is not limited to this embodiment. Prior to the description of the function of the plant operation support system, a configuration of an example of plant equipment to which the present embodiment is applied will be described with reference to FIG. In FIG. 2, the arrows between the components indicate the flow of the fluid.
[0021]
The plant equipment shown in FIG. 2 includes a distillation column 30, a condenser 31, a reflux tank 32, and a reboiler 33 as process devices.
[0022]
This plant supplies a feedstock comprising a mixture of components A and B to a distillation column 30 via a pipe 40, and in the distillation column 30, a liquid (distillate) having a high concentration of the low-boiling component A from the top of the column. The liquid having a high concentration of the high boiling point component B (boiler effluent) is taken out from the bottom of the column.
[0023]
At the bottom of the column, there is a reboiler 33 for heating and evaporating the bottoms. As a heating method, a heat exchanger method using steam is used. The bottom liquid is taken out from the bottom of the column through a pipe 41.
[0024]
On the other hand, a condenser 31 for condensing steam is provided at the top of the tower, and a heat exchange method using water is used as a cooling method. The condensed vapor is stored in a reflux tank 32, a part of which is refluxed to the distillation column 30 via a pipe 42, and the remainder is taken out from a pipe 43 as a distillate.
[0025]
The plant operation support system of the present embodiment applied to this plant has the same configuration as the plant operation support system shown in FIG.
[0026]
During the operation of the plant, it is assumed that an initiating event occurs in which the steam back pressure of the reboiler exceeds the lower limit and the temperature in the distillation column exceeds the lower limit. At this time, the warning message is printed, can also be informed that a failure has occurred in the operating state of the plant to the operator.
[0027]
Next, the collation engine 14 of the plant operation support system operates to extract from the knowledge database 12 diagnosis contents such as a target part corresponding to the occurrence of the state displacement of the plant, an event content, an abnormal factor, a ripple event, and necessary countermeasures. I do.
[0028]
The details of the abnormality, the necessary measures, and the results of the diagnosis such as future transmission events are displayed on the CRT screen 20. As for the display format, for example, as shown in FIG. 3, the display is performed from the top to the bottom in the order of the priority level 50 of the state displacement. Thus, the operator merely looks at the display, and the display column 51 of the displacement factor indicates, in order of priority, (1) abnormality of the steam source, (2) failure of the valve or instrument failure, and Can be recognized as (1) inspection of the receiving line or communication with the power, (2) inspection of the valve opening or adjustment of the manual valve.
[0029]
As for the necessary countermeasures, as shown in FIG. 4, a detailed display column 60 of the necessary countermeasures to be performed by the operator is displayed, and whether or not the countermeasures have been performed can be checked by the check box 61.
[0030]
As for the spread event, as shown in FIG. 5, a display column 70 of a fault tree diagram is displayed, and the flow of an abnormal phenomenon and the operation of the plant that can occur when the system is left without taking necessary measures can be confirmed. For example, if the steam source is left without taking necessary measures for abnormality (insufficient heat source), the feedstock to the distillation column flows down to the bottom of the column, and the column pressure cannot be controlled.
[0031]
While the process data is transmitting the alarm, the abnormality diagnosis of the plant is executed at regular intervals. The present diagnosis result is compared with the previous diagnosis result, and when the priority level of the state displacement changes, the change is displayed, for example, as shown in FIG.
[0032]
In this embodiment shown in FIG. 3, the first and second priority levels are switched, and the priority 53 at the time of the previous diagnosis and the mark 54 indicating a change in the priority indicating ascending or descending are displayed. ing. By observing the temporal change of the priority level, it is possible to predict which abnormality of which apparatus and which part will be diagnosed next as an abnormality having a higher priority level, and it is possible to take measures in advance.
[0033]
Process data related to the occurrence of state displacement can be automatically selected and displayed based on the knowledge database.
[0034]
FIG. 6 shows a display example of data in the present embodiment. By automatically selecting the process data that is transmitting the alarm and the process data that is likely to be transmitted, the process data to be monitored with priority to the operator can be presented. It should be noted that the process data 80 issuing the alarm can be emphasized by changing colors or blinking.
[0035]
With the above-described operation support system, the operator can diagnose the state of the plant in a short time with the same knowledge and experience as the experienced operator, and can appropriately cope with the state displacement.
[0036]
Next, a description will be given of a definition operation of a diagnosis content such as a target part, an event content, an abnormal cause, a spillover event, and necessary countermeasures, assuming a state of the process device in advance when a state displacement occurs in the plant. Hereinafter, this diagnosis content will be referred to as a knowledge aggregation sheet. All the definition operations are performed through the B & M function 13.
[0037]
In this embodiment, a knowledge information definition screen shown in FIG. 7 is prepared. This provides a means for easily defining the knowledge aggregation sheet for the selected device.
