JP2009175870A - Measurement instrument drift detection device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a measurement instrument drift detection device which has high general-purpose property for preventing the deterioration of the estimating precision of the true value of process data and the reliability of drift detection due to the correlation of the process data, and for making it unnecessary to prepare any drift decision index corresponding to the classification of the measuring instrument. <P>SOLUTION: The measurement instrument drift detection device 10 is configured to detect the drift of a measurement instrument by using process data to be measured by using a plurality of measuring instruments in which correlation is recognized in the process data as a target, and provided with a change detection means 101 for deciding the change of process data for each of the process data, and for obtaining the change as the detection result; and a drift decision means 102 for selecting one process data from the process data whose detection result has been shown, and for, when the detection results of the residual process data are matched with each other, deciding whether or not the detection result is matched with the detection result of one selected process data, and for deciding the presence of drift to the measuring instrument which has measured one selected process data under such conditions that the both detection results are not matched with each other. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a measuring instrument drift detection device that is provided in a plant facility and that detects a drift of a measuring instrument using a plurality of measuring instruments that are correlated with measured process data.

  Generally, in plant facilities such as a power plant, many types of measuring instruments such as pressure gauges, flow meters, thermometers and the like are often installed to ensure normal operation of the plant. These measuring instruments usually drift, and in order to measure the process data with the required accuracy, it is necessary to inspect and calibrate as necessary.

  If the inspection of the measuring instrument is performed based on the concept of time monitoring maintenance, it is important to set the inspection interval short with a sufficient margin in consideration that the drift of the measuring instrument does not progress in proportion to time. It becomes. For this reason, if maintenance is emphasized so that the measuring instrument is within the required accuracy range, the cost associated with the increase in the number of inspections, the risk of decreased plant safety due to human error during inspection, and the reduction in plant operating rate will be reduced. Will be invited.

  Conventionally, there has been known a measuring instrument drift detection device that determines such a problem without measuring the drift of the measuring instrument without inspecting and notifies that the measuring instrument has approached the required accuracy limit. For example, there is known one that estimates the drift amount by statistically calculating the most reliable estimated true value using a neural network or a linear model (see, for example, Patent Documents 1 and 2).

  It is also known to detect drift for this specific instrument using the correlation between process data measured by a specific instrument and process data measured by other instruments. (For example, see Patent Documents 3 and 4).

In addition, the drift of the measuring instrument is detected using a rule matrix created in advance (for example, see Patent Document 5), or the actual value for the true value when drift occurs using past process data for each measuring instrument. Known are those that detect the drift of a measuring instrument using a drift determination index preset for each measuring instrument, such as those that detect the drift of a measuring instrument using a probability density function of values (see, for example, Patent Document 6). It has been.
JP 2001-356818 A JP 2007-3381 A Special table 2002-509324 gazette JP-A-60-91289 Japanese Patent Laid-Open No. 7-63586 JP 2003-271231 A

  In the conventional instrument drift detection device, the accuracy of the true value estimation model is a factor that reduces the reliability of drift detection. For example, letting a neural network learn data is not simple. Especially assuming that the operation output of plant equipment changes in the middle, it becomes difficult to apply the neural network, and the accuracy of true value estimation It is not easy to secure enough.

  In addition, when using the correlation of process data measured at the time of drift detection, it is important to accurately model the correlation, and the accuracy of the correlation model or the strength of the correlation is a factor that reduces the reliability of drift detection. Become.

  Further, when a preset drift determination index is used, the reliability of drift detection depends on the accuracy of the index, and a great amount of labor is required to prepare a highly accurate index for each measuring instrument.

  The present invention has been made in view of the above problems, and avoids a decrease in the reliability of drift detection due to the accuracy of true value estimation of process data or the correlation of process data, and in addition, drift determination according to the type of measuring instrument. An object is to provide a highly versatile measuring instrument drift detection device that does not require an index.

  In order to solve the above-described problems, the present invention targets a plurality of measuring instruments that are provided in a plant facility and are correlated with measured process data, and detects drift of the measuring instrument using the process data. In the measuring instrument drift detection device, for each of the process data, a change in the process data is determined with the presence or absence of an increase or a decrease in the process data as a content, and a change detection using the change in the process data as a detection result Change detection means for performing processing, receiving the detection result, selecting one process data from the process data indicating the detection result, and determining whether the detection results of the remaining process data not selected match each other When this match is determined, the detection result and the detection result of the selected one process data are detected. Drift determination means for performing a drift determination process for determining that there is a drift with respect to the measuring instrument that has measured the one selected process data on the condition that the two detection results do not match And.

  According to the measuring instrument drift detection device of the present invention, it is possible to avoid a decrease in the reliability of drift detection due to the accuracy of estimation of the true value of process data or the correlation of process data, and in addition, a drift determination index corresponding to the type of measuring instrument. Versatility without the need for

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

[First Embodiment]
First, the structure of the measuring instrument drift detection apparatus 10 of 1st Embodiment is demonstrated.

  FIG. 1 is a functional block diagram of an information processing system 100 in the instrument drift detection device 10 of the first embodiment.

  The instrument drift detection device 10 of the first embodiment is applied to a water level gauge of a boiling water nuclear power plant (plant equipment), and the information processing system 100 includes a change detection means 101, as shown in FIG. Drift determination means 102. The measuring instrument drift detection device 10 of the present embodiment detects the drift of four water level meters (measuring instruments) that measure the reactor water level (process data). Note that these four water level gauges measure the same reactor water level although the installation locations are different, and thus the measured process data have a strong correlation with each other.

  The change detection means 101 of the information processing system 100 applies a statistical method to the time-series process data measured in the plant, thereby changing the process data indicating whether the process data has increased or decreased. Is determined as a detection result. In addition, the drift determination unit 102 of the information processing system 100 receives the detection result regarding the change of the process data from the change detection unit 101, and determines the drift of the measuring instrument using the detection result. The change detection unit 101 and the drift determination unit 102 of the information processing system 100 are realized as a program that operates on a computer.

  The change detection means 101 of the information processing system 100 includes a change detection parameter recording unit 103, an increase detection unit 104, and a decrease detection unit 105.

  The change detection parameter recording unit 103 of the change detection unit 101 records a parameter that is a determination criterion for a change in process data. The increase detection unit 104 of the change detection unit 101 generates a detection result for a change in process data such as the presence or absence of an increase in the process data using the process data. The decrease detection unit 105 of the change detection unit 101 generates a detection result for a change in process data such as the presence or absence of a decrease in process data using the process data.

