JP2006250541A - Calibration support device of detector and its method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a calibration support device of a detector and its method which can predict exact drift quantity by separately obtaining the drift quantity of the detector caused during startup and shutdown operation and the drift quantity caused during normal operation in the detectors of nuclear power plant and the like. <P>SOLUTION: In a statistical quantity evaluation means 14, a measured value in normal operation after the first regular inspection of a plurality of detectors, being the object of statistic quantity evaluation and a measured value during normal operation before the second regular inspection of a plurality of detectors are statistically processed to obtain the statistical drift quantity of a plurality of detectors causing during normal operation. The measured value of normal operation during normal operation after the first regular inspection and the measured value during normal operation after the second regular inspection, inspection results after drift control during the first regular inspection of a plurality of detectors and the inspection results before drift control during the second regular inspection of a plurality of detectors are statistically processed and the statistical quantity of drift quantity of a plurality of detectors caused in startup and shutdown is obtained. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a detector calibration support apparatus and method, and is particularly useful when applied to a large plant having a large number of detectors such as a nuclear power plant.

  A detector provided in a nuclear power plant or the like has a drift characteristic that its output (actual value) gradually shifts from a true value (true value) that should be detected with time. Accordingly, if the output is kept as it is, the output (actual measurement value) error of the detector becomes too large, which may hinder the monitoring and control of the plant based on the output (actual measurement value) of the detector.

  Therefore, with regard to the soundness confirmation of detectors used in nuclear power plants, etc., calibration (inspection) and drift adjustment work have been carried out when the operation of the plant or machinery has been stopped, such as during periodic inspections. It was. Specifically, for example, a reference signal generator is attached to a target detector in the plant to give a reference input signal, and whether the signal output from the detector at this time is within an allowable error compared to the reference output value. If it is not within the allowable error (within the allowable range), the setting is changed based on the adjustment procedure of the detector. There are various adjustment procedures for the detector. For example, some adjustments are made by operating an adjustment knob provided on the detector, and a dedicated tool is connected to change the set value in the semiconductor from the dedicated tool. There are also things.

  However, in the detector soundness confirmation method as described above, this cannot be performed unless the operation of the plants is stopped, which is not preferable from the viewpoint that the state of the detector is always suitably maintained. It is not preferable to frequently stop the operation of the plant for checking the soundness of the detector because the operation cost increases.

  From such a viewpoint, in recent years, attempts have been made to carry out detector health check online during plant operation. Patent Document 1 (Japanese Patent Laid-Open No. 10-104385) describes the necessity of calibration by comparing a trend of a detector signal stored in a database in the past with a current value or predicting a future drift amount. Therefore, a technique for calibrating during a periodic inspection and for providing a drift adjustment work plan is disclosed.

FIG. 17 shows a conceptual diagram of conventional drift amount prediction. As shown in FIG. 17A, conventionally, the drift characteristics of the detector have been evaluated based on the results of periodic inspection of the plant. That is, for example, as shown in FIGS. 17B to 17D, in a similar form and similar environmental conditions that occur between the N-th periodic inspection and the next N + 1-th periodic inspection. Assuming that the variance of the generated drift amount of the plurality of detectors used is σ ² and the standard deviation is σ, it was assumed that the drift adjustment of the detector was not performed in any of the subsequent N + 1 to N + 4 periodic inspections. If, because the variation in the generation amount of drift of the plurality of detectors can be regarded as a normal distribution, the variance sigma ² generation drift amount of the plurality of detectors 2 [sigma] ^2, 3 [sigma] ^2, the elapsed time and 4 [sigma] ² You may think that it grows proportionally. Therefore, as shown in FIG. 17 (e), the standard deviation σ of the drift amount generated by the plurality of detectors increases in proportion to the ½ power of the elapsed time. From this, the drift characteristic of the detector can be evaluated based on the results of the periodic inspection.

  Further, in Patent Document 2 (Japanese Patent Application No. 2004-301086), a plurality of true value estimation means are prepared, the applicability of each true value estimation means is evaluated to maintain accuracy, or data reconciliation techniques are combined. In addition to improving the accuracy of detector soundness evaluation, we are evaluating the aging and performance deterioration of plant equipment.

Japanese Patent Laid-Open No. 10-104385 Japanese Patent Application No. 2004-301086

  However, since Patent Document 1 does not distinguish between plant start / stop operation and normal operation, accurate prediction cannot be performed. Patent Document 2 also does not perform an evaluation that distinguishes between plant start / stop operation and normal operation.

  Therefore, in view of the above circumstances, the present invention is directed to a detector of a predetermined facility such as a nuclear power plant that is a calibration support target, and a drift amount of the detector generated during start / stop operation and a detector generated during normal operation. It is an object of the present invention to provide a detector calibration support apparatus and a method thereof that enable accurate prediction of a drift amount by separately determining the drift amount.

The detector calibration support apparatus according to the first aspect of the present invention that solves the above-described problems is used for a plurality of detectors (that is, in a similar format and in similar environmental conditions) that are objects of statistical evaluation provided in a predetermined facility. Detectors of any group of detectors grouped for each detector) measured values during normal operation after the first periodic inspection in the facility, and the plurality of detectors in the facility An actual measurement value during a normal operation before the second periodic inspection, which is a periodic inspection next to the first periodic inspection, an inspection result after drift adjustment in the first periodic inspection of the plurality of detectors, and Input means for inputting inspection results before drift adjustment at the time of the second periodic inspection of a plurality of detectors;
First storage means for storing the actual measurement value during the normal operation after the first periodic inspection and the actual measurement value during the normal operation before the second periodic inspection, which are input by the input means;
Second storage means for storing the inspection result after drift adjustment at the time of the first periodic inspection and the inspection result before drift adjustment at the time of the second periodic inspection, which are input by the input means;
Statistically processing the actual measurement value during the normal operation after the first periodic inspection and the actual measurement value during the normal operation before the second periodic inspection stored in the first storage means A statistical amount of the drift amount of the plurality of detectors generated during the normal operation, and the actual measurement value during the normal operation after the first periodic inspection stored in the first storage means and the The actual measurement value during the normal operation before the second periodic inspection, the inspection result after the drift adjustment in the first periodic inspection stored in the second storage means, and the second periodic inspection Statistically processing the inspection results before drift adjustment, and a statistic evaluation means for obtaining a statistic of the drift amount of the plurality of detectors generated during start / stop operation of the equipment,
Output means for outputting the statistical amount of the drift amount during the normal operation and the statistical amount of the drift amount during the start / stop operation obtained by the statistical amount evaluation means.

The detector calibration support apparatus according to the second aspect of the present invention is the detector calibration support apparatus according to the first aspect of the present invention, based on the statistic of the drift amount during the normal operation obtained by the statistic evaluation means. Adding a prediction means for predicting an increase in the drift amount of the plurality of detectors that occurs during normal operation in the case of extending the normal operation period;
The output means is also configured to output the prediction result of the prediction means.

The detector calibration support apparatus according to the third aspect of the present invention is the detector calibration support apparatus according to the first aspect of the present invention, wherein the input means also inputs detection signals of the plurality of detectors.
A true value estimation model for estimating a true value is used, a true value is estimated based on an actual measurement value of the detection signal input by the input means, and an uncertainty of the estimated true value is estimated by the statistic evaluation means A true value estimating means for calculating based on the statistic of the drift amount during the normal operation obtained in
The output means outputs the estimated true value obtained by the true value estimating means and the uncertainty of the estimated true value.

A detector calibration support apparatus according to a fourth aspect of the present invention is the detector calibration support apparatus according to the third aspect of the present invention, wherein the statistic of the drift amount during the normal operation obtained by the statistic evaluation means and the true value are calculated. Based on the uncertainty of the estimated true value obtained by the estimation means, adding a prediction means for predicting an increase in the drift amount of the plurality of detectors that occurs during the normal operation when extending the period of the normal operation,
The output means is also configured to output the prediction result of the prediction means.

The detector calibration support apparatus according to the fifth aspect of the present invention is the detection means according to the third or fourth aspect, wherein the plurality of facility-side input means inputs the detection signals of the detectors provided in the plurality of facilities, respectively. Connected to the input means of the calibration support device of the device through the communication means,
The input means of the detector calibration support apparatus according to the third or fourth aspect of the invention is characterized in that the detection signals of the detector are input from the plurality of equipment-side input means via the communication means. To do.

The detector calibration support apparatus of the sixth invention is the detector calibration support apparatus of the third, fourth or fifth invention,
The estimated true value obtained by the true value estimating means is used for monitoring or control in the facility.

The detector calibration support method according to the seventh aspect of the present invention includes an actual measurement value during normal operation after the first periodic inspection of the plurality of detectors to be subjected to statistical evaluation provided in a predetermined facility. And statistically processing the measured values during the normal operation before the second periodic inspection, which is the periodic inspection following the first periodic inspection in the equipment of the plurality of detectors, and occur during the normal operation. Obtaining a statistic of the drift amount of the plurality of detectors;
And the actual measurement value during the normal operation after the first periodic inspection, the actual measurement value during the normal operation before the second periodic inspection, and the drift during the first periodic inspection of the plurality of detectors. The inspection results after adjustment and the inspection results before drift adjustment at the time of the second periodic inspection of the plurality of detectors are statistically processed, and the plurality of detectors generated during the start / stop operation of the equipment It is characterized in that a statistical amount of drift is obtained.

