JP2010181949A - Gauge drift detection device and gauge drift detection method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a gauge drift detection device for reducing a workload in the case of selecting a drift detection object gauge or a drift determination model. <P>SOLUTION: This gauge drift detection device is provided with an HMI (Human Machine Interface) means 16 for outputting a display picture for presenting the list of drift determination object gauges, a display picture for presenting a drift determination model applicable to the drift determination of each gauge and a display picture for presenting a parameter applicable to a program for executing the drift determination model. Thus, a load is reduced in a work to extract a plurality of drift determination object gauges or a work to assign a proper drift determination model and parameter to each gauge. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a drift detection technique of a measuring instrument that is provided in a power plant, a chemical plant, or the like and measures a process amount of the plant.

Conventionally, as instrument drift detection technology,
(1) Calculate the deviation amount between the integrated amount or change amount of the process amount measured by a specific measuring instrument and the integrated amount or conversion amount of the process amount measured by a plurality of measuring instruments having a correlation with this measuring instrument. And what detects the drift of a measuring device using the calculated deviation amount is known (for example, refer to patent documents 1).

  (2) Also, obtain a statistical distribution of the process quantity measured by multiple instruments and a statistical distribution of the drift quantity of each instrument, and estimate the drift quantity of each instrument in light of these statistical distributions against probability theory. What is known is also known (see, for example, Patent Document 2).

Japanese Unexamined Patent Publication No. 7-63586 JP 2003-271231 A

  In the conventional measuring instrument drift detection technology, the drift detection of the measuring instrument is performed using the correlation of measuring instruments that measure the process quantities related to each other. For this reason, it is necessary to select a measuring instrument having a correlation when detecting drift of the measuring instrument, but this work burden increases as the number of measuring instruments provided in the plant increases. In particular, in a plant having thousands of measuring instruments such as a nuclear power plant, it is not easy to select correlated measuring instruments.

  Further, when an appropriate drift determination model is selected for each measuring instrument, the work load increases as the number of measuring instruments and the types of drift determination models increase. For example, a process quantity that is relatively important in a plant is often measured by a plurality of measuring instruments, and in this case, drift detection of the measuring instrument can be performed using a drift determination model that uses the correlation of the measuring instruments. it can. However, a process quantity having a relatively low importance is often measured by a single measuring instrument. In this case, it is necessary to detect drift of the measuring instrument using a drift determination model that does not use the correlation of the measuring instrument.

  The present invention has been made in view of the above circumstances, and an object thereof is to provide a measuring instrument drift detection device and a measuring instrument drift detection method that can reduce the work burden when selecting a target measuring instrument and a drift determination model for drift detection. And

  In order to solve the problems described above, in the measuring instrument drift detection device according to the present invention, process data receiving means for receiving each process data generated from a process amount of a plant measured by a plurality of measuring instruments, and the process data Process data recording means for recording process data received by the receiving means, drift determination method recording means for recording parameters used for a drift determination model and a program for executing the drift determination model as a drift determination method, and the process data recording Drift determination means for applying a drift determination technique recorded by the drift determination technique recording means to process data selected from the process data recorded by the means, and determining the drift of the measuring instrument; and the process data recording means Process data recorded by The drift determination model applicable to the process data and the parameters applicable to the program for executing the drift determination model are both displayed on the screen so that the selected drift determination model and parameters are stored in the drift determination method recording means. And a drift detection support means for recording.

In the instrument drift detection method according to the present invention,
(a) receiving each process data generated from the process amount of the plant measured by a plurality of measuring instruments, (b) recording the process data received in step (a), and (c) the above Presenting the process data recorded in step (b), a drift determination model applicable to the process data, and parameters applicable to a program for executing the drift determination model in a selectable manner, and (d) the step (c) And when a specific drift determination model and parameter are selected from the drift determination model and parameter presented in (1), the step of recording the selected drift determination model and parameter as a drift determination method, and (e) the step ( Apply the drift judgment method recorded in step (b) to the process data selected in c) , Characterized in that to execute the steps of performing drift determination instrument, to a computer.

  According to the present invention, it is possible to reduce the work load when selecting a drift detection target measuring instrument and a drift determination model.

The block diagram which shows the measuring device drift detection apparatus of 1st Embodiment. It is a figure which shows the recording format of the data in a process data recording means, (a) is a figure which shows a 1st recording structure, (b) is a figure which shows a 2nd recording format. It is a diagram showing a data recording structure in the drift determination technique recording means, (a) is a diagram showing a first recording structure, (b) is a diagram showing a second recording structure, (c) is a third recording structure Figure. The flowchart which shows the flow of the process performed by an HMI means. The figure which shows the basic screen which an HMI means outputs. The figure which shows the drift determination model selection screen which an HMI means outputs. The figure which shows the basic screen at the time of the drift determination execution which an HMI means outputs. The figure which shows the 2nd drift determination model selection screen which an HMI means outputs. The block diagram which shows the measuring device drift detection apparatus of 2nd Embodiment. The figure which shows a drift judgment model selection screen. The block diagram which shows the measuring device drift detection apparatus of 3rd Embodiment. The figure which shows the calculation concept of the correlation coefficient in a process data classification means. The figure which shows the state which displayed the similar process data display field on the drifted determination model selection screen. The block diagram which shows the measuring device drift detection apparatus of 4th Embodiment. The block diagram which shows the measuring device drift detection apparatus of 5th Embodiment. The figure which shows the recording format of the drift influence degree in a drift determination method recording means. The figure which shows the result of the drift progress prediction of the measuring device performed in 6th Embodiment.