  The drift determination means 102 of the information processing system 100 includes a data control unit 106, a coincidence determination unit 107, and a drift determination unit 108.

  The data control unit 106 of the drift determination unit 102 selects one process data from the process data indicating the detection result and sends it to the drift determination unit 108 of the drift determination unit 102, and the remaining process data not selected, That is, the flow of process data is controlled so that three process data are sent to the coincidence determination unit 107 of the drift determination means 102. The coincidence determination unit 107 of the drift determination unit 102 determines the coincidence of the detection results of the three process data changes received from the data control unit 106 of the drift determination unit 102. Then, the drift determination unit 108 of the drift determination unit 102 determines the drift of the measuring instrument according to a predetermined condition.

  Next, the effect | action of the measuring device drift detection apparatus 10 of 1st Embodiment is demonstrated.

  FIG. 2 is a flowchart showing processing executed in the information processing system 100 according to the first embodiment. FIG. 2A shows a flow of processing (change detection processing) executed in the change detection unit 101 of the information processing system 100. The flowchart (B) is a flowchart showing the flow of processing (drift determination processing) executed in the drift determination means 102 of the information processing system 100. Hereinafter, each step will be described.

(Change detection process)
The SPRT method (sequential probability ratio test) executed in the increase detection unit 104 of the change detection means 101 of the information processing system 100 and used to determine whether or not process data has increased will be described. The determination of whether or not the process data is reduced executed by the reduction detection unit 105 of the change detection unit 101 is based on the same method.

In the SPRT method, the log likelihood ratio λ is used as a judgment index. The log likelihood ratio λ is “the process data is increasing relative to the probability P (γ | H ₀ ) obtained under the assumption H ₀ that the measured process data γ is“ the process data is not increasing ”. Is the logarithm of the ratio of the probability P (γ | H ₁ ) obtained under the assumption H ₁ . That is, it is expressed by the following formula. Note that λ can be calculated sequentially when the process data has no time-series dependency.

Assuming that γ (k) is measured as the kth data when the log likelihood ratio up to k−1th in the time series of the process data is λ (k−1), λ (k) is expressed by the following equation. Calculated.

Furthermore, when P (γ | H ₀ ) and P (γ | H ₁ ) follow a normal distribution, Equation (2) becomes as follows (see: Takamitsu Nakamizo, “Statistical Fault Detection Method for System”). "Measurement and control Vol.18, No.6, 471-480).

  Next, change detection processing using the SPRT method executed by the change detection means 101 of the information processing system 100 will be described with respect to process data measured by each water level gauge. For the sake of explanation, it is assumed that the four process data are L1, L2, L3, and L4.

<Preparation>
Step A1: a reference parameter from the process data L1~L4 as a pre-reference, i.e., calculates an average value a ₀ and a standard deviation sigma, and records the change detection parameter recording unit 103 of the change detection means 101. Furthermore, the assumed values when the process data L1 to L4 are assumed to be “increased”, that is, the average value a ₁ , the alternative hypothesis determination threshold A, and the null hypothesis determination threshold B are recorded in the change detection parameter recording unit 103. To do.

  Step A2: The increase detection unit 104 sets an initial value λ (0) = 0 of the log likelihood ratio λ by the SPRT method.

<Runtime processing>
The following processing procedure (step B1 to step B5) is repeated for each of the time series of the process data L1, L2, L3, and L4. Here, the process data L1, L2, L3, and L4 are represented as Ln as a representative.

  Step B1: The increase detection unit 104 of the change detection means 101 reads time-series k-th process data γ (k) for each of the time-series process data Ln.

  Step B2: Following Step B1, using the value of the log likelihood ratio λ (k−1) up to the previous time, the log likelihood ratio λ (k) is calculated according to Equation (3). The log likelihood ratio for the first time series k = 1 is λ (k−1) = λ (0) = 0.

  Step B3: Following step B2, the increase detection unit 104 determines whether the log likelihood ratio λ (k) is greater than the alternative hypothesis determination threshold A.

  Step B4: If the log likelihood ratio λ (k)> the alternative hypothesis determination threshold A, the increase detection unit 104 sets “increase in process data” as the detection result, and notifies the drift determination means 102 of this detection result. . Then, the log likelihood ratio λ (k) is set to 0, and the change detection process for the presence or absence of the increase is continued.

  Step B5: If the log likelihood ratio λ (k) <B, the increase detection unit 104 assumes that the detection of “increase in process data” is denied, and sets “no increase” in process data as a detection result, This detection result is notified to the drift determination means 102. Then, the number likelihood ratio λ (k) is set to 0, and the change detection process for the presence or absence of the increase is continued.

  On the other hand, the decrease detection unit 105 of the change detection unit 101 performs a change detection process on whether or not the process data Ln has decreased by the same procedure as in Steps B1 to B5, and the drift determination unit 102 receives “Decrease” or “ The detection result of “no decrease” is notified.

  That is, the change detection means 101 detects the detection result `` with increase '' or `` without increase '' in the increase detection unit 104, the detection result `` with decrease '' or `` without decrease '' in the decrease detection unit 105, Each is notified to the drift determination means 102.

(Drift judgment processing)
Next, the drift determination process in the drift determination means 102 of the information processing system 100 will be described. The drift determination unit 102 receives a detection result such as “with increase” or “without increase” with respect to each process data Ln from the change detection unit 101 of the information processing system 100, and repeatedly executes the following processing (step C1 to step C5). To do.

  Step C1: Receive a detection result for a change in the process data Ln from the change detection means 101.

  Step C2: Following Step C1, the data control unit 106 of the drift determination unit 102 selects process data one by one from the process data Ln, and the detection result regarding the selected process data is the drift determination unit of the drift determination unit 102. The result of detection of the remaining three process data not selected is sent to the coincidence determination unit 107.

  Step C3: Following step C2, the coincidence determination unit 107 of the drift determination unit 102 determines whether or not the detection results of the three process data coincide with each other. For example, when all the detection results of the three process data are “no increase” and “no decrease”, or “increase” and “no decrease”, it is determined that they match.

  Step C4: When the detection results of the three process data match each other, the coincidence determination unit 107 of the drift determination unit 102 sends the detection results of the three process data to the drift determination unit 108 of the drift determination unit 102. When the detection results of the three process data do not match each other, the coincidence determination unit 107 of the drift determination unit 102 does not send the detection results of any process data to the drift determination unit 108 of the drift determination unit 102, and the drift determination process The process returns to step C1.