  The detector calibration support method according to the eighth aspect of the present invention is the detector calibration support method according to the seventh aspect, wherein the normal operation period is extended based on the statistics of the drift amount during the normal operation. An increase in the drift amount of the plurality of detectors occurring during the normal operation is predicted.

  A detector calibration support method according to a ninth aspect of the present invention is the detector calibration support method according to the seventh aspect of the present invention, wherein a true value estimation model for estimating a true value is used and the detection signals of the plurality of detectors are actually measured. A true value is estimated based on the value, and an uncertainty of the estimated true value is calculated based on a statistic of the drift amount during the normal operation.

  A detector calibration support method according to a tenth aspect of the present invention is the detector calibration support method according to the ninth aspect of the present invention, based on the statistics of the drift amount during the normal operation and the uncertainty of the estimated true value, An increase in the drift amount of the plurality of detectors occurring during the normal operation when the normal operation period is extended is predicted.

  The detector calibration support method according to the eleventh aspect of the present invention provides the detection signal from the input means on the plurality of facilities side through which the detection signals of the detectors provided in each of the plurality of facilities are input via communication means. The detector calibration support method of the ninth or tenth invention is implemented by inputting.

The detector calibration support method of the twelfth aspect of the invention is the detector calibration support method of the ninth, tenth or eleventh aspect of the invention,
The estimated true value is used for monitoring or control in the facility.

  According to the detector calibration support apparatus of the first invention or the detector calibration support method of the seventh invention, the actual measurement values during normal operation after the first periodic inspection of a plurality of detectors to be subjected to statistical evaluation And statistically processing the actual measurement values during normal operation before the second periodic inspection of the plurality of detectors to obtain a statistic of drift amounts of the plurality of detectors that occur during normal operation, and The actual measurement value during normal operation after the first periodic inspection, the actual measurement value during normal operation before the second periodic inspection, and the inspection result after drift adjustment during the first periodic inspection of the plurality of detectors. And statistically processing the inspection results before drift adjustment at the time of the second periodic inspection of the plurality of detectors to obtain a statistic of the drift amount of the plurality of detectors generated during start / stop operation. In order to make it a feature, that is, drift generated during start / stop operation To determine and distinguish between statistics and typically occurs drift amount of statistics during operation, it can be appropriately estimated without normally overestimate such occurrence drift amount during operation. That is, each statistic can be used strictly according to the purpose, and the accuracy of evaluation of the drift amount can be improved. For example, when considering the extension of the normal operation period, it is possible to predict an increase in the drift amount during normal operation based on the statistics of the amount of drift generated during normal operation. Appropriate without overestimating the trend of increasing drift in the future compared to when extending the operation period (based on statistics of the drift amount generated for one operation cycle) without distinguishing from operation It becomes possible to evaluate.

  According to the detector calibration support apparatus of the second aspect of the invention or the calibration support method of the detector of the eighth aspect of the present invention, the normal operation period when the normal operation period is extended based on the statistics of the generated drift amount during normal operation. In order to predict the increase in drift amount of multiple detectors that occur during operation, it is not necessary to distinguish between start operation, stop operation and normal operation (based on the statistics of the amount of drift generated for one operation cycle). ) Compared with the case of considering the extension of the operation period, it is possible to evaluate appropriately without overestimating the tendency of future drift increase. For this reason, the number of detectors that can be evaluated as calibration unnecessary (drift adjustment unnecessary) is increased (that is, the frequency of detector calibration (drift adjustment) is reduced), and whether the operation period can be extended from the viewpoint of detector calibration ( How much can be extended).

  According to the detector calibration support apparatus of the third invention or the detector calibration support method of the ninth invention, the true value estimation model for estimating the true value is used, and the actual measurement values of the detection signals of a plurality of detectors are used. A true value is estimated on the basis of this, and the uncertainty of the estimated true value is calculated based on the statistics of the drift amount during normal operation. Compared to the case where the uncertainty of the estimated true value is calculated without distinction (based on the statistic of the generated drift amount for one driving cycle), the estimated true value estimation section (the estimated true value taking into account the uncertainty) Range) can be reduced, and the amount of drift generated during normal operation can be evaluated and estimated more accurately.

  According to the detector calibration support apparatus of the fourth aspect of the invention or the calibration support method of the detector of the tenth aspect of the present invention, based on the statistics of the generated drift amount during normal operation and the uncertainty of the estimated true value, In order to predict an increase in the drift amount of a plurality of detectors that occurs during normal operation when extending the period, true value estimation (true value estimation means) and drift amount prediction (prediction means) By combining it, it is possible to evaluate the estimation interval of the drift amount more strictly (accurately), so the number of detectors that can be evaluated as calibration unnecessary (drift adjustment unnecessary) can be further increased, and whether the operation period can be extended (how much Can be extended more appropriately.

  According to the detector calibration support apparatus of the fifth aspect of the invention or the detector calibration support method of the eleventh aspect of the invention, the plurality of facility-side input means for inputting the detection signals of the detectors provided in the plurality of facilities, respectively. Since the detection signal is input through the communication means, the soundness of the detector can be evaluated with high reliability at a remote monitoring station or the like. In addition, a regular inspection plan can be formulated at a remote location. Furthermore, an effect that it is possible to deal with a plurality of facilities at one place is obtained, and efficient calibration support can be provided in terms of facility costs and labor costs.

  According to the detector calibration support apparatus of the sixth aspect of the invention or the detector calibration support method of the twelfth aspect of the invention, the estimated true value is used for monitoring or control in the facility. Further improvement can be achieved.

  Embodiments of the present invention will be described below in detail with reference to the drawings.

<Embodiment 1>
FIG. 1 is a block diagram showing the configuration of a detector calibration support apparatus according to Embodiment 1 of the present invention, FIG. 2 is a flowchart showing a processing flow of the calibration support apparatus, and FIG. 3 is an operation cycle of a nuclear power plant. FIG. 4 is a graph showing an example of an actual measurement value of a detector during normal operation of the nuclear power plant, and FIG. 5 is a graph showing an example of an inspection result at a periodic inspection in the nuclear power plant.

  As shown in FIG. 1, the detector calibration support apparatus 10 according to the first embodiment includes an input unit 11, a first database 12 as a first storage unit, and a second database as a second storage unit. Database 13, statistic evaluation means 14, and output means 17. The calibration support apparatus 10 is configured by a computer such as a personal computer, and is installed in a nuclear power plant or other appropriate place (offline) to support calibration of a detector provided in the nuclear power plant. Used for.

  The input means 11 is an interface connected to an input device such as a keyboard (not shown), and during normal operation after the first periodic inspection in the nuclear power plant of a plurality of detectors to be subjected to statistical evaluation of the nuclear power plant. Output (actual value) of the plurality of detectors in the normal operation before the second periodic inspection, which is a periodic inspection following the first periodic inspection in the nuclear power plant, Inspection results after drift adjustment (drift amount) at the time of the first periodic inspection of a plurality of detectors, and inspection results (drift amount) before drift adjustment at the time of the second periodic inspection of the plurality of detectors Enter each. Note that these data are manually input by an operator using the input device, for example.

  As shown in FIG. 3, the operation state of the nuclear power plant includes the state of the start-up operation from the stop state until the electric load (output) reaches 100% (normal operation) (the start-up operation period is, for example, 1 to About 2 weeks), and a state of normal operation in which the electric load (output) is kept constant at 100% (currently, the normal operation period of the pressurized water nuclear power plant (PWR) is 13 months), There is a stop operation state (a stop operation period is, for example, about 1 to 2 weeks) from the normal operation state until the electric load (output) starts to be reduced to a stop state. And such an operating state (start-up operation, normal operation and stop operation) has a period of a periodic inspection (for example, a stop period of 40 to 50 days) in which a calibration test of the detector and inspection of other devices are performed. It is an operation cycle that is repeated between the two. FIG. 3 illustrates the n-th periodic inspection, the next n + 1-th periodic inspection, and the operating state before and after these periodic inspections.

  The above-mentioned first periodic inspection means any one periodic inspection (for example, n-th periodic inspection) or multiple times (for example, 1st to n-th) periodic inspection, and the first periodic inspection The second periodic inspection that is the next periodic inspection is a periodic inspection (for example, the (n + 1) th periodic inspection) or a plurality of times (for example, the second to the (n + 1) th) following the one of the periodic inspections. Means periodic inspection. The plurality of detectors to be subjected to the statistic evaluation means detectors of each group grouped for a plurality of detectors used in a similar format and under similar environmental conditions. Examples of the detector type include a capacitance type, a force balance type, and a semiconductor type. As a plurality of detectors used in a similar format and under similar environmental conditions, for example, a water level detector, a flow rate detector, a pressure detector multiplexed (such as quadruple) to ensure high reliability. Etc. It should be noted that which detectors and which detectors actually belong to the same group may be appropriately set according to detector specifications or experiments.