  An embodiment of a measuring instrument drift detection device (measuring instrument drift detection method) according to the present invention will be described with reference to the accompanying drawings.

[First embodiment]
FIG. 1 is a block diagram showing a measuring instrument drift detection device 10 of the first embodiment.

  As shown in FIG. 1, the instrument drift detection apparatus 10 of the present embodiment includes a process data receiving unit 11, a process data recording unit 12, a drift determination method recording unit 13, a drift determination unit 14, and a drift determination method. Setting means 15 and HMI (Human Machine Interface) means 16 are provided.

  The process data receiving means 11 is configured to have a network interface, and periodically receives each process data generated from the process amount of the plant measured by a plurality of measuring instruments.

  The process data recording means 12 is constituted by a database management system provided with a large capacity recording device. The process data recording unit 12 records the process data received by the process data receiving unit 11. As the recording device, a hard disk device, a semiconductor recording device, a magneto-optical disk device, a magnetic tape device, or the like is used.

Data recorded by the process data recording means 12 is exemplified in the following (A1) to (A4).
(A1) Process ID: An identification name that uniquely identifies process data.
(A2) Process data name: Name given to process data.
(A3) Process data value: Process data generated from the process amount required for measurement, and functions of measurement start date / time and measurement end date / time of the process amount.
(A4) System name …… System name of the plant related to the process quantity.

  FIG. 2 is a diagram showing a recording structure of data (A1) to (A4) in the process data recording means 12, (a) is a diagram showing a first recording structure, and (b) is a diagram showing a second recording structure. is there.

  As shown in FIG. 2, two types of recording structures of data (A1) to (A4) in the process data recording means 12 are prepared.

  The first recording structure 100 includes a column 101 and a column 102 as shown in FIG. In each column, a process ID and a process data name are recorded.

  As shown in FIG. 2B, the second recording structure 200 includes a column 201, a column 202, a column 203, and a column 204. In each column, a process ID, a measurement start date / time of a process amount for generating process data, a measurement end date / time of the process amount, and a process data value are recorded. This process ID is the same process ID recorded in the first recording structure 100.

The process data recording means 12 has the following functions (B1) to (B4) in addition to the data (A1) to (A4) recording functions.
(B1) When a process ID, a process amount measurement time, and a process data value are given, these pieces of information are recorded.
(B2) When a process ID is given, the process data name corresponding to the process ID is output externally.
(B3) When a process ID and a process amount measurement time are given, a process data value corresponding to the process ID and the process amount measurement time is externally output.
(B4) When a process ID and a process amount measurement time are given, a process data value corresponding to the process ID and the process amount measurement time is output to the outside.

  The drift determination technique recording means 13 is constituted by a database management system, like the process data recording means 12.

  The drift determination method recording means 13 records each data of the drift determination model applied to the drift determination of the measuring instrument and the parameters input to the program for executing the drift determination model. Recording of data in the drift determination method recording unit 13 is performed by the drift determination method setting unit 15. Each data recorded in the drift determination method recording means 13 is used for drift determination in the drift determination means 14.

  The data recorded by the drift determination method recording means 13 is exemplified in the following (C1) to (C6).

  (C1) Process ID: An identification name for uniquely identifying process data, and the same process ID as that recorded in the process data recording means 12.

  (C2) Process data name: The name given to the process data and the same process data recorded in the process data recording means 12.

  (C3) System name: The system name of the plant related to the process quantity, and the same system name as that recorded in the process data recording means 12.

  (C4) Measuring instrument name: The name of the measuring instrument that measures process data, and the name of the measuring instrument that is subject to drift determination.

(C5) Drift determination model: A model for determining drift.
(i) “Threshold”: Drift by comparing the process data value generated from the process quantity measured by the measuring instrument with the lower and upper limit values set in advance for each process data value. A drift determination model for determining the presence or absence of
(ii) “Comparison with reference period” …… Drift determination model that determines the presence or absence of drift by comparing the process data value in the reference period and the process data value in the verification period using a statistical method. The reference period is a period arbitrarily selected from the process amount measurement periods. The verification period is a period in which the presence or absence of the drift of the measuring instrument is determined, and is a period arbitrarily selected from the process quantity measurement periods.
(iii) “Correlation”: A so-called correlation method, which is a drift determination model that determines the presence or absence of drift using correlation between a plurality of process data. For example, a drift determination model that calculates the correlation coefficient of each process data and determines the presence or absence of drift from the change in the correlation coefficient.