  Step C5: The drift determination unit 108 of the drift determination unit 102 matches the detection result of one selected process data selected and sent directly by the data control unit 106 of the drift determination unit 102 with the drift determination unit 102 The three detection results that coincide with each other sent from the determination unit 107 are compared, and it is determined that there is a drift for the measuring instrument that measures one process data selected according to the drift determination table (see FIG. 3).

  In other words, the drift determination means 102 of the information processing system 100 determines, based on changes in the remaining three process data that are not selected, when determining whether or not there is a drift for one selected process data.

  FIG. 3 is a diagram showing a drift determination table used in the drift determination process in the drift determination means 102 of the information processing system 100. FIG. 3 relates to one process data selected by the data control unit 106 of the drift determination means 102 among the process data Ln. Hereinafter, the contents of the drift determination table will be described using some examples.

  In the drift determination process according to this drift determination table, the detection result for the change of one selected process data is “no increase” and “no decrease”, and the detection result for the three process data changes not selected is the same. If there is no increase and no decrease, it is determined that there is no drift with respect to the measuring instrument that has measured one selected process data. That is, it can be said that there is a high possibility that the process data has not changed on the basis of the detection result of the change in the three process data not selected. At this time, since it is determined that the process data of the selected one process data is not changed, it is determined that the measuring instrument is normal and no drift is detected.

  On the other hand, the detection result for the change of one selected process data is “no increase” and “no decrease”, and the detection result for three unselected process data changes is “with increase” and “no decrease”. If there is, a drift toward the decreasing side is determined with respect to the measuring instrument that has measured one selected process data. That is, it can be said that there is a high possibility that the process data has increased on the basis of the detection results of the three process data changes. At this time, since it is determined that the selected one process data has not increased, this measuring instrument determines that the process data is underestimated and determines that there is a drift toward the decreasing side.

  Also, the detection result for the change of one selected process data is “with increase” and “no decrease”, and the detection result for the change of three process data not selected is also “with increase” and “no decrease”. ", It is determined that the process data itself has increased. That is, it can be said that there is a high possibility that the process data has increased on the basis of the detection results of the three process data changes. At this time, since it is determined that one selected process data has also increased, it is determined that there is a high possibility that the process data itself has increased, and it is determined that the process data has increased.

  Next, the effect of the measuring instrument drift detection apparatus 10 of 1st Embodiment is demonstrated.

  In the instrument drift detection device 10 of the first embodiment, the effects listed below can be obtained.

  (1) An instrument drift detector that is provided in a boiling water nuclear power plant and that detects four instrument measuring instruments whose correlation is recognized in the measured process data, and detects the drift of the instrument using the process data. 10, a change detection unit 101 that performs a change detection process that determines whether or not the process data has increased or decreased for each of the process data, and detects a change in the process data as a detection result; The detection result is received, one process data is selected from the process data indicating the detection result, and it is determined whether or not the detection results of the remaining three process data not selected match, and this match is determined. If this detection result matches the detection result of one selected process data, it is determined whether or not both are detected. Since there is provided drift determination means 102 for performing drift determination processing for determining that there is a drift for the measuring instrument that has measured one selected process data on condition that the results do not match, the accuracy of the true value estimation of the process data Or the fall of the reliability of drift detection resulting from the correlation of process data can be avoided, and in addition, the versatility can be improved without requiring a drift determination index corresponding to the type of measuring instrument.

  That is, a change in process data such as whether or not each process data has increased or decreased is determined (change detection process), and the presence or absence of drift is determined by comparing the determined process data (drift determination process). This comparison is performed by selecting one process data measured by a specific measuring device to be subjected to the determination of whether or not there is a drift and comparing it with a change in the remaining process data not selected. Also, this comparison is made when the remaining unselected process data changes match, so these unselected remaining process data changes are more likely due to plant process variations rather than instrument drift. Reliable as a criterion for determining the presence or absence of drift.

  That is, in the measuring instrument drift detection device of the present invention, when determining the presence or absence of drift of the measuring instrument, the change in process data that can be determined without using the correlation of the measured process data is used, and the true value is estimated. There is no need to do. Further, in the comparison, a universal index is used regardless of the type of measuring instrument, such as whether the process data has increased or decreased.

  (2) Since the change detection unit 101 performs change detection processing using a statistical method using a sequential probability ratio test for each process data, the change of the process data can be determined with high reliability. .

[Second Embodiment]
The second embodiment is an example in which a change confirmation unit 301 is added to the configuration of the instrument drift detection device 10 of the first embodiment.

  First, the configuration of the measuring instrument drift detection device 10A of the second embodiment will be described.

  FIG. 4 is a functional block diagram of an information processing system 100A in the measuring instrument drift detection apparatus 10A of the second embodiment.

  The information processing system 100A of the measuring instrument drift detection apparatus 10A includes a change confirmation unit 301 as shown in FIG. In an actual plant, even if the measuring instrument is not drifting, process data may change due to some transient phenomenon. The change confirmation unit 301 removes the influence of such a transient phenomenon.

  When the change confirmation unit 301 of the information processing system 100A receives the detection result on whether or not the process data has increased or decreased from the change detection unit 101, confirms the validity of the detection result, and the detection result is valid The detection result is sent to the drift determination means 102 of the information processing system 100A only (confirmation process).

  The confirmation processing in the change confirmation means 301 is performed when the determination of whether there is an increase or decrease in process data in the change detection process in the change detection means 101 exceeds a preset reference number N. This is done by determining that it is valid.

  The change confirmation unit 301 of the information processing system 100A includes a confirmation parameter recording unit 302, a determination history recording unit 303, and a change confirmation unit 304.

  The confirmation parameter recording unit 302 of the change confirmation unit 301 records parameters used for confirming the change of process data, and the determination history recording unit 303 of the change confirmation unit 301 records the determination history of process data change. Then, the change confirmation unit 304 of the change confirmation unit 301 confirms whether the change in the process data is due to the drift of the measuring instrument.

  The change confirmation unit 301 of the information processing system 100A is realized as a program that operates on a computer. Since other configurations are the same as those of the first embodiment, the corresponding components are denoted by the same reference numerals and description thereof is omitted.

  Next, the operation of the measuring instrument drift detection device 10A of the second embodiment will be described.