In addition, the measured values during normal operation after the first periodic inspection of the plurality of detectors described above are, for example, the time point B in FIG. 3, that is, during normal operation after the nth periodic inspection (from startup operation to normal operation). Are the outputs (actually measured values) of a plurality of detectors at the time immediately after. For example, the measured values are exemplified in FIG. FIG. 4 illustrates measured values of the quadruple detectors X ₁ , X ₂ , X ₃ , and X ₄ . Similarly, the measured values during normal operation before the second periodic inspection of the plurality of detectors are, for example, at time C in FIG. 3, that is, during normal operation before the (n + 1) th periodic inspection (from normal operation to stop operation). Is the output (actual value) of a plurality of detectors at the time immediately before.

Also, the inspection result (drift amount) after the drift adjustment at the time of the first periodic inspection of the plurality of detectors described above is, for example, time A in FIG. 3 (time immediately before starting the startup operation), that is, the first time. The second detector drift amount test result for confirmation after the detector drift adjustment based on the test result of the detector drift amount (deviation amount of the actual measured value of the detector with respect to the true value) It is. The inspection result (drift amount) before adjustment at the time of the second periodic inspection of the plurality of detectors is, for example, time D in FIG. 3 (time immediately after the stop operation is finished), that is, drift of the first detector. It is the said test result before performing drift adjustment of a detector based on the test result of quantity. FIG. 5 illustrates the inspection results (drift amounts Z ₁ , Z ₂ , Z ₃ , Z ₄ ) before drift adjustment of the quadruple detectors X ₁ , X ₂ , X ₃ , X ₄ . Of course, in the inspection result after the drift adjustment, the drift amount is smaller than in the illustrated example. In the inspection of the detector during the periodic inspection, the full range of 0 to 100% is divided into five stages (or six stages), and the drift amount of each detector is inspected at each point. In FIG. 5, the drift amount of each detector X ₁ , X ₂ , X ₃ , X ₄ is divided into 5 steps (0%, 25%, 50%, 75%, 100%) between 0% to 100% full range. The inspection result when inspecting is illustrated.

  In addition, the input unit 11 designates to the statistic evaluation unit 14 which statistic evaluation is to be performed based on the actual measurement value in the normal operation and the periodic inspection result in which operation cycle with respect to which group of detectors. . That is, the input unit 11 designates the evaluation target to the statistic evaluation unit 14. For example, the operator manually inputs the evaluation target into the input unit 11 using the input device.

  The first database 12 is a storage unit that stores data for statistical evaluation, and is an actual value that is input by the input unit 11 during normal operation after the first periodic inspection, and before the second periodic inspection. The actual measured value during normal operation is stored and stored. The second database 13 is also a storage means for storing data for statistic evaluation, the inspection result after the drift adjustment at the time of the first periodic inspection inputted by the input means 11, and the time of the second periodic inspection. Stores and stores the inspection result before the drift adjustment at. The first database 12 and the second database 13 are configured by a storage device such as a hard disk.

Then, the statistic evaluation unit 14 performs normal operation after the first periodic inspection of the plurality of detectors to be evaluated, extracted from the first database 12 based on the designation of the evaluation target from the input unit 11. Statistically processing the actual measurement value and the actual measurement value of the plurality of detectors during the normal operation before the second periodic inspection, and calculating a statistic (drift amount) of the plurality of detectors generated during the normal operation. Dispersion σ ₂ ² and standard deviation σ ₂ , which are indices indicating the degree of variation in the distribution, and average μ _2, which is an index representing the center of the distribution. Further, in the statistic evaluation means 14, the actual measurement values during normal operation after the first periodic inspection of the plurality of detectors extracted from the first database 12 based on the designation of the evaluation object from the input means 11 and Measured values during normal operation before the second periodic inspection of the plurality of detectors, inspection results after drift adjustment at the first periodic inspection of the plurality of detectors extracted from the second database 13, and the Statistically processing the drift amount of the plurality of detectors generated during start-up / stop operation of the nuclear power plant by statistically processing the inspection results before the drift adjustment at the second periodic inspection of the plurality of detectors. Dispersion σ _e ² and standard deviation σ _e , which are indices indicating the degree of variation in the quantity distribution, and average μ _e, which is an index representing the center of the distribution, are obtained. The statistic evaluation means 14 includes a memory that stores a statistical processing program, a CPU that executes the statistical processing program, and the like. Details of the statistical processing in the statistic evaluation means 14 will be described later.

In the statistic evaluation means 14, the statistic of the drift amount during start-up operation (such as variance σ ₁ ² ) and the statistic of the drift amount during stop operation (such as variance σ ₃ ² ) are individually obtained. Without determining the statistic of the amount of drift (such as variance σ _e ² ) during start / stop operation that combines start-up operation and stop-operation, is based on the following knowledge.

That is, the following equation ( ¹⁾ is used by using dispersions σ ₁ ² , σ ₂ ² , and σ ₃ ² of drift amounts in the start-up operation, normal operation, and stop operation from the n-th periodic inspection to the (n + 1) -th periodic inspection shown in FIG. When the variance σ ² of the overall drift amount is obtained from 1), the variance σ ² obtained from this equation (1) is the inspection result (drift amount) after the drift adjustment in the n-th periodic inspection and the n + 1-th periodic interval. The variance (σ ² ) of the overall drift amount (for one operating cycle) obtained from the inspection result (drift amount) before the drift adjustment in the inspection did not coincide with the variance σ ² . Therefore, as a result of investigating the drift characteristics at the start operation and the drift characteristics at the stop operation, they are not independent, but are related to each other.If the drift amount increases during the start operation, the drift amount decreases during the stop operation, It has been found that there is a relationship in which if the drift amount decreases during start-up operation, the drift amount increases during stop operation. Therefore, the variance σ _e ² of the generated drift amount at the start / stop operation combined with the start operation and the stop operation is obtained, and this variance σ _e ² and the variance σ ₂ ² of the drift amount at the normal operation are used. obtains the variance sigma ² overall drift amount from the following equation (2), and variance sigma ² obtained from the equation (2), the inspection results after drift adjustment in the n times periodic inspection (drift amount) and the n + 1 As a result of comparing the overall drift amount variance σ ² obtained from the inspection result (drift amount) before the drift adjustment in the periodic inspection, the ^two values were almost the same.

In addition, when comparing the standard deviation σ ₂ and variance σ ₂ ² of the drift amount during normal operation with the standard deviation σ _e and variance σ _e ² of the drift amount during start / stop operation, the former is compared to the latter It was also found that the value would be small. This is because the detector environment (temperature and pressure of the process being measured, etc.) is relatively large during start-up and stop operations, and a relatively large drift occurs during normal operation. This is probably because the environment of the detector hardly changes, so that there is not much drift. In [Table 1], standard deviations σ _e and σ ₂ were obtained as statistics of each drift amount during start / stop operation and normal operation for a detector of a certain type (form) A, B, and C. An example is shown. From [Table 1], although the difference varies depending on the type of detector, the standard deviation σ ₂ of the drift amount during normal operation is greater for each type A, B, C detector. It can be seen that the standard deviation σ _e of the drift amount during the stop operation is smaller.

The output means 17 is a statistic (standard deviation σ ₂ , variance σ ₂ ² , average μ ₂ ) of the generated drift amount during normal operation obtained by the statistic evaluation means 14 and statistics of the generated drift amount during start / stop operation. The quantity (standard deviation σ _e , variance σ _e ² , average μ _e ) is output to, for example, a display (monitor) or a printer (not shown).

The processing flow of the calibration support apparatus 10 will be described with reference to FIG. 2. In the first embodiment, first, the input unit 11 designates an evaluation target for the statistic evaluation unit 14 (step S11). As a result, in the statistic evaluation means 14, based on the designation of the evaluation target from the input means 11, during the normal operation after the first periodic inspection of the plurality of detectors that are the statistic evaluation targets extracted from the first database 12. And the statistical values of the drift amounts of the plurality of detectors generated during the normal operation (standard) by statistically processing the actual measurement values of the plurality of detectors and the actual measurement values during the normal operation before the second periodic inspection of the plurality of detectors. (Deviation σ ₂ , variance σ ₂ ² , average μ ₂ ), and the measured values during normal operation after the first periodic inspection of the plurality of detectors to be statistically evaluated extracted from the first database 12 And the actual measurement values during normal operation before the second periodic inspection of the plurality of detectors, the inspection results after the drift adjustment at the first periodic inspection of the plurality of detectors extracted from the second database 13, and During a second periodic inspection of the plurality of detectors It delivers the inspection result of the previous drift adjustment is treated statistically to determine the drift amount of statistics of the plurality of detectors that occurs when starting and stopping operation (standard deviation sigma _e, variance sigma _e ^2, mean mu _e) (Steps S12, 13, 14). Then, the output means 17 calculates the statistics (standard deviation σ ₂ , variance σ ₂ ² , average μ ₂ ) of the generated drift amount during the normal operation obtained by the statistic evaluation means 14 and the occurrence during the start / stop operation. The statistics of the drift amount (standard deviation σ _e , variance σ _e ² , average μ _e ) are output (step S15).

  If new actual measurement values and new periodic inspection results are obtained in the future operation cycle (after the evaluation time point), these new actual measurement values and periodic inspection results are also stored in the first database via the input means 11. 12 and stored in the second database 13 and used for statistic evaluation in the statistic evaluation means 14.

  Here, the statistical processing in the statistic evaluation means 14 will be described in detail.