(C6) Parameter: This parameter is required by the program that executes the drift judgment model, and is prepared for each drift judgment model. Hereinafter, the term “parameter” simply refers to this parameter. The contents of the parameters are exemplified in the following (i) to (iii). (i) Parameters related to the drift judgment model “threshold”:
・ Identification information of drift judgment model ・ Threshold ・ Process data sampling period and sampling period
(ii) Parameters related to drift judgment model “comparison with reference period”:
・ Significance level ・ Base period ・ Test period
(iii) Parameters related to the drift judgment model “correlation”:
・ Drift judgment model identification information ・ Process data ID to identify correlated process data
-Sampling cycle and sampling period of reference process data-Sampling cycle and sampling period of process data to be verified

  FIG. 3 is a diagram showing a data recording structure in the drift determination method recording means 13, wherein (a) shows a first recording structure, (b) shows a second recording structure, and (c) shows a third recording structure. It is a figure which shows a recording structure.

  As shown in FIG. 3, three types of recording structures of data (C1) to (C6) in the drift determination method recording means 13 are prepared.

  As shown in FIG. 3A, the first recording structure 300 includes a column 301, a column 302, a column 303, a column 304, a column 305, a column 306, and a column 307. In each column, a process ID, a process data name, a system name, a measuring instrument name, a group number, a drift determination model, and a parameter are recorded. The group number is a common number given to process data related to each other.

  As shown in FIG. 3B, the second recording structure 400 includes a column 401, a column 402, and a column 403. In each column, a group number, a drift determination model, and parameters are recorded. The group number is the same group number recorded in the first recording structure 300.

  As shown in FIG. 3C, the third recording structure 500 includes a column 501 and a column 502. In each column, a drift determination model and a code entity of a program that executes the drift determination model are recorded. The drift determination model is the same drift determination model as that recorded in the first recording structure 300.

  The relationship between the first recording structure 300 to the third recording structure 500 will be described. For example, when “correlation” is selected as the drift determination model, the process data having process IDs LT01A, LT01B, and LT01C are correlated with each other as shown in FIG. The group number “1” is recorded. The group number “2” is recorded on the same basis. For process data of group numbers “1” and “2”, a second recording structure 400 is prepared. Each process data is recorded in a state where a common drift determination model “3” is assigned in the second recording structure 400, as shown in FIG. In this example, the drift determination model “3” is assigned the drift determination model “correlation”.

  On the other hand, the process data whose process IDs are FT01, FT02, and PT01 have no correlation with each other, so that the group number “NULL” is recorded in the column 305. A third recording structure 500 is prepared for the process data of the group number “NULL”. As shown in FIG. 3C, each process data is recorded in a state where a code entity of a program that executes a drift determination model selected by an operator of the instrument drift detection apparatus 10 is assigned.

  That is, the drift determination method recording means 13 separately records process data having correlation with each other and process data having no correlation with each other, and maintains correspondence between each process data and the drift determination model applied to the process data. Record in the correct state.

  The drift determination unit 14 is configured as a program constructed by a server-side script of a Web server.

  The drift determination means 14 performs drift determination on the target measuring instrument selected by the operator as the target of drift determination. This drift determination is performed by applying the drift determination model and parameters recorded in the drift determination method recording means 13 to the process data selected by the operator among the process data recorded in the process data recording means 12.

  In the drift determination means 14, the correspondence between the process data and the measuring instrument that has measured the process data, the drift determination model applied to each process data, and the parameters used in this drift determination model are the drift determination method recording means 13. Read from the recording structure.

  The drift determination technique setting means 15 is configured as a program described by a server-side script of a Web server.

  This drift determination method setting means 15 has mainly two functions. One is a function of extracting process data from the process data recording unit 12 in response to a request from the HMI unit 16 and transmitting the extracted process data to the HMI unit 16. As will be described later, the HMI means 16 displays the process data received from the drift determination method setting means 15 on the display unit so that the operator can easily select the process data used for the drift determination.

  The other is a function of receiving the data (C1) to (C6) described above from the HMI means 16 and recording these data in the drift determination technique recording means 13.

  The HMI means 16 is constituted by a personal computer equipped with a Web browser, and includes a display unit, a recording unit, and an input operation unit (all not shown). Note that a liquid crystal display or a CRT display is used as the display unit, and a pointing device such as a keyboard, a mouse, or a track pad is used as the input operation unit. The display unit is used for output of a selection screen of a drift determination model and parameters applied to drift determination. The input operation unit is provided for the operator to select each data according to the selection screen output to the display unit.

  The drift detection assisting means 21 includes a human-machine interface means 16 for selectively displaying parameters applicable to a program for executing the selected drift determination model, and a drift determination for the selected specific drift determination model and parameters. And a drift determination technique setting 15 to be recorded in the technique recording means 13.

  The drift detection support means 21 is a screen that allows selection of any of process data recorded by the process data recording means 12, a drift determination model applicable to the process data, and a parameter applicable to a program for executing the drift determination model. It has a function of displaying and selecting the selected drift determination model and parameters in the drift determination method recording means 13.

  FIG. 4 is a flowchart showing the flow of processing executed by the HMI means 16. FIG. 5 is a diagram showing a basic screen 20 output by the HMI means 16. FIG. 6 is a diagram showing a first drift determination model selection screen 30 output by the HMI means 16. FIG. 7 is a diagram showing a basic screen 20a at the time of execution of drift determination output from the HMI means 16. FIG. 8 is a diagram showing a second drift determination model selection screen 30a output from the HMI means 16. Hereinafter, each step in the flowchart shown in FIG. 4 will be described with reference to FIGS.