  FIG. 6 is a flowchart showing a flow of processing (confirmation processing) executed by the change confirmation unit 301 of the second embodiment in FIG. 5, and each step will be described below.

(Confirmation process)
<Preparation>
Step D1: The reference number N of determinations is recorded in the confirmation parameter recording unit 302 of the change confirmation unit 301.

  Step D2: Number of determinations = 0 is recorded in the determination history recording unit 303 of the change confirmation unit 301 that records the number of determinations regarding whether process data has increased or decreased.

<Runtime>
For example, in order to confirm the validity of the detection result regarding the increase in process data, the following confirmation processing is executed.

  Step E1: In the information processing system 100A, the change confirmation unit 304 of the change confirmation unit 301 receives a detection result on whether or not process data has increased from the change detection unit 101.

  Step E2: Following the step E1, if the received detection result is “increase” in the process data, the determination history recording unit 303 of the change confirmation unit 301 counts up the increase determination count by one and sets the determination count to the current Update to the number of counts.

  Step E3: Following step E2, the change confirmation unit 304 of the change confirmation unit 301 determines whether or not the increase determination number has exceeded the reference number N recorded in the confirmation parameter recording unit 302 of the change confirmation unit 301.

  Step E4: Following step E3, the change confirmation unit 304 of the change confirmation unit 301 determines that the detection result of “with increase” by the change detection process is valid when the increase determination count exceeds the reference count N. The detection result “increase” is sent to the drift determination means 102.

  Step E5: Following step E4, the number of increase determinations = 0 is recorded in the determination history recording unit 303 of the change confirmation unit 301.

  In the case where the detection results received from the change detection means 101 in step E1 are “no increase”, “with reduction”, and “no reduction”, the same confirmation processing as in steps E1 to E5 is performed. Since other operations are the same as those of the first embodiment, description thereof is omitted.

  However, the drift determination unit 102 performs a drift determination process using the same method as in the first embodiment on the detection result determined to be valid in the confirmation process in the change confirmation unit 301 for each process data.

  Next, the effect of the measuring instrument drift detection apparatus 10A of the second embodiment will be described.

  In the instrument drift detection device 10A of the second embodiment, in addition to the effects (1) and (2) of the first embodiment, the following effects can be obtained.

  (3) Counts the number of times that the process data increase or decrease is continuously determined in the change detection process of the change confirmation means 301, and if the counted number exceeds the preset reference number N, the process It is provided with a change confirmation means 301 that performs a detection result confirmation process for determining that a detection result having data increase or decrease is valid, and the drift determination means 102 performs drift determination processing for the valid detection result. Therefore, the drift of the measuring instrument can be determined by removing the influence of the transient process fluctuation.

  (4) The change confirmation unit 301 receives the detection result from the change detection unit 101, and sends only the detection result determined to be valid in the confirmation process to the drift determination unit 102. The drift determination unit 102 receives the change determination unit 301 from the change confirmation unit 301 Since the drift determination process is performed on the detection result, the effect shown in (3) can be easily obtained.

[Third Embodiment]
The third embodiment is an example in which the configuration of the change confirmation unit 301 of the second embodiment is modified.

  First, the structure of the measuring instrument drift detection apparatus 10B of 3rd Embodiment is demonstrated.

  FIG. 6 is a functional block diagram of the information processing system 100B in the measuring instrument drift detection apparatus 10B of the third embodiment.

  As shown in FIG. 6, the information processing system 100B of the instrument drift detection device 10B includes a determination frequency confirmation unit 305 instead of the change confirmation unit 304 of the change confirmation unit 301 of the second embodiment. . The determination frequency confirmation unit 305 removes the influence of the transient phenomenon by a method different from that of the change confirmation unit 304 of the second embodiment.

  When the increase determination frequency or the decrease determination frequency calculated based on the detection result received from the change detection unit 101 of the information processing system 100B exceeds a preset reference frequency, the change confirmation unit 301B of the information processing system 100B The received detection result is regarded as valid, and the detection result is sent to the drift determination means 102 of the information processing system 100B. This increase determination frequency is defined as the number of determinations with an increase with respect to the number of determinations of whether there is an increase or not when the increase detection unit 104 of the change detection unit 101 determines that there is an increase. This is defined as the number of determinations with a decrease with respect to the number of determinations of whether or not there is a decrease when “decrease is present” is determined by the decrease detection unit 105 of 101. Since other configurations are the same as those of the second embodiment, the corresponding components are denoted by the same reference numerals and description thereof is omitted.

  Next, the operation of the instrument drift detection device 10B of the third embodiment will be described.

(Confirmation process)
The SPRT method executed in the change confirmation means 301B of the instrument drift detection apparatus 100B and used for the confirmation process of the increase determination frequency and the decrease determination frequency will be described.

If the process data fluctuates in a constant state, the increase determination frequency follows a binomial distribution. Accordingly, "increase determination frequency is p" assume the assumption H _p of, consider a hypothetical H _q as "increase determination frequency is q", the p <q. In a normal state where transient fluctuations in process data occur, “increase” is determined by p <q. On the other hand, in a state where the drift of the measuring instrument is occurring, “increase” is determined by p> q.

When the increase determination frequency is y, the log likelihood ratio μ (k) of the probabilities P (y | Hp) and P (y | Hq) determined under the assumption Ha (a = p or q) is It is calculated as follows. Note that k indicates a time series.

Substituting these into equation (5) yields:

  When n = 1, that is, when the expression (6) is calculated every time the change detecting means 101 of the information processing type 100B performs the process once, it is as follows.

When the change detection means 101 of the information processing type 100B determines “increase”.
In this case, the increase determination frequency y is 1/1 = 1, and the equation (6) is as follows.

When the change detection means 101 of the information processing type 100B determines “no increase”.
In this case, the increase determination frequency y is 0/1 = 0, and Expression (6) is as follows.

  FIG. 7 is a flowchart showing the flow of processing (confirmation processing) executed by the change confirmation means 301B of the second embodiment. Each step will be described below. Here, the confirmation process for the increase determination frequency will be described.

<Preparation>
Step F1: The reference frequency p in the standard state and the reference frequency q serving as a reference for confirming the validity of the detection result “with increase” by the change detection process are recorded in the confirmation parameter recording unit 302 of the change confirmation unit 301B. . In addition, the confirmation parameter recording unit 302 records a threshold A for confirming the detection result “with increase” and a threshold B for negating the detection result “with increase” with respect to the log likelihood ratio μ regarding the increase determination frequency. .