[0038]
The operator calls the B & M function 13 from the computer system 2 of the present embodiment on the man-machine device 3 to display a graphic screen common to the operation monitoring screen as shown in FIG. Perform definition operation via. The designation of the device to be diagnosed can be performed by, for example, double-clicking an area where the target device is displayed using a mouse.
[0039]
By this designation, a definition screen (an example of a heat exchanger in FIG. 7) 90 of the knowledge information of the corresponding device is output on the screen of the display unit 20. The items defined here can be divided into two items: common information 91 and knowledge editing information 100.
[0040]
In FIG. 7, a definition screen of a definition field 91 of common information includes a process name input area 92, a group name input area 93, a sheet name input area 94, an execution cycle input area 95, a screen number input area 96, And a start mode input area 97 when the system is started.
[0041]
According to this definition example, it is a definition screen of a knowledge aggregation sheet for diagnosing the operation state of the distillation column for the apparatus group of "Group 2" in the "A plant".
[0042]
Also, it is shown that the diagnosis is executed every "300" seconds during the normal state and every "60" seconds during the occurrence of the abnormality, and the diagnosis is automatically executed when the diagnosis system is started ("START"). As the screen number, the number “1000” specified at the time of creating the graphic screen shown in FIG. 2 is automatically input.
[0043]
In FIG. 7, the definition screen of the definition field 100 of the knowledge editing information includes an input area 101 for information about the device, an input area 102 for information about the cause, an input area 103 for alarm signal information, an input area 104 for information about the spread, And an input area 105 for information on the measures. The definition of the knowledge editing information is made for each target part of the device to be diagnosed.
[0044]
According to this definition example, it is a definition screen of a knowledge aggregation sheet for abnormalities at “steam supply side boundary” in “reboiler RB-101”.
[0045]
In the input of the alarm signal information, the tag to be a condition target of the alarm output and the transmission degree (1 to 3) are input. According to the definition example of the alarm signal information, there are three tags that are the condition targets of the alarm output. For example, for the tag “TRB001”, the lower limit “L” of the process data “PV” and the transmission degree “3” Is defined.
[0046]
The input of information on the propagation defines future propagation events, their severity, and the conditions for propagation. The screen number “5000” in the display column 70 of the fault tree diagram shown in FIG.
[0047]
The knowledge aggregation sheet defined as described above can be displayed by the B & M function 13 in two tabular forms, a common information display column 110 and a knowledge edit information display column 111 as shown in FIG. It is.
[0048]
FIG. 8 shows a case where the number of pieces of knowledge editing information is one. However, when a plurality of pieces of knowledge editing information are defined, the correlation can be confirmed by inputting the information in a table format. As a result, the operator can efficiently create the knowledge aggregation sheet.
[0049]
Next, the matching engine will be described. The matching engine 14 evaluates the knowledge aggregation sheet stored in the knowledge database 12 using the plant data stored in the plant database 11, searches for a condition that matches the current plant state, and compares the diagnosis result with the inference result database. 15 is output.
[0050]
Data necessary for evaluation of the knowledge aggregation sheet is obtained from the plant database 11 and converted into internal data used for collation.
[0051]
For analog data, a conversion graph as shown in FIG. 9 is created by dividing the upper and lower limits into three equal parts. Using the created conversion graph, the obtained analog data is converted into internal data from -3 to 3. For example, in FIG. 9, when the value of the analog data is A, the value of the internal data is 2.
[0052]
Since digital data is represented by only two values, ON and OFF, the value of the internal data is set to 3 when the acquired tag data is ON and to 0 when the acquired tag data is OFF.
[0053]
FIG. 10 shows a method for evaluating the knowledge aggregation sheet.
[0054]
(1) The internal data obtained and converted from the plant database is converted into a vector sequence.
[0055]
(2) The transmission degree defined at the time of inputting the alarm signal information on the knowledge aggregation sheet is converted into a vector sequence C.
[0056]
(3) An evaluation point is calculated by the inner product of the vector sequence of the internal data and the vector sequence of the transmission degree. In the case of this embodiment, the evaluation point is 16.
[0057]
The priority of the output of the collation result is determined by the following method.
[0058]
(1) Priority is given to the condition where the number of matches between the warning signal on the knowledge aggregation sheet and the warning transmission of the process data is large.
[0059]
(2) For each condition, the condition with the larger evaluation score calculated by the method described above is prioritized.
[0060]
(3) When the number of coincidences and the evaluation points of alarm transmissions are the same, the condition with the greater “severity of future transmission event” is given priority.
[0061]
(4) If the condition cannot be determined in (1), (2) and (3), the conditions are output in the order of description.
[0062]
【The invention's effect】
According to the present invention, the guidance for preventing the propagation of the state displacement is displayed, and the data related to the occurrence of the state displacement is displayed, so that the burden on the operator when the state displacement occurs is reduced.