  Step S1 is a step of setting an initial value. In step S 1, the process data value, the process data name, and the system name acquired from the process data recording unit 12 by the drift determination method setting unit 15 are acquired from the determination method setting unit 15.

  Step S2 is a step of updating the display screen on the display unit of the HMI means 16 following Step S1, and displays the basic screen 20 shown in FIG. 5 when the measuring instrument drift detection device 10 is activated. In the process data selection field 22 of the basic screen 20, a list of process data acquired in step S1 is displayed. This list includes process data names and system names.

  Step S3 is a step following step S2 in which the operator waits for a process data selection request. The operator can request the HMI unit 16 to select desired process data by operating the input operation unit of the HMI unit 16. The same applies to the case where each HMI means 16 is requested to select each data in each subsequent step.

  Step S4 is a step of determining whether or not there is a process data selection request by the operator following step S3.

  Step S5 is a step performed when <Yes> in step S3, and is a step of selecting the process data that has received the selection request. And it transfers to step S2 and the drift determination model selection screen 30 shown in FIG. 6 is displayed.

  In step S5, the trend of the process data that has received the selection request is displayed in the trend display field 21 of the basic screen 20. While the process data is not requested to be selected, the trend display field 21 is blank. FIG. 5 is an example in which process data identified by “FT-0002 transfer pump flow rate” is selected. The process data selection field 22 highlights the item of the process data requested to be selected.

  Step S6 is a step performed when <No> in step S3, and is a step of determining whether or not there is a scroll request in the process data selection field 22.

  Step S7 is a step performed when <Yes> is determined in step S6, and is a step of scrolling the display area of the process data selection field 22 according to the position of the scroll bar 23 moved by the operator.

  Step S8 is a step performed when <No> is determined in step S6, and is a step of determining whether or not there is a request for selecting a drift determination model.

  Step S9 is a step that is performed when <Yes> is determined in step S8, and is a step of selecting the drift determination model that has received the selection request. Here, the operator can request the HMI means 16 to select a drift determination model according to the display of the drift determination model selection screen 30 shown in FIG. The drift determination model displayed on the drift determination model selection screen 30 is a drift determination model applicable to the process data selected in step 5.

  The drift determination model selection screen 30 mainly includes a drift determination model selection field 31 and a parameter selection field 32. The OK button 33 determines the current selection request, and the cancel button 34 is selected when canceling the current selection request. The end button F is selected when canceling the current screen display.

  Of the drift determination models displayed in the drift determination model selection field 31, for example, when “comparison with the reference period” is requested to be selected, the color tone of the check box on the left side of the item is changed. At the same time, a parameter selection field 32 for selecting a parameter required by the program that executes the drift determination model “comparison with reference period” that has received the selection request is displayed.

  In the displayed parameter selection field 32, for example, the operator can request to select the parameter “significance level”. When the OK button 33 is selected by the operator, the process proceeds to step S2, and the basic screen 20 shown in FIG. 5 is displayed. In the process data selection field 22 of the basic screen 20 at this time, the drift determination model “comparison with the reference period” received by the operator is displayed.

  Step S10 is a step performed when <No> is determined in step S8, and is a step of determining whether or not the operator has requested to select “start date and time”. The “starting date / time” means the starting date / time of the reference period necessary for the drift determination model “comparison with the reference period”. The operator can make a selection request for “starting date and time” by moving the starting point bar 24a.

  Step S11 is a step performed when <Yes> is determined in step S8, and is a step of changing the “starting point date / time” according to the position of the starting point bar 24a moved by the operator. In addition, the “start date / time” corresponding to the position of the start point bar 24 a is numerically displayed in the start point date / time display field 26 a of the basic screen 20. Further, the regression line formula and standard deviation of the process data at the interval between the currently selected “start date and time” and “end date and time” are calculated and displayed in the regression line formula display field 28.

  Steps S12 and S13 are steps related to the selection of “end point date and time”, and are the same steps as steps S10 and S11, and thus description thereof is omitted. The “end point date and time” means the end point date and time of the reference period used in the drift determination model “comparison with reference period”.

  When the selection of the “end point date / time” of the reference period is completed, the HMI means 16 starts and displays the start point bar 24b for selecting the “start point date / time” of the test period used for drift determination on the basic screen 20a, as shown in FIG. The end point bar 25b for selecting the “end date and time” of the period is displayed. The steps related to the selection of the “start date and time” and “end date and time” for the test period are the same as those in steps S10 to S13, and the description thereof will be omitted.

  Step S14 is a step performed when <No> is determined in step S12, and the currently selected drift determination model “comparison with reference period”, parameters “significance level”, “start date and time”, and “end point” This is the step of ending the “date and time” selection process.

  Step S15 is a step of transmitting the data selected in each step to the drift determination method setting means 15. The drift determination method setting means 15 records each received data in the drift determination method recording means 13. The drift determination unit 14 performs drift determination of the target measuring device using each data recorded in the drift determination method recording unit 13.

  When step S15 is completed, the result of the drift determination is displayed in the determination result display field 29 of the basic screen 20a (see FIG. 7).

  FIG. 8 shows a drift determination model selection screen 30a when the drift determination model “threshold” is selected in step S9. When the drift determination model “threshold value” is selected, the “upper limit value” and “lower limit value” of the threshold value are displayed in the parameter selection field 32 of the drift determination model selection screen 30a.