  Step F2: The initial value μ (0) = 0 of the log likelihood ratio μ (k) is recorded in the determination history recording unit 303 of the change confirmation unit 301B.

<Runtime>
Step G1: The determination frequency confirmation unit 305 of the change confirmation unit 301B receives a detection result about whether or not there is an increase from the change detection unit 101.

  Step G2: Subsequent to step G1, the determination frequency confirmation unit 305 of the change confirmation unit 301B determines the value μ (k−1) from the determination history recording unit 303 of the change confirmation unit 301B to the previous log likelihood ratio μ (k). Take out.

  Step G3: Following step G2, the determination frequency confirmation unit 305 of the change confirmation unit 301B calculates a log likelihood ratio μ (k) with respect to the detection result regarding the presence or absence of the k-th increase by the equation (7), Μ (k) is recorded in the determination history recording unit 303 of the change confirmation unit 301B. If the detection result is “no increase”, the log-likelihood ratio μ (k) is calculated by the equation (8).

  Step G4: Following step G3, the determination frequency confirmation unit 305 of the change confirmation unit 301B determines the content of the log likelihood ratio μ (k).

  Step G5: If μ (k)> A in Step G4, the determination frequency confirmation unit 305 of the change confirmation unit 301B confirms that the detection result “with increase” is valid, and drifts the detection result “with increase”. The determination unit 102 is notified. Then, μ (k) recorded in the determination history recording unit 303 of the change confirmation unit 301B is reset.

  Step G6: On the other hand, if μ (k) <B in Step G4, the determination frequency confirmation unit 305 of the change confirmation unit 301B negates the detection result “with increase” and sets the detection result “without increase” as the drift determination unit. Notify 102. Then, μ (k) recorded in the determination history recording unit 303 of the change confirmation unit 301B is reset.

  The confirmation processing in the change confirmation means 301B described above follows the processing in steps G1 to G6 even when the reduction determination frequency is targeted.

  Next, the effect of the measuring instrument drift detection apparatus of 3rd Embodiment is demonstrated.

  In the instrument drift detection device 10B of the third embodiment, in addition to the effects (1) and (2) of the first embodiment, the following effects can be obtained.

  (5) The increase determination frequency is the number of determinations with an increase with respect to the number of determinations of whether process data has been increased or not in the change detection process of the change confirmation unit 301B, and the determination number with decrease is the decrease determination frequency with respect to the determination number with or without decrease. When the increase determination frequency or decrease determination frequency is calculated, and the calculated increase determination frequency or decrease determination frequency exceeds a preset reference frequency, there is an increase or decrease in the change detection process of the change confirmation unit 301B. A change confirmation means 301B for performing a confirmation process for validating the detection result having the content of presence, and the drift determination means 102 performs the drift determination process for the valid detection result, so that the process transient The drift of the measuring instrument can be determined by removing the influence of fluctuations.

  (6) The change confirmation unit 301B receives the detection result from the change detection unit 101, and sends only the detection result determined to be valid in the confirmation process to the drift determination unit 102. The drift determination unit 102 receives the change result from the change confirmation unit 301B. Since the drift determination process is performed on the detection result, the effect shown in (5) can be easily obtained.

[Fourth Embodiment]
The fourth embodiment is an example in which parameter adjustment means 401 is added to the configuration of the measuring instrument drift detection device 10A of the second embodiment.

  First, the structure of the measuring instrument drift detection apparatus 10C of 4th Embodiment is demonstrated.

  FIG. 8 is a functional block diagram of an information processing system 100C in the instrument drift detection device 10C of the fourth embodiment.

  The information processing system 100C of the instrument drift detection device 10C of the fourth embodiment includes parameter adjustment means 401 as shown in FIG. The parameter adjusting means 401 adjusts a parameter that is a target of change detection when the process of the boiling water nuclear power plant to which the instrument drift detection device 10C of the present embodiment is applied changes.

  The parameter adjustment unit 401 of the information processing system 100C includes a parameter change unit 402 and a reference value calculation unit 403. The parameter adjustment unit 401 is realized as a program that operates on a computer. Since other configurations are the same as those of the second embodiment, the corresponding components are denoted by the same reference numerals and description thereof is omitted.

  Next, the operation of the measuring instrument drift detection device 10C of the fourth embodiment will be described.

  FIG. 9 is a flowchart showing the flow of processing (parameter adjustment processing) executed by the parameter adjustment means 401 of the fourth embodiment. Each step will be described below.

(Parameter adjustment processing)
Step H1: Each time the drift determination means 102 makes a drift determination, the parameter changing unit 402 of the parameter adjustment means 401 receives the result of the drift determination. Here, as shown in FIG. 3, the drift determination includes “process data increase” and “process data decrease” indicating an increase / decrease in process data itself, which is caused by a process change and is not related to the drift of the measuring instrument. .

  Step H2: Following Step H1, the parameter changing unit 402 of the parameter adjusting unit 401 determines whether or not the drift determination received in Step H1 is process data increase or process data decrease.

  Step H3: Following Step H2, when the result of the drift determination received by the parameter changing unit 402 of the parameter adjusting unit 401 is process data increase or process data decreasing, the parameter changing unit 402 of the parameter adjusting unit 401 Parameters (average value, standard deviation) corresponding to new process data after the process variation calculated by the calculation unit 403 are acquired.

Step H4: Subsequent to Step H3, the parameter changing unit 402 of the parameter adjusting means 401 uses the new average value and standard deviation acquired in Step H3 to assume the average value a ₁ , process data when the increase in process data is assumed. An average value a ₂ is calculated when the decrease is assumed to be sent to the change detecting means 101. In the present embodiment, the average value a ₁ is obtained by adding twice the standard deviation to the average value of the new process data, and the average value a ₂ is twice the standard deviation of the average value of the new process data. It is calculated by subtracting.

  Here, the reference value calculation unit 403 of the parameter adjustment unit 401 constantly receives the process data, and calculates the average value and standard deviation of the process data using the latest 500 pieces of data. In response to a request from the parameter changing unit 402 of the parameter adjusting unit 401, these values are sent to the change detecting unit 101.