  As described above, the statistic evaluation means 14 calculates the statistic of the generated drift amount during the normal operation and the statistic of the generated drift amount during the start / stop operation in the designated operation cycle. Specifically, it is as follows.

但し，Y₁，Y₂は第ｎ回定期検査後の通常運転時及び第ｎ＋１回定検前の通常運転時のプロセス値の真値とする。
一方，確率変数ＡとＢの間に成立する統計上の定理

において、

と考えれば、多数の多重検出器についてβ_ijを求め、その標準偏差をσ_Rとして

と考えることができる。即ち、各検出器の第ｎ回定検後の通常運転時のドリフト量と第ｎ＋１回定検前の通常運転時のドリフト量の差((X_i2−Y₂)−(X_i1−Y₁))は統計的にσ₂ ²(=σ_R ²／２)のばらつきを持つこととなる。即ち、多数の多重検出器の実測値から通常運転時に発生したドリフト量の標準偏差を算出することができる。 <Statistics of the amount of drift generated during normal operation>
Now, let X ₁₁ , X ₂₁ , X ₃₁ , and X ₄₁ be measured values during normal operation after the nth periodic inspection for a quadruple detector. Further, when the measured values during normal operation before the (n + 1) th periodic inspection are X ₁₂ , X ₂₂ , X ₃₂ , and X ₄₂ , the variation β of the measured values between the multiple detectors is expressed by the following formula (3) to It can be developed as in (6).

Y ₁ and Y ₂ are the true values of the process values during normal operation after the nth periodic inspection and during normal operation before the (n + 1) th regular inspection.
On the other hand, a statistical theorem that holds between random variables A and B

In

If β _ij is obtained for a large number of multiple detectors, and its standard deviation is σ _R

Can be considered. That is, the difference between the drift amount during normal operation after the n-th regular inspection of each detector and the drift amount during normal operation before the (n + 1) -th regular inspection ((X _i2 −Y ₂ ) − (X _i1 −Y ₁ )) is to have a variation in statistically ^{_{σ 2 2 (= σ R 2}} /2). That is, the standard deviation of the drift amount generated during normal operation can be calculated from the measured values of a large number of multiple detectors.

<Statistics of drift amount during start / stop operation>
The standard deviation of the drift amount can be obtained during start / stop operation in the same way as during normal operation. In this case, the inspection result at the n-th periodic inspection (after adjustment), the actual measurement value at the normal operation after the n-th regular inspection, the actual measurement value at the normal operation before the n + 1-th periodic inspection, the n + 1-th periodic inspection By subtracting the difference between multiple detectors that occurred during normal operation using the test results at the time (before adjustment), the amount of drift that occurs only during start / stop operation is obtained, and the standard deviation is calculated.

  In addition, the inspection at the time of regular inspection is divided into 5 steps (or 6 steps) between 0 to 100% full range as described above, and the drift amount is measured at each point. For example, the operation point P) illustrated in FIG. 5 is not inspected. For this reason, the drift amount at the operation point during normal operation (operation point P in FIG. 5) is obtained by interpolation calculation or the like for the drift amount at the five-stage (or six-stage) inspection point and used for statistical evaluation. . Of course, if the operation point at the time of normal operation is inspected, the inspection result (drift amount) of the operation point may be used for the statistic evaluation.

従って、多数の多重検出器についてβ_iSSを求め、その標準偏差をσ_SSとすれば

と考えることができる。即ち，各検出器の起動運転時及び停止運転時に発生したドリフト量は統計的にσ_T ²(=σ_SS ²／２)のばらつきを持つこととなる。即ち、多数の多重検出器の実測値と定期検査の検査結果から起動・停止運転時に発生したドリフト量の標準偏差を算出することができる。 The calculation of the standard deviation will be described in detail. The drift amount of the operation point during normal operation in the inspection result at the n-th periodic inspection (after adjustment) for a certain quadruple detector is Z ₁₀ , Z ₂₀ , Z _30. , Z ₄₀ . In addition, when the drift amount of the operation point during normal operation in the inspection result at the (n + 1) th periodic inspection (before adjustment) is Z ₁₃ , Z ₂₃ , Z ₃₃ , Z ₄₃ , multiple detectors during start / stop operation A change amount β _iSS of the actually measured value between can be developed as in the following equations (11) to (14).

Therefore, if β _iSS is obtained for many multiple detectors and its standard deviation is σ _SS ,

Can be considered. That is, startup operation and drift amount generated when stopping the operation of the detector becomes to have a variation in statistically _{^{σ T 2 (= σ SS 2}} /2). That is, the standard deviation of the drift amount generated during the start / stop operation can be calculated from the actual measurement values of a large number of multiple detectors and the inspection results of the periodic inspection.

  As described above, according to the detector calibration support apparatus or the method thereof according to the first embodiment, after the first periodic inspection of the plurality of detectors to be subjected to the statistic evaluation in the statistic evaluation means 14. Statistically processing the actual measurement value during normal operation and the actual measurement value during normal operation before the second periodic inspection of the plurality of detectors, and calculating drift amounts of the plurality of detectors that occur during normal operation. Statistics are obtained, and measured values during normal operation after the first periodic inspection, measured values during normal operation before the second periodic inspection, and during the first periodic inspection of the plurality of detectors. The drift amount of the plurality of detectors generated during start / stop operation is statistically processed by the inspection result after the drift adjustment in the first and the inspection result before the drift adjustment in the second periodic inspection of the plurality of detectors. Since it is characterized by calculating the statistic of Effect can be obtained.

  That is, in the first embodiment, since the statistical amount of the generated drift amount at the start / stop operation and the statistical amount of the generated drift amount at the normal operation are separately determined, the generated drift amount at the normal operation is excessively large. It can be estimated appropriately without evaluation. That is, each statistic can be used strictly according to the purpose, and the accuracy of evaluation of the drift amount can be improved. For example, when considering the extension of the normal operation period, it is possible to predict an increase in the drift amount during normal operation based on the statistics of the amount of drift generated during normal operation. Appropriate without overestimating the trend of increasing drift in the future compared to when extending the operation period (based on statistics of the drift amount generated for one operation cycle) without distinguishing from operation It becomes possible to evaluate.

<Embodiment 2>
6 is a block diagram showing a configuration of a detector calibration support apparatus according to Embodiment 2 of the present invention, FIG. 7 is a flowchart showing a processing flow of the calibration support apparatus, and FIG. 8 is a prediction of the calibration support apparatus. It is explanatory drawing of the drift amount prediction in a means. In FIG. 6, the same components as those in FIG. 1 are denoted by the same reference numerals, and detailed description thereof is omitted.

  As shown in FIG. 6, the detector calibration support apparatus 20 according to the second embodiment includes an input unit 11, a first database 12 as a first storage unit, and a second database as a second storage unit. Database 13, statistic evaluation means 14, prediction means 25, and output means 27. That is, the calibration support apparatus 20 is characterized in that a prediction means 25 is further added to the calibration support apparatus 10 (see FIG. 1) of the first embodiment. The calibration support apparatus 20 is also configured by a computer such as a personal computer, and is installed in the nuclear power plant or other appropriate location (offline) to support calibration of the detector provided in the nuclear power plant. Used for.

Then, in the prediction means 25, when the period of normal operation is extended based on the statistics (standard deviation σ ₂ ) of the drift amount generated during the normal operation of the plurality of detectors to be evaluated obtained by the statistic evaluation means 14. An increase in the drift amount of the plurality of detectors that occurs during normal operation (for example, when the normal operation period is extended from 13 months to 18 months) is predicted. Note that the prediction means 25 includes a memory that stores a prediction processing program, a CPU that executes the prediction processing program, and the like.

Here, the prediction process of the prediction means 25 will be described in detail. The amount of drift generated at the date and time of evaluation (arbitrary date and time), where T is the period used as the standard for calculating the statistics of the amount of drift generated during normal operation and T _eval is the difference between the starting point of evaluation and the date and time of evaluation The standard deviation σ _i is calculated by the following equation (16). As shown in Expression (16), the standard deviation σ _i of the generated drift amount increases in proportion to the 1/2 power of time. Note that the period T used as a standard for calculating the statistic of the generated drift amount during normal operation is, for example, an actual measurement at the beginning (time B) and end (time C) of the normal operation period of 13 months as illustrated in FIG. If the statistic of the amount of drift generated during normal operation is calculated based on the value, it will be 13 months. In the case of this example, the starting point of the evaluation is the initial stage (time B) of the normal operation period of 13 months. In this example, when the normal operation period is to be extended from 13 months to, for example, 18 months, the difference _Teval between the starting point of the evaluation and the date and time at the time of evaluation is 18 months.

  In general, if ± 3σ is used, the reliability can be evaluated with a reliability of 99.7%. Thus, as shown in FIG. 8, how the drift amount increases with the extension of the normal operation period (the drift amount in the range of ± 3σ). It is possible to predict when the drift amount (the drift amount in the range of ± 3σ) deviates from the allowable range (exceeds the allowable line). Although it is considered that it is most desirable to evaluate in the range of ± 3σ, it is not necessarily limited to this, and for example, evaluation may be performed in the range of ± 2σ or ± 4σ depending on allowable conditions. .