(Function)
In the instrument drift detection apparatus 10 of the present embodiment, two data flows are formed. One data flow relates to process data. This process data flows along the path of process data receiving means 11 → process data recording means 12 → drift determining means 14, and is used for drift determination in the drift determining means 14. The process data used for this drift determination is process data related to the target measuring instrument selected by the operator among the process data recorded in the process data recording means 12. When “correlation” is used as the drift determination model, other correlated process data is included, but this process data is also selected by the operator. The process data is selected according to the basic screen 20 (see FIG. 5) output by the HMI means 16.

  Another data flow relates to the drift determination model and parameters required by the program that executes the drift determination model. This drift determination model and parameters flow through a route of HMI means 16 → drift determination method setting means 15 → drift determination method recording means 13 → drift determination means 14 and are used for drift determination in the drift determination means 14. The drift determination model and parameters used for this drift determination are the drift determination model and parameters selected by the operator. The selection of the drift determination model and parameters is performed according to the drift determination model selection screen 30 (see FIG. 6) on the display unit of the HMI means 16.

  Here, the basic screen 20 is first output to the display unit of the HMI means 16. In the process data selection field 22 of the basic screen 20, all the process data recorded by the process data recording means 12 are displayed together with the identification information. The operator can select process data related to the target measuring instrument to be subjected to drift determination according to the basic screen 20.

  When specific process data is selected in the process data selection field 22, a drift determination model selection screen 30 is displayed on the display unit of the HMI means 16. In the drift determination model selection field 31 in this screen, a drift determination model applicable to the drift determination of the measuring instrument related to the process data selected by the operator is displayed. The operator can select a drift determination model according to the drift determination model selection screen 30.

  When a specific drift determination model is selected in the drift determination model selection field 31, a parameter selection field 32 is displayed on the drift determination model selection screen 30. In the parameter selection field 32 in this screen, parameters that can be input to the program that executes the drift determination model selected by the operator are displayed. The operator can select necessary parameters according to the drift determination model selection screen 30.

  When specific process data is selected in the process data selection field 22, the trend of the process data selected by the operator is displayed in the trend display field 21 of the basic screen 20. For this reason, for example, when the drift determination model “comparison with reference value” is selected, the reference period and the test period can be set while visually checking the trend of the process data.

(effect)
As a result, the measuring instrument drift detection device 10 can reduce the work burden when selecting a target measuring instrument or drift determination model for input drift detection.

  Further, the process data recorded by the process data recording means 12 is displayed on the screen so as to be selectable together with the process identification information. Next, on the condition that specific process data is selected from the process data displayed on the screen, a drift determination model applicable to the drift determination of the measuring instrument related to the selected process data is displayed on the screen. . Then, on the condition that a specific drift determination model is selected from the drift determination models displayed on the screen, parameters applicable to the program that executes the selected drift determination model are displayed on the screen. That is, since the operator can detect drift according to the display screen output by the HMI means 16, it is possible to reduce erroneous input of data necessary for drift determination.

[Second Embodiment]
The second embodiment is an example in which a process data correlation evaluation unit 17A is added to the configuration of the instrument drift detection device 10 of the first embodiment. In addition, the same structure as 1st Embodiment attaches | subjects the same code | symbol to a corresponding structure, abbreviate | omits description, The structure which changed the structure of 1st Embodiment, or was newly added attaches "A" to the code | symbol end. To explain.

  FIG. 9 is a block diagram showing a system configuration of the instrument drift detection apparatus 10A of the second embodiment. FIG. 10 is a diagram showing a drift determination model selection screen 30A output by the HMI means 16.

  As shown in FIG. 9, the instrument drift detection apparatus 10A of the present embodiment includes a process data correlation evaluation unit 17A. The process data correlation evaluation unit 17A performs correlation evaluation between the process data recorded in the process data recording unit 12A, and creates a list of process data having high correlation. The correlation evaluation between the process data is performed using, for example, a neural network.

  The HMI unit 16 refers to the result of the correlation evaluation performed by the process data correlation evaluation unit 17A, and displays the process data whose degree of correlation exceeds the set standard on the screen.

(Function)
When the drift determination model “correlation” is selected in step S9 in the flowchart of FIG. 4, a correlation display field 35A is displayed on the drift determination model selection screen 30A as shown in FIG. In the correlation display field 35A, process data relating to a measuring instrument whose correlation coefficient with the target measuring instrument exceeds a set standard is displayed. FIG. 10 shows an example in which 0.95 is set as the setting reference. Since other operations are the same as those of the first embodiment, description thereof is omitted.

(effect)
As a result, when the correlation method is used to determine the drift of the target measuring instrument using the correlation with other measuring instruments as the drift determination model, the workload of selecting a measuring instrument with a high correlation with the target measuring instrument can be reduced. .

[Third embodiment]
The third embodiment is an example in which process data classification means 18B is added to the configuration of the instrument drift detection device 10 of the first embodiment. In addition, the same structure as 1st Embodiment attaches | subjects the same code | symbol to a corresponding structure, abbreviate | omits description, and the structure which changed the structure of 1st Embodiment or added newly is "B" at the code | symbol end. Will be described.