<Change detection process>
In the information processing system 100C, the change detecting means 101 receives the average value a _{1 of} the new process data from the parameter adjusting means 401, the standard deviation, the average value a ₁ when the process data is “increased”, and the process data is “decreased”. When the average value a ₂ is received, these average values are recorded as new parameters in the change detection parameter recording unit 103 of the change detection means 101, and the same changes as described in the first embodiment are recorded. Perform detection processing.

  Next, the effect of the measuring instrument drift detection apparatus 10C of the fourth embodiment will be described.

  In the instrument drift detection apparatus 10C of the fourth embodiment, in addition to the effects of (1) and (2) of the first embodiment and (3) and (4) of the second embodiment, the following effects are obtained. Obtainable.

  (7) When the measured process data itself changes, the parameter used for determining the change of the process data in the change detection process of the change detecting means 101 is adjusted to the parameter corresponding to the new changed process data. Since the parameter adjusting means 401 is provided, the presence / absence of drift of the measuring instrument can be determined even when the operation state of the plant changes during the process and the process amount changes.

[Fifth Embodiment]
The fifth embodiment is an example in which representative value calculation means 501 is added to the configuration of the instrument drift detection device 10 of the first embodiment.

  First, the structure of the instrument drift detection apparatus 10D of 5th Embodiment is demonstrated.

  FIG. 10 is a functional block diagram of an information processing system 100D in the instrument drift detection device 10D of the fifth embodiment.

  The instrument drift detection device 10D of the fifth embodiment includes representative value calculation means 501. The instrument drift detection device 10D of this embodiment is applied to the measurement of the differential pressure between the containment vessel of the boiling water nuclear power plant and the external pressure, and the process data measured by a plurality of instruments has a strong correlation and It is applied to the scene where those process data changes in time series.

  The representative value calculation unit 501 of the information processing system 100D includes a representative value calculation unit 502 and a deviation calculation unit 503.

  The representative value calculator 502 of the representative value calculator 501 calculates representative values of a plurality of process data measured by a plurality of measuring instruments. The deviation calculation unit 503 of the representative value calculation unit 501 calculates the deviation of each process data from the representative value.

  The representative value calculation means 501 of the information processing system 100D is realized as a program that operates on a computer. Since other configurations are the same as those of the first embodiment, the corresponding components are denoted by the same reference numerals and description thereof is omitted.

  Next, the operation of the instrument drift detection device 10D of the fifth embodiment will be described.

  FIG. 11 is a flowchart showing a flow of processing (representative value calculation processing) executed by the representative value calculation means 501 of the fifth embodiment. Each step will be described below. FIG. 12 is a diagram showing data relating to the differential pressure between the storage container and the outside measured by four measuring instruments. FIG. 12A shows representative values M of the differential pressures P1 to P2 and the differential pressures P1 to P4. The figure shown, (B) is a figure which shows the deviation from the typical value M regarding the differential pressure P1-P4 shown to (A).

<Representative value calculation processing>
Step I1: The representative value calculator 502 and the deviation calculator 503 of the representative value calculator 501 acquire process data at a constant period.

  Step I2: Following step I1, the representative value calculation unit 502 of the representative value calculation unit 501 calculates the representative value M of the acquired process data, and sends the calculated representative value M to the deviation calculation unit 503 of the representative value calculation unit 501. send. As the representative value M, an average value of the differential pressures P1 to P4 may be used, but in the present embodiment, an intermediate value of the differential pressures P1 to P4 is used.

  Step I3: Following step I2, the deviation calculation unit 503 of the representative value calculation means 501 calculates and calculates the deviation (see FIG. 12B) from the representative value M acquired for the differential pressures P1 to P4 as the process data. The deviation is sent to the change detection means 101.

<Change detection process>
The change detection means 101 of the information processing system 100D performs the above-described change detection process for the presence or absence of increase or decrease of the deviation instead of process data. Then, the detection result is sent to the drift determination means 102 of the information processing system 100D.

<Drift judgment processing>
The drift determination unit 102 of the information processing system 100D performs the above-described drift determination process on the detection result regarding the deviation received from the change detection unit 101 of the information processing system 100D.

  Next, the effect of the instrument drift detection apparatus 10D of the fifth embodiment will be described.

  In the instrument drift detection device 10D of the fifth embodiment, the following effects can be obtained in addition to the effects (1) and (2) of the first embodiment.

  (8) A representative value calculation unit 501 for calculating a representative value of the process data from the process data and a deviation from the representative value is provided, and the change detection unit 101 replaces the determination of the change of the process data with the increase of the deviation. The change in deviation is determined based on the presence / absence and the presence / absence of a decrease, and a change detection process using the change in deviation as a detection result is performed, and the drift determination means 102 receives the detection result on the change in deviation and receives the detected Since the drift determination process is performed on the result, the drift of the measuring instrument can be determined even when the change in the process amount is large.

[Sixth Embodiment]
The sixth embodiment is an example in which data selection means 601 is added to the configuration of the instrument drift detection device 10 of the first embodiment.

  First, the structure of the measuring instrument drift detection apparatus 10E of 6th Embodiment is demonstrated.

  FIG. 13 is a functional block diagram of the information processing system 100E in the measuring instrument drift detection apparatus 10E of the sixth embodiment.

  The instrument drift detection device 10E of the information processing system 100E includes data selection means 601. The instrument drift detection device 10E of this embodiment selects only process data when the generator output of the nuclear power plant is within a certain range, and performs drift determination processing and the like. The process data relates to the water level as in the first embodiment.

  The data selection unit 601 of the information processing system 100E includes a selection condition recording unit 602 and a data selection unit 603.

  The selection condition recording unit 602 of the data selection unit 601 records a selection condition for selecting process data. A plurality of selection conditions are set. For this reason, the selection condition recording unit 602 of the data selection means 601 is provided in a multiplexed number corresponding to the number of selection conditions.

  The data selection unit 603 of the data selection unit 601 selects only process data when the condition data is input together with the process data and the condition data matches the selection condition recorded in the selection condition recording unit 602 of the data selection unit 601. This is sent to the change detection means 101.

  The data selection means 601 of the information processing system 100E is realized as a program that operates on a computer. Since other configurations are the same as those of the first embodiment, the corresponding components are denoted by the same reference numerals and description thereof is omitted.

However, as shown in FIG. 13, the change detection parameter recording unit 103E of the change detection means 101 is provided with the number of selection conditions for the process data multiplexed in correspondence with the setting of a plurality of process data selection conditions. It is done. That is, in each change detection parameter recording unit 103A, the average value a ₀ as a reference, the standard deviation σ, the average value a ₁ when the process data is “increased”, and the process data “there is“ decrease ”are assumed. the average value a ₂ when the recording.