In the output means 27, the statistics (standard deviation σ ₂ , variance σ ₂ ² , average μ ₂ ) of the generated drift amount during normal operation obtained by the statistic evaluation means 14 and the generated drift amount during start / stop operation are calculated. The statistics (standard deviation σ _e , variance σ _e ² , average μ _e ) are output to, for example, a display (monitor) or a printer (not shown), and further, the prediction means 25 increases the amount of drift generated during normal operation. The prediction result (how the drift amount (± 3σ value) increases and when the drift amount (± 3σ increase amount) deviates from the allowable range) is also output to a display (monitor), a printer, or the like.

The processing flow of the calibration support apparatus 20 will be described with reference to FIG. 7. In the second embodiment, first, the input unit 11 designates an evaluation target for the statistic evaluation unit 14 (step S21). As a result, in the statistic evaluation means 14, based on the designation of the evaluation target from the input means 11, during the normal operation after the first periodic inspection of the plurality of detectors that are the statistic evaluation targets extracted from the first database 12. And the statistical values of the drift amounts of the plurality of detectors generated during the normal operation (standard) by statistically processing the actual measurement values of the plurality of detectors and the actual measurement values during the normal operation before the second periodic inspection of the plurality of detectors. Deviation σ ₂ , variance σ ₂ ² , average μ ₂ ), and extracted from the first database 12 and the actual values and the plurality of detections during normal operation after the first periodic inspection of the plurality of detectors. Measured values during normal operation before the second periodic inspection of the detector, the inspection results after the drift adjustment of the plurality of detectors extracted from the second database 13 at the first periodic inspection, and the plurality of detectors Drift adjustment at the second periodic inspection Treated the previous inspection results statistically, the statistics of the drift amount of the plurality of detectors that occurs when starting and stopping operation (standard deviation sigma _e, variance sigma _e ^2, mean mu _e) obtaining the (step S22 23, 24).

Thereafter, the prediction unit 25 generates the normal operation period when the normal operation period is extended based on the statistics (standard deviation σ ₂ ) of the generated drift amount during the normal operation obtained by the statistic evaluation unit 14. An increase in the drift amount of the plurality of detectors is predicted (step S25). Then, the output means 17 calculates the statistics (standard deviation σ ₂ , variance σ ₂ ² , average μ ₂ ) of the generated drift amount during normal operation obtained by the statistic evaluation means 14, and the occurrence detection during start / stop operation. Statistic (standard deviation σ _e , variance σ _e ² , average μ _e ) of the drift amount of the vessel, and further, the prediction result of the increase in the generated drift amount during normal operation in the prediction means 25 is also output (step S26). ).

  As described above, according to the detector calibration support apparatus or the method thereof according to the second embodiment, the normal operation period is extended by the prediction unit 25 based on the statistics of the generated drift amount during the normal operation. In this case, the increase in the drift amount of the plurality of detectors occurring during the normal operation is predicted, so that the start operation and the stop operation are not distinguished from the normal operation (statistics of the generated drift amount for one operation cycle). Compared with the case of considering the extension of the operation period (based on the amount), it is possible to appropriately evaluate without overestimating the tendency of the future drift amount to increase. For this reason, the number of detectors that can be evaluated as calibration unnecessary (drift adjustment unnecessary) is increased (that is, the frequency of detector calibration (drift adjustment) is reduced), and whether the operation period can be extended from the viewpoint of detector calibration ( How much can be extended).

<Embodiment 3>
FIG. 9 is a block diagram showing a configuration of a detector calibration support apparatus according to Embodiment 3 of the present invention, FIG. 10 is a flowchart showing a processing flow of the calibration support apparatus, and FIG. It is explanatory drawing which shows an example of the true value estimation model used with a value estimation means. In FIG. 9, the same parts as those in FIG. 1 are denoted by the same reference numerals, and detailed description thereof is omitted.

  As shown in FIG. 9, the detector calibration support apparatus 30 according to the third embodiment has an input means 11, a first database 12 as a first storage means, and a second database as a second storage means. Database 13, statistic evaluation means 14, true value estimation means 36, and output means 37. That is, the calibration support apparatus 30 is characterized in that a true value estimation means 36 is further added to the calibration support apparatus 10 (see FIG. 1) of the first embodiment. The calibration support apparatus 30 is also configured by a computer such as a personal computer, and is installed in the nuclear power plant (online) and used to support calibration of the detector provided in the nuclear power plant.

  The input means 11 is an interface connected to an input device such as a keyboard (not shown) as in the first embodiment. The nuclear power plant 1 is a plurality of detectors that are subject to statistical evaluation in the nuclear power plant 1. Measured values during normal operation after the first periodic inspection in, and during normal operation before the second periodic inspection, which is the periodic inspection next to the first periodic inspection in the nuclear power plant 1 of the plurality of detectors , Actual inspection value of the plurality of detectors, and inspection result after drift adjustment (drift amount) at the time of the first periodic inspection, and inspection before drift adjustment at the time of the second periodic inspection of the plurality of detectors Enter the result (drift amount). Further, in the third embodiment, the input means 11 is also connected to a large number of detectors provided in the nuclear power plant 1 and also inputs detection signals output from these detectors. The input means 11 sends the detection signal input from each detector to the true value estimation means 36 and the first database 12 as an actual measurement value. In the first database 12, detection signals (actual measurement values) input from the detectors are also stored as statistics evaluation data.

Then, the true value estimation means 36 uses a true value estimation model for estimating the true value, and the true value is based on the detection signals (actual values) of the plurality of detectors to be evaluated by the statistic input inputted by the input means 11. , And the uncertainty of the estimated true value is calculated based on the statistics (standard deviation σ ₂ ) of the drift amounts of the plurality of detectors generated during the normal operation obtained by the statistic evaluation means 14. The estimated drift amount can be calculated from the estimated true value obtained by the true value estimating means 36 and the actually measured value. The true value estimation means 36 includes a memory for storing a programmed true value estimation model, a CPU for executing a program for the true value estimation model, and the like.

  More specifically, the true value estimation model used by the true value estimation means 36 is obtained by adjusting or learning parameters in the model in advance, and is, for example, a linear model or a neural network. Construction of the true value estimation model, that is, adjustment and learning of the estimation model is performed using a detector signal (actual measurement value) obtained in normal operation of the nuclear power plant 1.

  FIG. 11 is a schematic diagram showing the concept of a neural network as an example of a true value estimation model. Since the neural network itself is known, detailed description of the neural network is omitted. In FIG. 11, the four nodes at the left end constitute an input layer, and the four nodes at the right end constitute an output layer. Between the input layer and the output layer is an intermediate layer. In this neural network, when the measured value of the detector signal is input to the input layer, the calculation is performed based on the weight, bias, conversion function, etc. inside the network, and the estimated true value of each detector signal is output from the output layer. The Here, the number of nodes in the input layer matches the number of nodes in the output layer, and the estimated true value for the input signal to the i-th node is output to the i-th node in the output layer. Note that the number of nodes in the input layer, intermediate layer, and output layer varies depending on the number of input signals to the network, and is not limited to the number shown in FIG.

Further, the estimated true value obtained by the true value estimating means 36 has uncertainty. The uncertainties σ _k , _j of the estimated true value Y _k , _j of the detector A _j are the drifts generated in the indicated value X _i of the detector A _i to the estimated true value Y _k , _j via the true value estimation model k. If the influencing sensitivities are f _k and _ij , they can be calculated by the following equation (17).

However, σ _i (that is, σ ₁ , σ ₂ ... Σ _n ) on the right side is a statistic (standard deviation σ ₂ ) of the generated drift amount during normal operation, and increases in proportion to the 1/2 power of time. Become. That is, assuming that the period used as the standard for calculating the statistic of the drift amount during normal operation is T, and the difference between the data date and time used to construct the true value estimation model and the date and time of evaluation is T _now , the following equation (18) Calculate with The period T used as a standard for calculating the statistic of the amount of drift generated during normal operation is, for example, measured values at the beginning (time B) and the end (time C) of the normal operation period of 13 months as illustrated in FIG. If the statistic of the amount of drift generated during normal operation is calculated based on this, it will be 13 months. In the case of this example, the data date and time used for constructing the true value estimation model is the initial value when the true value estimation model is constructed based on, for example, the actual measurement value of the normal operation period of 13 months (time B). Date and time (date and time at time B). In the case of this example, the date and time at the time of evaluation is the date and time at which the true value is estimated by the true value estimating means 36 and the uncertainty of the estimated true value is evaluated (at any point in the normal operation period of 13 months). The difference T _now between the data date and time used for constructing the true value estimation model and the date and time at the time of evaluation is, for example, 13 when the true value is estimated by the true value estimating means 36 and the uncertainty of the estimated true value is evaluated. If five months have passed since the beginning of the normal operation period of the month (time B), it is 5 months.

ここでｆ_kは真値推定モデルｋである。また、記号^Tは転置を意味する。
そして、検出器Ａ_iの実測値Ｘ_iにおいて発生したドリフトが真値推定モデルｋを経て推定真値Ｙ_k、_jに影響する感度ｆ_k、_ijは、次式（２０）で求めることができる。

Also, the sensitivity f _k, will be described more in detail _ij, now, detector A _1, · · ·, measured values X ₁ of A _n, · · ·, the X _n was vector representation, X = [X _1, X ₂ ,... X _n ] ^T , the estimated value Y _k = [Y _k , 1, Y _k , 2,..., Y _k , _n ] ^T obtained by the true value estimation model k is expressed by the following equation (19). .