  FIG. 11 is a block diagram showing a system configuration of the instrument drift detection device 10B of the third embodiment. FIG. 12 is a diagram showing a calculation concept of the correlation coefficient in the process data classification unit 18B. FIG. 13 is a diagram showing a state in which the similar process data display field 36B is displayed on the drift determination model selection screen 30B.

  As shown in FIG. 11, the instrument drift detection apparatus 10B of the present embodiment includes process data classification means 18A.

  As shown in FIG. 12, the process data classification means 18B first classifies the process data into several groups according to the type of measuring instrument. FIG. 12 exemplifies process data classified into flow meter groups.

  Next, the correlation coefficient of the process data belonging to the same group is calculated. FIG. 12 shows the process data identified by the “FT-0002 transfer pump flow rate” measured by the target flow meter to be subjected to drift determination, and the “FT-0003 transfer water flow rate” measured by another measuring instrument, Correlation coefficients with each process data identified by “FT-0008 heat exchange inlet flow rate” and “FT-0009 heat exchange outlet flow rate” are “0.98”, “0.85”, and “0.84”, respectively. ”Is an example calculated.

  Finally, process data having a correlation coefficient equal to or greater than a set standard is determined as correlated process data. When the setting standard is set to “0.8”, the process data identified by “FT-0002 transfer pump flow rate” is “FT-0003 transfer water flow rate”, “FT-0008 heat inlet flow rate” and “ It is determined that there is a correlation with all of the process data identified by “FT-0009 Heat exchange outlet flow rate”.

  In the present embodiment, when specific process data is selected from the process data displayed on the screen in step S5 in the flowchart shown in FIG. 4, the HMI means 16 displays a similar process data display field 36B on the drift determination model selection screen 30B. indicate. Then, the HMI means 16 refers to the result of the correlation evaluation performed by the process data classification means 18B, and sets the process data belonging to the same classification as the classification to which the selected process data belongs to the correlation with the selected process data. The number is displayed in the similar process data display field 36B together with the number.

(Function)
The operator can confirm the display of the similar process data display field 36B and know the process data having a strong correlation with the selected process data. An appropriate drift determination model can be used based on the strength of the correlation in determining the drift of the measuring instrument related to the selected process data.

  Figure 13 assigns the drift judgment model "correlation" to the process data identified by "FT-0002 transfer pump flow rate" and identifies by "FT-0008 heat inlet flow rate" and "FT-0009 heat exchange outlet flow rate" This is an example in which a drift determination model “threshold value” is assigned to the processed process data. Since other operations are the same as those of the first embodiment, description thereof is omitted.

(effect)
As a result, the work burden of assigning an appropriate drift determination model for each measuring instrument can be reduced.

[Fourth embodiment]
The fourth embodiment is an example in which a connection information recording unit 19C is added to the configuration of the measuring instrument drift detection device 10A of the second embodiment. In addition, the same structure as 2nd Embodiment attaches | subjects the same code | symbol to a corresponding structure, abbreviate | omits description, and the structure which changed the structure of 2nd Embodiment or added newly is "C" at the code | symbol end. Will be described.

  FIG. 14 is a block diagram showing a system configuration of a measuring instrument drift detection apparatus 10C of the fourth embodiment.

  As shown in FIG. 14, the instrument drift detection device 10 </ b> C of this embodiment includes a connection information recording unit 19 </ b> C.

The connection information recording unit 19 </ b> C records connection information of equipment and piping used for plant process execution. The connection information recorded by the connection information recording means 19C is exemplified in the following (D1) to (D4).
(D1) Connection information for equipment and piping used for process execution involving fluid movement.
(D2) Equipment / piping connection information used for process execution involving object rotation and vibration.
(D3) Device / wiring connection information for process execution with current change.
(D4) Equipment / piping arrangement and connection information for process execution involving heat transfer.
The HMI means 16 displays device / piping connection information (D1) to (D4) on the screen in response to the operator's request.

(effect)
As a result, it is possible to reduce the work burden of estimating the correlation strength of the measuring instrument based on the operator's experience. That is, the work burden of assigning an appropriate drift determination model to the target measuring instrument is reduced.

[Fifth Embodiment]
The fifth embodiment is an example in which a drift influence evaluation unit 20D is added to the configuration of the instrument drift detection device 10 of the first embodiment. In addition, the same structure as 1st Embodiment attaches | subjects the same code | symbol to a corresponding structure, abbreviate | omits description, and the structure which changed the structure of 1st Embodiment or added newly is "D" at the code | symbol end. Will be described.

  FIG. 15 is a block diagram showing a system configuration of an instrument drift detection device 10D of the fifth embodiment. FIG. 16 shows the information recorded by the drift determination method recording means 13 and shows the influence of the drift of the measuring instrument on the plant operation.

  As shown in FIG. 15, the instrument drift detection device 10 </ b> D of this embodiment includes a drift influence evaluation unit 20 </ b> D.

  The drift influence evaluation means 20D evaluates the influence that the drift of the target measuring instrument determined by the drift determination means 14 has on the operation of the plant.

  The drift determination technique recording means 13 has a drift influence degree. The drift influence degree recording structure 600 includes a column 601, a column 602, a column 603, and a column 604, as shown in FIG. In each column, a process ID, a + drift importance level, a -drift importance level, and a location URL of related information are recorded in advance. The process ID is the same process ID as in the first embodiment, and is an identification name that uniquely identifies process data.