[0063]
In addition, it is possible to standardize countermeasures when a state displacement occurs, and to pass on the operating knowledge of a skilled operator. Further, thereby, a plant operation support system can be realized.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a configuration of an embodiment of a plant operation support system according to the present invention.
FIG. 2 is a block diagram showing a configuration of an example of plant equipment to which the present invention is applied.
FIG. 3 is an explanatory diagram showing an example of a display screen configuration of a diagnosis result according to an embodiment of the present invention.
FIG. 4 is an explanatory diagram showing an example of a display screen configuration of a necessary measure among the diagnostic results according to the embodiment of the present invention.
FIG. 5 is an explanatory diagram showing an example of a display screen configuration of a ripple event among diagnosis results according to an embodiment of the present invention.
FIG. 6 is an explanatory diagram showing a display example of process data according to one embodiment of the present invention.
FIG. 7 is an explanatory diagram showing an example of a definition screen configuration of a knowledge aggregation sheet according to an embodiment of the present invention.
FIG. 8 is an explanatory diagram showing a display example in a table format of a knowledge aggregation sheet according to an embodiment of the present invention.
FIG. 9 is an explanatory diagram showing an example of a method of converting analog data according to an embodiment of the present invention.
FIG. 10 is an explanatory diagram illustrating an example of a method for evaluating a knowledge aggregation sheet according to an embodiment of the present invention.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 ... Plant operation support system, 2 ... Computer system, 3 ... Man-machine device, 4 ... Plant, 5 ... Instrumentation control system, 6 ... Other operation support systems, 10 ... Communication device, 11 ... Plant database, 12 ... Knowledge Database, 13 B & M (builder & maintenance) function, 14 arithmetic processing unit (collating engine), 15 inference result database, 20 display unit, 21 operation unit, 30 distillation column, 31 condenser, 32 reflux Tank, 33 ... reboiler, 40, 41, 42, 43 ... piping, 50 ... priority display field, 51 ... displacement factor display field, 52 ... necessary countermeasure display field, 53 ... priority at previous diagnosis, 54: a mark indicating a change in priority, 60: a detailed display column of necessary measures, 61: a check box, 70: a display column of a fault tree diagram, 80: an alarm is transmitted Process data, 90 ... Knowledge information definition screen, 91 ... Common information definition field, 92 ... Process name input area, 93 ... Group name input area, 94 ... Sheet name input area, 95 ... Execution cycle Input area, 96: screen number input area; 97, start mode input area at system startup; 100, knowledge editing information definition field; 101, apparatus information input area; 102, cause-related information input area; 103: an input area for alarm signal information; 104, an input area for information on propagation; 105, an input area for information on coping; 110, a display field for common information; 111, a display field for knowledge editing information.

Claims (5)

When a state displacement occurs in an operating plant, a plant operation support system including a computer system that evaluates the state displacement of the plant and predicts the spread of an event and outputs a coping method,
A plant database for storing process data collected during plant operation,
A knowledge database that presumes contents when a state displacement occurs in the plant and stores diagnosis contents such as target parts, event contents, abnormal factors, spillover events and necessary countermeasures in a knowledge aggregation sheet ,
A knowledge database storing the diagnostic contents, and an arithmetic processing unit that evaluates current data from plant data stored in the plant database and searches for diagnostic contents that match the current plant state,
An inference result database that stores a diagnosis result of the arithmetic processing device ;
In the arithmetic processing unit,
The upper and lower limits of the analog data of the plant are divided into three equal parts, the analog data is converted into internal data from -3 to 3, and a vector sequence of the internal data is obtained,
An evaluation point is calculated based on an inner product of the vector sequence of the acquired internal data and the vector sequence of the transmission degree defined at the time of inputting the alarm signal information of the knowledge aggregation sheet, and the alarm signal information of the knowledge aggregation sheet having the largest evaluation point is calculated. A plant operation support system, wherein priority processing is performed . The plant operation support system according to claim 1, wherein the arithmetic processing unit has a function of determining a priority level of a state displacement to be eliminated from a diagnosis result. 3. The plant operation support system according to claim 2, wherein the arithmetic processing unit has a function of comparing a current diagnosis result with a previous diagnosis result and detecting a change in a priority level of a state displacement. 4. The plant operation support system according to claim 1, wherein the arithmetic processing unit has a function of automatically selecting process data related to occurrence of a state displacement from a diagnosis result based on a knowledge database. 5. The plant operation support system according to any one of claims 1 to 4, further comprising a man-machine device having a display unit and an operation unit, wherein the man-machine device acquires a calculation result from an inference result database and controls plant operation. Plant operation support system including a display unit for displaying data for a plant.

Images