Here, f _k is a true value estimation model k. The symbol ^T means transposition.
The sensitivities f _k and _ij that the drift generated in the actual measurement value X _i of the detector A _i affects the estimated true values Y _k and _j through the true value estimation model k can be obtained by the following equation (20). .

In the output means 37, the statistics (standard deviation σ ₂ , variance σ ₂ ² , average μ ₂ ) of the generated drift amount at the normal operation obtained by the statistic evaluation means 14 and the generated drift amount at the start / stop operation are calculated. Statistics (standard deviation σ _e , variance σ _e ² , average μ _e ) are output to, for example, a display (monitor) or a printer (not shown), and further, an estimated true value or an estimated true value obtained by the true value estimating means 36. Uncertainty is also output to a display (monitor) or printer.

The processing flow of the calibration support apparatus 30 will be described with reference to FIG. 10. In the third embodiment, first, the input unit 11 captures a detection signal (actual measurement value) from a detector provided in the nuclear power plant 1. (Step S31). Further, the input means 11 designates an evaluation target for the statistic evaluation means 14. As a result, in the statistic evaluation means 14, based on the designation of the evaluation target from the input means 11, during the normal operation after the first periodic inspection of the plurality of detectors that are the statistic evaluation targets extracted from the first database 12. And the statistical values of the drift amounts of the plurality of detectors generated during the normal operation (standard) by statistically processing the actual measurement values of the plurality of detectors and the actual measurement values during the normal operation before the second periodic inspection of the plurality of detectors. Deviation σ ₂ , variance σ ₂ ² , average μ ₂ ), and extracted from the first database 12 and the actual values and the plurality of detections during normal operation after the first periodic inspection of the plurality of detectors. Measured values during normal operation before the second periodic inspection of the detector, the inspection results after the drift adjustment of the plurality of detectors extracted from the second database 13 at the first periodic inspection, and the plurality of detectors Drift adjustment at the second periodic inspection Treated the previous inspection results statistically, the statistics of the drift amount of the plurality of detectors that occurs when starting and stopping operation (standard deviation sigma _e, variance sigma _e ^2, mean mu _e) obtaining the (step S32 33, 34).

Thereafter, the true value estimating means 36 estimates the true value based on the actual measurement value of the detection signal input by the input means 11 using the true value estimation model for estimating the true value, and this estimated true value. Is calculated based on the statistic (standard deviation σ ₂ ) of the generated drift amount during the normal operation obtained by the statistic evaluation means 14 (step S35). Then, the output means 37 calculates the statistics (standard deviation σ ₂ , variance σ ₂ ² , average μ ₂ ) of the generated drift amount during normal operation obtained by the statistic evaluation means 14 and the generated drift during start / stop operation. The quantity statistics (standard deviation σ _e , variance σ _e ² , average μ _e ) are output, and the estimated true value obtained by the true value estimating means 36 and the uncertainty of the estimated true value are also output (step S36).

  As described above, according to the detector calibration support apparatus or method of the third embodiment, the true value estimation means 36 uses a true value estimation model for estimating the true value, and a plurality of detectors. The true value is estimated based on the actual measurement value of the detection signal, and the uncertainty of the estimated true value is calculated based on the statistic of the drift amount during normal operation. Compared to calculating the estimated true value uncertainty (based on the statistics of the drift amount generated for one operating cycle) without distinguishing between stop operation and normal operation, the estimated true value estimation interval (uncertainty) The range of the estimated true value in consideration of the above can be reduced, and the drift amount generated during normal operation can be evaluated and estimated more accurately.

<Embodiment 4>
FIG. 12 is a block diagram showing a configuration of a detector calibration support apparatus according to Embodiment 4 of the present invention, FIG. 13 is a flowchart showing a processing flow of the calibration support apparatus, and FIG. It is a block diagram which shows the example of application of the estimation true value obtained by a value estimation means. In FIG. 12, the same components as those in FIG. 1 are denoted by the same reference numerals, and detailed description thereof is omitted.

  As shown in FIG. 12, the detector calibration support apparatus 40 according to the fourth embodiment includes an input means 11, a first database 12 as a first storage means, and a second database as a second storage means. Database 13, statistic evaluation means 14, prediction means 45, true value estimation means 46, and output means 47. That is, the calibration support apparatus 40 is characterized in that a prediction means 45 and a true value estimation means 46 are further added to the calibration support apparatus 10 (see FIG. 1) of the first embodiment. Furthermore, the calibration support apparatus 40 is characterized in that a prediction means 45 is further added to the calibration support apparatus 30 (see FIG. 9) of the third embodiment. The calibration support device 40 is also configured by a computer such as a personal computer, and is installed in the nuclear power plant (online) and is used to support calibration of the detector provided in the nuclear power plant.

  The input means 11 is an interface connected to an input device such as a keyboard (not shown) as in the first embodiment. The nuclear power plant 1 is a plurality of detectors that are subject to statistical evaluation in the nuclear power plant 1. Measured values during normal operation after the first periodic inspection in, and during normal operation before the second periodic inspection, which is the periodic inspection next to the first periodic inspection in the nuclear power plant 1 of the plurality of detectors , Actual inspection value of the plurality of detectors, and inspection result after drift adjustment (drift amount) at the time of the first periodic inspection, and inspection before drift adjustment at the time of the second periodic inspection of the plurality of detectors Enter the result (drift amount). Further, in the fourth embodiment, the input means 11 is also connected to a number of detectors provided in the nuclear power plant 1, and the detection signals output from these detectors are also input. The input means 11 sends the detection signal input from each detector to the true value estimation means 46 and the first database 12 as an actual measurement value. In the first database 12, detection signals (actual measurement values) input from the detectors are also stored as statistics evaluation data.

但し、右辺のσ_i（即ちσ₁，σ₂・・・σ_n）は通常運転時の発生ドリフト量の統計量（標準偏差σ₂）であり，時間の１／２乗に比例して大きくなる。即ち、通常運転時の発生ドリフト量の統計量算定の基準とした期間をＴとし、真値推定モデル構築に使用したデータ日時と評価時点の日時の差をＴ_nowとすると、次式（２２）で計算する。通常運転時の発生ドリフト量の統計量算定の基準とした期間Ｔとは、例えば図３に例示するように１３箇月の通常運転期間の初期（Ｂ時点）と終期（Ｃ時点）の実測値に基づいて通常運転時の発生ドリフト量の統計量を算出した場合には１３箇月とする。また、この例の場合、真値推定モデル構築に使用したデータ日時とは、１３箇月の通常運転期間の例えば初期（Ｂ時点）の実測値に基づいて真値推定モデルを構築したときには当該初期の日時（Ｂ時点の日時）となる。また、この例の場合、評価時点の日時とは真値推定手段４６で真値を推定して当該推定真値の不確かさを評価する日時（１３箇月の通常運転期間の何れかの時点）であり、真値推定モデル構築に使用したデータ日時と評価時点の日時の差Ｔ_nowとは、真値推定手段４６で真値を推定して当該推定真値の不確かさを評価した日時が例えば１３箇月の通常運転期間の初期（Ｂ時点）から５箇月経過した時点であるとすると、５箇月である。

The true value estimating means 46 uses a true value estimation model for estimating the true value, estimates the true value based on the actual value of the detection signal input by the input means 11, and is uncertain about the estimated true value. This is calculated based on the statistic (standard deviation σ ₂ ) of the drift amounts of the plurality of detectors generated during the normal operation obtained by the statistic evaluation means 14. The true value estimating means 46 is the same as the true value estimating means 36 (see FIG. 9) provided in the calibration support apparatus 30 of the third embodiment. As described in the third embodiment, the estimated true value obtained by the true value estimating means 46 has uncertainty. The uncertainties σ _k , _j of the estimated true values Y _k , _j of the detector A _j are the drifts generated in the indicated values X _i of the detectors A _i to the estimated true values Y _k , _j via the true value estimation model k. If the influencing sensitivities are f _k and _ij , they can be calculated by the following equation (21).

However, σ _i (that is, σ ₁ , σ ₂ ... Σ _n ) on the right side is a statistical amount (standard deviation σ ₂ ) of the generated drift amount during normal operation, and increases in proportion to the 1/2 power of time. Become. That is, assuming that the period used as the standard for calculating the statistic of the drift amount during normal operation is T, and the difference between the data date and time used to construct the true value estimation model and the date and time at the time of evaluation is T _now , the following equation (22) Calculate with The period T used as a standard for calculating the statistic of the amount of drift generated during normal operation is, for example, measured values at the beginning (time B) and the end (time C) of the normal operation period of 13 months as illustrated in FIG. If the statistic of the amount of drift generated during normal operation is calculated based on this, it will be 13 months. In the case of this example, the data date and time used for constructing the true value estimation model is the initial value when the true value estimation model is constructed based on, for example, the actual measurement value of the normal operation period of 13 months, for example, the initial (time B). Date and time (date and time at time B). In the case of this example, the date and time at the time of evaluation is the date and time at which the true value is estimated by the true value estimating means 46 and the uncertainty of the estimated true value is evaluated (at any point in the normal operation period of 13 months). Yes, the difference T _now between the data date and time used to construct the true value estimation model and the date and time at the time of evaluation is, for example, 13 when the true value is estimated by the true value estimating means 46 and the uncertainty of the estimated true value is evaluated. If five months have passed since the beginning of the normal operation period of the month (time B), it is 5 months.