  Next, the operation of the instrument drift detection device 10E of the sixth embodiment will be described.

  FIG. 14 is a flowchart showing the flow of processing (data selection processing) executed by the data selection means 601 of the sixth embodiment. Each step will be described below.

<Data selection process>
Step J1: In the selection condition recording unit 602 of the data selection means 601, two conditions “generator output 98% or more” and “generator output 50% or more and 55% or less” are selected as selection conditions for selecting process data. Record.

  Step J2: Following step J1, the data selection unit 603 of the data selection unit 601 acquires the condition data together with the process data. This condition data is the same kind of generator output as the selection condition.

  Step J3: Following step J2, the data selection unit 603 of the data selection unit 601 determines whether the acquired condition data satisfies the selection condition recorded in the selection condition recording unit 602 of the data selection unit 601. .

  Step J4: Following step J3, the data selection unit 603 of the data selection means 601 determines that the acquired condition data is “generator output of 98% or more” or “generator output of 50% or more and 55% or less”. Then, the selection condition “generator output 98% or more” or “generator output 50% or more and 55% or less” is sent to the change detection means 101 together with the process data. On the other hand, if the acquired condition data does not satisfy the selection condition, the process data is not sent to the change detection means 101.

<Change detection process>
The increase detection unit 104 and decrease detection unit 105 of the change detection unit 101 perform change detection processing on the process data sent from the data selection unit 603 of the data selection unit 601. That is, the increase detection unit 104 and the decrease detection unit 105 of the change detection unit 101 perform the process data within the range of the selection conditions “generator output 98% or more” and “generator output 50% or more and 55% or less”. Then, change detection processing is performed.

  Next, the effect of the instrument drift detection apparatus 10E of the sixth embodiment will be described.

  In the instrument drift detection device 10E of the sixth embodiment, in addition to the effects (1) and (2) of the first embodiment, the following effects can be obtained.

  (9) Process data that is acquired from the plant and that is different from the process data is acquired as condition data together with the process data, and it is determined whether or not the condition data satisfies one or more preset selection conditions Since the data selection means 601 that allows the change detection process for the process data acquired together with the condition data only when the condition data satisfies the selection condition, the process drift is eliminated and the instrument drift is reduced. Can be judged.

  As mentioned above, although the instrument drift detection apparatus of this invention has been demonstrated based on 1st Embodiment-6th Embodiment, about a specific structure, it is not restricted to these embodiment, Claim of Claim Design changes and additions are allowed without departing from the spirit of the invention according to each claim.

  In the first embodiment, in the drift determination process, four process data measured by four measuring instruments are targeted, one process data is selected from the four process data in which the detection results are shown, and the remaining unselected Although the example in which it is determined whether or not the detection results of all three process data match and the subsequent processing is continued has been shown, some of the detection results (for example, two) of the remaining three not selected are It may be determined whether or not they match and the subsequent processing may be continued.

  In the first embodiment, the example in which the measuring instrument drift detection device 10 is applied to a water level gauge of a boiling water nuclear power plant has been described. However, for example, it is provided in a plan facility such as a boiling water nuclear power plant or a thermal power plant. You may apply to the thermometer (measurement device) which measures the temperature of the heat exchanger to be manufactured. In this case, it can be applied as follows.

  FIG. 15 is a diagram showing a plant heat exchanger 100 to which the measuring instrument drift detection device 10 of the present invention is applied. FIG. 16 is a diagram showing a drift determination table of a thermometer used in the drift determination process of the drift determination means 102 of the present invention. The heat exchanger 100 includes a body 111 of the heat exchanger 100, a pipe 112 of the heat exchanger 100 through which the low temperature fluid flows, and a pipe 113 of the heat exchanger 100 through which the high temperature fluid flows.

The heat exchanger 100 is provided with a first thermometer 114 and a second thermometer 115. The first thermometer 114 measures the exit temperature of the cryogenic fluid (measured value is T _C1A ). The second thermometer 115 measures the outlet temperature (T _C1B ) of the cryogenic fluid. Since the temperatures T _C1A and T _C1B measured by the two thermometers are temperatures related to the same fluid, they have a strong correlation with each other.

The measured value measured by the thermometer 116 that measures the inlet temperature of the cold fluid is T _C2 , the measured value measured by the thermometer 117 that measures the inlet temperature of the hot fluid is T _H1 , and the outlet temperature of the hot fluid is The measured value measured by the measuring thermometer 118 is T _H2 , the measured value measured by the flow meter 119 that measures the flow rate of the cryogenic fluid is F _C , and the measured value is measured by the flow meter 120 that measures the flow rate of the hot fluid. _{Let the} value be F _H.

Here, from the heat energy conservation law in the heat exchanger 100, the outlet temperature of the low-temperature fluid can be obtained by the following equation. Here, the specific heat is constant.

The change detecting means 101 determines whether or not there is an increase or a decrease in the low temperature fluid outlet temperatures T _C1A and T _C1B and the low temperature fluid outlet temperature Tcc calculated in accordance with the equation (9). The detection result of is generated. Then, the drift determination means 102 performs a drift determination process on the detection result, and drifts with respect to the thermometer 114 that measures the outlet temperature T _C1A of the low temperature fluid and the thermometer 115 that measures the outlet temperature T _C1B of the low temperature fluid. The presence or absence of is determined. This determination follows the drift determination table of FIG.

  In the third embodiment, the increase determination frequency is defined by the number of determinations with an increase with respect to the number of determinations of whether there is an increase or not when the increase detection unit 104 of the change detection unit 101 determines “increase”. The determination frequency is defined as the number of determinations with a decrease relative to the number of determinations of whether or not there is a decrease when the decrease detection unit 105 of the change detection means 101 determines that there is a decrease, but there are various definitions of the increase determination frequency Can be considered. For example, the increase determination frequency may be determined based on “with increase” per M increase determination times. Since the increase determination frequency follows a binomial distribution, when the False-alarm rate is α, the standard deviation σ of the increase determination frequency when M determinations are made is σ = √ [α (1-α) / M]. . For example, a method of confirming that the detection result by the change detection means 101 is valid when the increase determination frequency exceeds α + 3σ can be considered.