  The + drift importance indicates the importance of the influence of the drift on the operation of the plant when the measuring instrument drifts in the direction of measuring a process amount larger than the true process amount. -The drift importance indicates the importance of the influence of the drift on the operation of the plant when the measuring instrument drifts in a direction indicating a process amount smaller than the true process amount. The related information is detailed information regarding the effect of instrument drift on plant operation.

The present embodiment is an example in which the drift importance is divided into three levels, and the contents of each level are determined as follows.
Level 1: The impact on the plant is large, and continuation of plant operation is not permitted.
Level 2: Although there is an impact on the plant, the plant operation can be continued by taking appropriate measures.
Level 3: There is almost no impact on the plant.

  According to the example illustrated in FIG. 16, the drift influence evaluation unit 20D evaluates the drift influence degree as “level 2” if the measuring instrument related to the process data identified by the process ID “FT01” drifts to the plus side. That is, although it has an influence on the plant, it is evaluated that the plant operation can be continued by taking an appropriate measure. On the other hand, if this measuring instrument drifts to the minus side, the drift influence degree is evaluated as “level 1”. That is, it is evaluated that there is almost no influence on the plant.

  Then, the HMI means 16 displays on the screen the result of the drift influence evaluation performed by the drift influence evaluation means 20D in response to the operator's request. The mode of screen display is not particularly limited. For example, a button for displaying the result of the drift influence evaluation is displayed on the basic screen 20a (see FIG. 7) at the time of executing the drift determination, and the result of the drift influence evaluation is displayed at an appropriate position on the basic screen 20 when this button is selected. You may make it display. At this time, related information regarding the influence of the drift of the measuring instrument on the operation of the plant or the location URL of this related information may be displayed.

(effect)
The operator can grasp the influence of the drift of the target measuring instrument on the plant operation after executing the drift determination of the target measuring instrument.

[Sixth Embodiment]
The sixth embodiment is an example in which a drift progress prediction function is added to the instrument drift detection device 10 of the first embodiment. In addition, since the structure of the measuring device drift detection apparatus of this embodiment is the same as that of 1st Embodiment, description is abbreviate | omitted.

  The HMI means 16 in the instrument drift detection apparatus 10 (see FIG. 1) of the present embodiment has a drift progress prediction model. The drift progress prediction model regressively analyzes the past process volume measurement history by the least square method, and predicts that the confidence interval, that is, the interval where the true value of the future drift is entered, and the confidence interval exceeds the drift tolerance. It is to calculate the time to be done.

  In the instrument drift detection device 10 of the present embodiment, the operator can select a drift progress prediction method from among the drift determination models displayed on the screen by the HMI means 16. When the drift progress prediction method is selected by the operator, the drift determination method setting means 15 stores the drift progress prediction method and the parameters for prediction input to the program for executing the drift progress prediction method. 13 is recorded. Then, the drift determination unit 14 predicts the drift progress of the measuring instrument using the drift progress prediction method and the prediction parameters recorded in the drift determination method recording unit 13.

(Function)
In the measuring instrument drift detection apparatus 10 of the present embodiment, after the drift determination unit 14 determines the drift of the measuring instrument, the future drift progress is predicted for the measuring instrument for which the drift is determined. Since other operations are the same as those of the first embodiment, description thereof is omitted.

  FIG. 17 is a diagram illustrating a result of prediction of drift progress of the measuring instrument.

  FIG. 17 is a graph in which confidence intervals are plotted with time as the horizontal axis. The horizontal axis shows the period from the past 6 months to the future 6 months centering on the present. The vertical axis indicates the drift amount of the measuring instrument. The solid line La in the figure indicates the upper limit value of the confidence interval, and the solid line Lb indicates the lower limit value of the confidence interval. When the drift progress prediction of the measuring instrument is performed, the graph shown in FIG. 17 is output. At this time, when the allowable range of the drift amount of the measuring instrument is set to -2% to + 2%, the operator can predict that the upper limit of the confidence interval will exceed the allowable range after about three months.

(effect)
As a result, the operator can take appropriate measures before the drift develops greatly, and the operation of the plant can be maintained healthy.

  As mentioned above, although the measuring instrument drift detection apparatus based on this invention has been demonstrated based on 1st Embodiment-6th Embodiment, about a specific structure, it is not restricted to these embodiments, Claims Modifications and additions of the design are permitted without departing from the spirit of the invention according to the claims.

  For example, the drift determination technique recording unit 13 and the process data recording unit 12 may be configured by one database management system.

  The drift determination unit 14 may be configured as one application of the HMI unit 16.

  The drift determination method setting means 15 may be configured as an application program for a personal computer. Alternatively, when the HMI means 16 is configured by a personal computer, the drift determination method setting means 15 may be configured as one application program in the HMI means 16.