Then, in the predicting means 45, the statistics of the drift amounts (standard deviation σ ₂ ) of the plurality of detectors generated during the normal operation obtained by the statistic evaluating means 14 and the estimated true value obtained by the true value estimating means 46 are calculated. Based on the uncertainty, an increase in the drift amount of the plurality of detectors occurring during the normal operation when the normal operation period is extended (for example, when the normal operation period is extended from 13 months to 18 months) is predicted. Note that the prediction unit 45 includes a memory that stores a prediction processing program, a CPU that executes the prediction processing program, and the like. Here, the prediction process of the prediction means 45 will be described in detail. Assuming that the future date and time to be evaluated is T _future, the uncertainty at this date and time T _future can be obtained by the following equation (23) in consideration of the current estimated true value uncertainty and the future drift amount during normal operation. it can. That is, as shown in FIG. 8, it is possible to predict how the drift amount (± 3σ value) increases and when the drift amount (± 3σ increase amount) deviates from the allowable range. In the case of the above example, if the normal operation period is to be extended from 13 months to, for example, 18 months, T _future is 18 months.

In the output means 47, the statistics (standard deviation σ ₂ , variance σ ₂ ² , average μ ₂ ) of the generated drift amount during normal operation obtained by the statistic evaluation means 14 and the generated drift amount during start / stop operation are calculated. The statistics (standard deviation σ _e , variance σ _e ² , average μ _e ) are output to, for example, a display (monitor) or a printer (not shown), and further the estimated true value and the estimated true value obtained by the true value estimating means 46 Uncertainty of value and prediction result of increase in generated drift amount during normal operation in predicting means 45 (how drift amount (± 3σ value) increases, when drift amount (± 3σ increase amount)) Is out of the permissible range) is also output to a display (monitor) or a printer.

The processing flow of the calibration support apparatus 40 will be described with reference to FIG. 13. In the fourth embodiment, first, the input unit 11 captures a detection signal (actual value) from a detector provided in the nuclear power plant 1. (Step S41). Further, the input means 11 designates an evaluation target for the statistic evaluation means 14. As a result, in the statistic evaluation means 14, based on the designation of the evaluation target from the input means 11, during the normal operation after the first periodic inspection of the plurality of detectors that are the statistic evaluation targets extracted from the first database 12. And the statistical values of the drift amounts of the plurality of detectors generated during the normal operation (standard) by statistically processing the actual measurement values of the plurality of detectors and the actual measurement values during the normal operation before the second periodic inspection of the plurality of detectors. (Deviation σ ₂ , variance σ ₂ ² , average μ ₂ ), and the measured values during normal operation after the first periodic inspection of the plurality of detectors to be statistically evaluated extracted from the first database 12 And the actual measurement values during normal operation before the second periodic inspection of the plurality of detectors, the inspection results after the drift adjustment at the first periodic inspection of the plurality of detectors extracted from the second database 13, and During a second periodic inspection of the plurality of detectors It delivers the inspection result of the previous drift adjustment is treated statistically to determine the drift amount of statistics of the plurality of detectors that occurs when starting and stopping operation (standard deviation sigma _e, variance sigma _e ^2, mean mu _e) (Steps S42, 43, 44).

Thereafter, the true value estimation means 46 estimates the true value based on the actual measurement value of the detection signal input by the input means 11 using the true value estimation model for estimating the true value, and this estimated true value Is calculated based on the statistic (standard deviation σ ₂ ) of the generated drift amount during normal operation obtained by the statistic evaluation means 14 (step S45). Subsequently, the predicting means 45 is based on the statistical amount (standard deviation σ ₂ ) of the generated drift amount during normal operation obtained by the statistic evaluating means 14 and the uncertainty of the estimated true value obtained by the true value estimating means 46. Thus, an increase in the drift amount of the plurality of detectors that occurs during the normal operation when the normal operation period is extended is predicted. Then, the output means 47 calculates the statistics (standard deviation σ ₂ , variance σ ₂ ² , average μ ₂ ) of the generated drift amount during normal operation obtained by the statistic evaluation means 14 and the generated drift during start / stop operation. Statistic of quantity (standard deviation σ _e , variance σ _e ² , average μ _e ), and further, the estimated true value obtained by the true value estimating means 46 and the uncertainty of the estimated true value, and the predicting means 45 The predicted result of the increase in the amount of drift generated during normal operation is also output.

  The estimated true values output from the output means 47 and 37 of the calibration support apparatus 40 of the fourth embodiment and the calibration support apparatus 30 of the third embodiment can also be used for monitoring and control. FIG. 15 shows an example in which the estimated true value is used for control. The control device 90 shown in the figure has a calibration processing function 91 and various control functions 92. The control device 90 performs predetermined control (for example, pressure control, flow rate control, water level control, etc.) based on the detection signal (actual value) of the detector 81. At this time, the calibration processing function 91 performs calibration. The detection signal (actual value) of the detector 81 is calibrated based on the estimated true value input from the support device 40 or 30, and the various control functions 92 perform predetermined control (based on the calibrated detection signal (actual value). For example, pressure control, flow rate control, water level control, etc.) are performed.

  As described above, according to the detector calibration support apparatus or method of the fourth embodiment, the prediction means 45 and the true value estimation means 46 are provided, and the prediction means 45 generates during normal operation. Based on the statistics of the drift amount and the uncertainty of the estimated true value, it is characterized by predicting an increase in the drift amount of a plurality of detectors that occurs during the normal operation when extending the normal operation period, By combining true value estimation (true value estimation means) and drift amount prediction (prediction means), it is possible to evaluate the drift amount estimation section more strictly (accurately), so calibration is not required (drift adjustment is not required) ) And the number of detectors that can be evaluated can be further increased, and whether or not the operation period can be extended (how much can be extended) can be evaluated more appropriately.

  Further, as illustrated in FIG. 14, the reliability of the nuclear power plant 1 can be further improved by using the estimated true value for monitoring or control in the nuclear power plant 1.

<Embodiment 5>
FIG. 15 is a block diagram showing a configuration of a detector calibration support apparatus according to Embodiment 5 of the present invention. As shown in the figure, Embodiment 5 of the present embodiment processes the detection signals of a large number of detectors 55 provided in each of a plurality of nuclear power plants A, B. Therefore, the detector calibration support apparatus 50 according to the fifth embodiment includes the communication means 52 provided on each plant A, B... Side and the communication means 53 for exchanging signals. . Since the other configuration of the calibration support apparatus 50 is the same as that of the calibration support apparatus 40 of the fourth embodiment, detailed description thereof is omitted here. Further, as the calibration support apparatus in this case, the calibration support apparatus 30 of the third embodiment can be applied.

In each plant A, B,..., An input means 51 for inputting a detection signal (actual value) of each detector 55 and a communication means 53 of the present calibration support apparatus 50 installed at a remote place. Communication means 52 and output means 54. The communication means 52 on each plant A, B... Side and the communication means 53 on the calibration support apparatus 50 side are configured to be capable of bidirectional communication. The statistics (standard deviation σ ₂ , variance σ ₂ ² , average μ ₂ ) of the generated drift amount during normal operation and the generated drift amount during start / stop operation obtained by the statistic evaluation means 14 of the calibration support device 50 Statistic (standard deviation σ _e , variance σ _e ² , average μ _e ), estimated true value obtained by the true value estimating means 46 and uncertainty of the estimated true value, and the amount of drift generated during normal operation in the predicting means 45 Is predicted from the communication means 53 on the calibration support device 50 side to the communication means 52 on each plant A, B... Side, and the output means 54 on each plant A, B. For example, the data is output to a display (monitor) or a printer on the side of each plant A, B,.

  As described above, according to the detector calibration support apparatus or the method thereof in the fifth embodiment, the detection signals of the detectors 55 provided in each of the plurality of nuclear power plants A, B,. Since detection signals are input via the communication means 52, 53 from the input means 51 on the input side of the plurality of nuclear power plants A, B,..., The soundness of the detector 55 is increased at a remote monitoring station or the like. Can be evaluated by reliability. In addition, a regular inspection plan can be formulated at a remote location. Further, an effect that it is possible to deal with a plurality of nuclear power plants A, B,... At one place is obtained, and efficient calibration support can be provided in equipment costs and labor costs.

Here, based on FIG. 16, a schematic configuration of a nuclear power plant to which the calibration support apparatuses 10 to 50 of the first to fifth embodiments are applied will be described. [Table 2] shows the measurement point names and points of the detector shown in FIG.

  As shown in FIG. 16, the heat generated in the nuclear reactor 101 is absorbed by the primary coolant (borate water). The primary coolant is forcibly circulated by the primary coolant pump 103 in the reactor cooling system 104 including the reactor 101, the steam generator 102, and the primary coolant pump 103. A pressurizer 105 for adjusting the pressure is connected to the piping of the reactor cooling system 104. On the other hand, the secondary coolant (water) supplied into the steam generator 102 evaporates due to heat input from the primary coolant and is sent to a steam turbine (not shown) as steam. As a result, the steam turbine is rotated by the steam, and the generator is rotated by the steam turbine to generate electric power.