  In the sixth embodiment, an example in which two selection conditions are used has been described, but one or three or more may be used. The selection condition need not be the same type of data, and different types of data such as the generator output and the reactor coolant temperature may be used.

  In the first to sixth embodiments, the change detection means 101, the drift determination means 102, the change confirmation means 301 and 301B, the parameter adjustment means 401, the representative value calculation means 501, and the data selection means 601 operate on a computer. Although an example realized as a program has been shown, it can also be realized as an electronic circuit using a semiconductor element such as an ASIC or FPGA.

  In the first embodiment to the sixth embodiment, an example of targeting four measuring instruments has been shown, but there is no reason limited to four, and a plurality of measuring instruments can be targeted.

  In the first to sixth embodiments, each means such as the change detection means 101 and the drift determination means 102 has a different function, but it goes without saying that it may be regarded as a single means.

The functional block diagram of the information processing system in the measuring instrument drift detection apparatus of 1st Embodiment. It is a flowchart which shows the process performed in the information processing system of 1st Embodiment, (A) is a flowchart which shows the flow of the process (change detection process) performed in the change detection means of an information processing system, (B) The flowchart which shows the flow of the process (drift determination process) performed in the drift determination means of an information processing system. The figure which shows the drift determination table | surface used by the drift determination process in the drift determination means of an information processing system. The functional block diagram of the information processing system in the measuring instrument drift detection apparatus of 2nd Embodiment. The flowchart which shows the flow of the process (confirmation process) performed in the change confirmation means of 2nd Embodiment. The functional block diagram of the information processing system in the measuring instrument drift detection apparatus of 3rd Embodiment. The flowchart which shows the flow of the process (confirmation process) performed in the change confirmation means of 2nd Embodiment. The functional block diagram of the information processing system in the measuring instrument drift detection apparatus of 4th Embodiment. The flowchart which shows the flow of the process (parameter adjustment process) performed in the parameter adjustment means of 4th Embodiment. The functional block diagram of the information processing system in the measuring instrument drift detection apparatus of 5th Embodiment. The flowchart which shows the flow of the process (representative value calculation process) performed in the representative value calculation means of 5th Embodiment. It is a figure which shows the data which concern on the differential pressure | voltage of the storage container measured with four measuring instruments, and the exterior, (A) is a figure which shows the representative value M of differential pressure P1-P2 and differential pressure P1-P4, (B (A) is a figure which shows the deviation from the typical value M regarding the differential pressure P1-P4 shown to (A). The functional block diagram of the information processing system in the measuring instrument drift detection apparatus of 6th Embodiment. The flowchart which shows the flow of the process (data selection process) performed in the data selection means of 6th Embodiment. The figure which shows the heat exchanger of the plant to which the measuring instrument drift detection apparatus of this invention was applied. The figure which shows the drift determination table | surface of the thermometer used in the drift determination process of the drift determination means of this invention.

Explanation of symbols

10, 10A, 10B, 10C, 10D, 10E Measuring instrument drift detection device 100, 100A, 100B, 100C, 100D, 100E Information processing system 101 Change detection means 102 Drift determination means 301, 301A Change confirmation means 401 Parameter adjustment means 501 Representative Value calculation means 601 Data selection means

Claims (8)

In a measuring instrument drift detection device for detecting a drift of a measuring instrument using a plurality of measuring instruments that are provided in a plant facility and are correlated with measured process data.
A change detecting means for determining a change in process data with the presence or absence of an increase or a decrease in process data for each of the process data, and performing a change detection process with the change in the process data as a detection result;
The detection result is received, one process data is selected from the process data indicating the detection result, and it is determined whether or not the detection results of the remaining process data that are not selected match each other. If the detection result matches the detection result of the selected one process data, the selected one process data is determined on condition that the two detection results do not match. Drift determination means for performing a drift determination process on the measured measuring instrument as having drift; and
An instrument drift detection device characterized by comprising: In the measuring instrument drift detection device according to claim 1,
The measuring device drift detecting apparatus, wherein the change detecting means performs the change detecting process by applying a statistical method using a sequential probability ratio test to each of the process data. In the measuring instrument drift detection device according to claim 1,
In the change detection process of the change detection means, the number of times the presence or absence of the process data is continuously determined is counted, and the increase in the process data when the counted number exceeds a preset reference number It is provided with a change confirmation means for performing a detection result confirmation process for determining that a detection result having presence or absence is valid.
The drift determination unit performs the drift determination process on the effective detection result as an object. In the measuring instrument drift detection device according to claim 1,
When the number of determinations with an increase with respect to the number of determinations of whether or not there is an increase in process data in the change detection process of the change detection means is an increase determination frequency, and the number of determinations with a decrease with respect to the number of determinations with or without a decrease is a decrease determination frequency,
When the increase determination frequency or the decrease determination frequency is calculated and the calculated increase determination frequency or decrease determination frequency exceeds a preset reference frequency, the detection result including the increase or decrease in the change detection process Is provided with a change confirmation means for confirming the detection result for determining that the
The drift determination unit performs the drift determination process on the effective detection result as an object. In the measuring instrument drift detection device according to claim 3 or claim 4,
The change confirmation unit receives the detection result from the change detection unit, and sends the detection result determined to be valid in the confirmation process to the drift determination unit.
The instrument drift detection apparatus, wherein the drift determination unit performs the drift determination process on the detection result received from the change confirmation unit. In the measuring instrument drift detection device according to claim 1,
When the measured process data itself changes, the parameter used for determining the change of the process data in the change detection process of the change detecting means is adjusted to a parameter corresponding to the changed new process data. A measuring instrument drift detecting device comprising a parameter adjusting means. In the measuring instrument drift detection device according to claim 1,
A representative value calculating means for calculating a representative value of the process data from the process data and a deviation from the representative value;
The change detection means, instead of determining the change in the process data, determines a change in the deviation with the presence or absence of the increase in the deviation or the presence or absence of the decrease, and performs a change detection process using the change in the deviation as a detection result. Done
The instrument drift detection device, wherein the drift determination unit receives a detection result of the change in the deviation and performs the drift determination process on the received detection result. In the measuring instrument drift detection device according to claim 1,
Whether the process data is acquired from the plant equipment and is different from the process data as condition data, and whether the condition data satisfies one or more preset selection conditions Only when it is determined that the condition data satisfies the selection condition, data selection means for allowing change detection processing in the change detection means for the process data acquired together with the condition data is provided. A measuring instrument drift detector.

Images