10, 10A, 10B, 10C, 10D Measuring instrument drift detector 11 Process data receiving means 12 Process data recording means 13 Drift determination technique recording means 14 Drift determination means 15 Drift determination technique setting means 16 HMI (Human Machine Interface) means 17A Process data correlation evaluation means 18B Process data classification means 19C Connection information recording means 20D Drift influence evaluation means 21 Drift detection support means

Claims (13)

Process data receiving means for receiving each process data generated from a process amount of a plant measured by a plurality of measuring instruments;
Process data recording means for recording process data received by the process data receiving means;
A drift determination method recording means for recording a drift determination model and a parameter used in a program for executing the drift determination model as a drift determination method;
Drift determination means for applying a drift determination method recorded by the drift determination method recording means to process data selected from the process data recorded by the process data recording means, and performing drift determination of the measuring instrument;
The process data recorded by the process data recording means, the drift determination model applicable to the process data, and the parameters applicable to the program for executing the drift determination model are both displayed on the screen and the selected drift determination is possible. Drift detection support means for recording the model and parameters in the drift determination technique recording means;
An instrument drift detection device characterized by comprising: The drift detection support means is
The process data recorded by the process data recording means is displayed on the screen so as to be selectable together with the process identification information, and is related to the selected process data on condition that specific process data is selected from the process data displayed on the screen. The drift judgment model applicable to the drift judgment of the instrument is displayed on the screen, and the selected drift judgment model is selected on the condition that a specific drift judgment model is selected from the drift judgment models displayed on the screen. Human-machine interface means for selectively displaying parameters applicable to the program to be executed on the screen;
Drift determination method setting means for recording the selected specific drift determination model and parameters in the drift determination method recording means;
The instrument drift detection device according to claim 1, wherein: A process data correlation evaluation unit for performing correlation evaluation of the process data recorded by the process data recording unit;
When the correlation detection method is selected from the drift determination models displayed on the screen, the drift detection support means refers to the result of the correlation evaluation performed by the process data correlation evaluation means, and The instrument drift detection device according to claim 1, wherein process data having a correlation degree exceeding a set criterion is displayed on a screen so as to be selectable. The process data recorded by the process data recording means is classified according to the type of measuring instrument that measures the process data, and includes process data classification means for performing correlation evaluation of process data belonging to each classification,
When specific process data is selected from the process data displayed on the screen, the drift detection support means refers to the result of correlation evaluation performed by the process data classification means, and the selected process data belongs to 2. The instrument drift detection device according to claim 1, wherein process data belonging to the same classification as the classification is displayed on a screen so as to be selectable together with the degree of correlation with the selected process data. Connection information recording means for recording connection information of equipment and piping used for process execution of the plant,
2. The measuring instrument drift detection device according to claim 1, wherein the drift detection support means displays the connection information of the device / pipe on a screen in response to an operator's request. A drift influence evaluating means for evaluating the influence of the drift of the measuring instrument determined by the drift determining means on the operation of the plant,
The instrument drift detection device according to claim 1, wherein the drift detection support means displays the result of the influence evaluation performed by the drift influence evaluation means on the screen in response to an operator's request.   The instrument drift detection apparatus according to claim 1, wherein the drift determination model includes a drift progress prediction method that predicts drift progress of the instrument using a process data measurement history. (a) receiving each process data generated from the process quantity of the plant measured by a plurality of measuring instruments;
(b) recording the process data received in step (a);
(c) The process data recorded in step (b), the drift determination model applicable to the process data, and the parameters applicable to the program for executing the drift determination model are presented in a selectable manner;
(d) recording a selected drift determination model and parameter as a drift determination method when a specific drift determination model and parameter are selected from the drift determination model and parameter presented in step (c); ,
(e) applying the drift determination technique recorded in step (b) to the process data selected in step (c), and performing a drift determination of the instrument;
A drift detection method for a measuring instrument characterized by causing a computer to execute. In step (c),
The process data recorded in the step (b) is presented together with the process identification information, and on the condition that the specific process data is selected from the presented process data, the drift determination of the instrument related to the selected process data is performed. The applicable drift judgment model is presented, and the parameters applicable to the program that executes the selected drift judgment model are presented on condition that a specific drift judgment model is selected from the presented drift judgment models. The instrument drift detection method according to claim 8, wherein: (f) causing the computer to perform a correlation evaluation of the process data recorded in step (b),
In step (c), when a correlation method is selected from the drift determination model presented in this step, the result of the correlation evaluation performed in step (f) is referred to, and the selected process data and 9. The instrument drift detection method according to claim 8, wherein process data whose correlation degree exceeds a set criterion is presented in a selectable manner. (g) The process data recorded in step (b) is classified according to the type of measuring instrument that measured the process data, and the computer is caused to perform a step of performing correlation evaluation of the process data belonging to each classification,
In step (c), when specific process data is selected from the process data presented in this step, the result of the correlation evaluation performed in step (g) is referred to, and the selected process data 9. The instrument drift detection method according to claim 8, wherein process data belonging to the same classification as the classification to which the data belongs is presented together with a degree of correlation with the selected process data. (h) Let the computer execute the step of recording the connection information of the equipment and piping used for the plant process execution,
9. The instrument drift detection method according to claim 8, wherein in the step (c), connection information of the equipment / pipe recorded in the step (h) is presented. (i) causing the computer to execute a step of evaluating the influence of the drift of the measuring instrument determined in the step (e) on the plant operation,
9. The measuring instrument drift detection method according to claim 8, wherein the step (c) presents the result of the influence evaluation performed in the step (i).

Images