  Here, T601 and T602 are detectors for detecting the high temperature side temperature and the low temperature side temperature of the primary coolant, respectively, F601 and F602 are detectors for detecting the steam generator feed flow rate, and F603 and F604 are all for steam generation. P601 to P603 are detectors that detect the steam generator pressure, U600 is a detector that detects the reactor output, and L601 to L603 are detectors that detect the pressurizer water level. It is. Numerous other detectors are provided in the nuclear power plant. Of these detectors, a plurality of detectors used in a similar format and under similar environmental conditions are grouped, and the detectors of each group are subjected to statistical evaluation. For example, since triple pressure detectors P601 to P603 are used in a similar format and under similar environmental conditions (detect the same process value), statistics as detectors belonging to the same group Subject to evaluation.

  At present, the nuclear power plant is operated as a power generation facility for base load, and the electric load (output) remains almost constant during normal operation, so the present invention is particularly applicable to the nuclear power plant. However, it is not necessarily limited to this, and it consists of regular inspection and operation at the time of stopping (operation that consists of normal operation and starting operation of starting operation and constant output (including almost constant output)) The present invention is useful when applied to various facilities (such as gas turbine power generation facilities) that repeat the pattern.

  The present invention relates to a detector calibration support apparatus and method, and is useful when applied to support calibration of a detector in a facility having a large number of detectors such as a nuclear power plant.

It is a block diagram which shows the structure of the calibration assistance apparatus of the detector which concerns on Example 1 of this invention. It is a flowchart which shows the processing flow of the said calibration assistance apparatus. It is explanatory drawing which shows the operation cycle etc. of a nuclear power plant. It is a graph which shows the example of an actual measurement value of a detector at the time of normal operation of a nuclear power plant. It is a graph which shows the example of an inspection result at the time of periodic inspection in a nuclear power plant. It is a block diagram which shows the structure of the calibration assistance apparatus of the detector which concerns on Example 2 of this invention. It is a flowchart which shows the processing flow of the said calibration assistance apparatus. It is explanatory drawing of the drift amount prediction in the prediction means of the said calibration assistance apparatus. It is a block diagram which shows the structure of the calibration assistance apparatus of the detector which concerns on Example 3 of Embodiment of this invention. It is a flowchart which shows the processing flow of the said calibration assistance apparatus. It is explanatory drawing which shows an example of the true value estimation model used with the true value estimation means of the said calibration assistance apparatus. It is a block diagram which shows the structure of the calibration assistance apparatus of the detector which concerns on Example 4 of this invention. It is a flowchart which shows the processing flow of the said calibration assistance apparatus. It is a block diagram which shows the application example of the estimated true value obtained by the true value estimation means of the said calibration assistance apparatus. It is a block diagram which shows the structure of the calibration assistance apparatus of the detector which concerns on Example 5 of this invention. 1 is a schematic configuration diagram of a nuclear power plant to which a calibration support apparatus according to an embodiment of the present invention is applied. It is a conceptual diagram of the conventional drift amount prediction.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1, A, B Nuclear power plant 10 Detector calibration assistance apparatus 11 Input means 12 1st database 13 2nd database 14 Statistics evaluation means 17 Output means 20 Detector calibration assistance apparatus 25 Predictive means 27 Output means 30 Detector calibration support apparatus 36 True value estimation means 37 Output means 40 Detector calibration support apparatus 45 Prediction means 46 True value estimation means 47 Output means 50 Detector calibration support apparatus 51 Input means 52, 53 Communication means 54 Output means 55 Detector 81 Detector 90 Control device 91 Calibration processing function 92 Various control operations 101 Reactor 102 Steam generator 103 Primary coolant pump 104 Reactor cooling system 105 Pressurizer

Claims (12)

Measured values at the time of normal operation after the first periodic inspection in the equipment of the plurality of detectors to be subjected to statistical evaluation provided in the predetermined equipment, and the first in the equipment of the plurality of detectors Measured values during normal operation before the second periodic inspection, which is a periodic inspection next to the periodic inspection, inspection results after drift adjustment at the first periodic inspection of the plurality of detectors, and the plurality of detections Input means for inputting each of the inspection results before the drift adjustment at the time of the second periodic inspection of the container;
First storage means for storing the actual measurement value during the normal operation after the first periodic inspection and the actual measurement value during the normal operation before the second periodic inspection, which are input by the input means;
Second storage means for storing the inspection result after drift adjustment at the time of the first periodic inspection and the inspection result before drift adjustment at the time of the second periodic inspection, which are input by the input means;
Statistically processing the actual measurement value during the normal operation after the first periodic inspection and the actual measurement value during the normal operation before the second periodic inspection stored in the first storage means A statistical amount of the drift amount of the plurality of detectors generated during the normal operation, and the actual measurement value during the normal operation after the first periodic inspection stored in the first storage means and the The actual measurement value during the normal operation before the second periodic inspection, the inspection result after the drift adjustment in the first periodic inspection stored in the second storage means, and the second periodic inspection Statistically processing the inspection results before drift adjustment, and a statistic evaluation means for obtaining a statistic of the drift amount of the plurality of detectors generated during start / stop operation of the equipment,
A detector comprising: an output means for outputting the statistics of the drift amount during the normal operation and the statistics of the drift amount during the start / stop operation obtained by the statistics evaluation means. Calibration support device. In the detector calibration support apparatus according to claim 1,
Predicting an increase in the drift amount of the plurality of detectors that occurs during the normal operation when the normal operation period is extended based on the statistics of the drift amount during the normal operation obtained by the statistic evaluation unit Add a prediction method to
A detector calibration support apparatus, wherein the output means outputs the prediction result of the prediction means. In the detector calibration support apparatus according to claim 1,
The input means is configured to also input detection signals of the plurality of detectors,
A true value estimation model for estimating a true value is used, a true value is estimated based on an actual measurement value of the detection signal input by the input means, and an uncertainty of the estimated true value is estimated by the statistic evaluation means A true value estimating means for calculating based on the statistic of the drift amount during the normal operation obtained in
A detector calibration support apparatus, wherein the output means outputs the estimated true value obtained by the true value estimating means and the uncertainty of the estimated true value. In the detector calibration support apparatus according to claim 3,
In the case of extending the period of the normal operation based on the statistical amount of the drift amount during the normal operation obtained by the statistic evaluation unit and the uncertainty of the estimated true value obtained by the true value estimation unit Adding a prediction means for predicting an increase in drift amount of the plurality of detectors that occurs during the normal operation;
A detector calibration support apparatus, wherein the output means outputs the prediction result of the prediction means. A plurality of equipment-side input means that respectively input detection signals of detectors provided in each of the plurality of equipment and an input means of the calibration support apparatus according to claim 3 or 4 are connected via communication means. And
The detection means characterized in that the input means of the calibration support apparatus according to claim 3 or 4 is configured to input detection signals of the detectors from the plurality of equipment side input means via the communication means. A calibration support device. In the detector calibration support apparatus according to claim 3, 4 or 5,
A detector calibration support apparatus, wherein the estimated true value obtained by the true value estimating means is used for monitoring or control in the facility. Measured values at the time of normal operation after the first periodic inspection in the equipment of the plurality of detectors to be subjected to statistical evaluation provided in the predetermined equipment, and the first in the equipment of the plurality of detectors Statistically processing the measured value at the time of normal operation before the second periodic inspection, which is the periodic inspection next to the periodic inspection, to determine the statistic of the drift amount of the plurality of detectors that occurs during the normal operation,
And the actual measurement value during the normal operation after the first periodic inspection, the actual measurement value during the normal operation before the second periodic inspection, and the drift during the first periodic inspection of the plurality of detectors. The inspection results after adjustment and the inspection results before drift adjustment at the time of the second periodic inspection of the plurality of detectors are statistically processed, and the plurality of detectors generated during the start / stop operation of the equipment A detector calibration support method characterized by obtaining a drift statistic. The detector calibration support method according to claim 7,
A detector that predicts an increase in drift amount of the plurality of detectors that occurs during the normal operation when the period of the normal operation is extended based on a statistic of the drift amount during the normal operation. Calibration support method. The detector calibration support method according to claim 7,
A true value estimation model for estimating a true value is used, a true value is estimated based on measured values of detection signals of the plurality of detectors, and an uncertainty of the estimated true value is determined in the normal operation. A detector calibration support method, which is calculated based on a drift statistic. The detector calibration support method according to claim 9,
Based on the statistics of the drift amount during the normal operation and the uncertainty of the estimated true value, an increase in the drift amount of the plurality of detectors that occurs during the normal operation when the period of the normal operation is extended. A detector calibration support method characterized by predicting.   11. The detection device according to claim 9 or 10, wherein the detection signals are input via communication means from the input means on the plurality of equipment sides that respectively input detection signals of detectors provided in each of the plurality of facilities. A calibration support method for a detector, characterized in that a calibration support method for the detector is implemented. The detector calibration support method according to claim 9, 10 or 11,
A detector calibration support method, wherein the estimated true value is used for monitoring or control in the facility.

Images