KR102580902B1 - System for modeling failure prediction of vanadium detector in CANDU and method therefor - Google Patents

Abstract

The present invention relates to a heavy water reactor vanadium detector failure prediction modeling system that accurately evaluates failure predictions and establishes a replacement plan accordingly. According to the present invention, it is possible to accurately calculate the lifespan of the vanadium detector and determine the replacement time to prevent unnecessary investment and to maintain the FLX program stably so that there are no problems in checking operation restriction conditions such as reactor control, channels, and bundle output. , it is effective in accurately evaluating failure predictions and establishing replacement plans accordingly.

Description

Heavy water vanadium detector failure prediction modeling system and method {System for modeling failure prediction of vanadium detector in CANDU and method therefor}

The present invention relates to a heavy water reactor vanadium detector failure prediction modeling system, and more specifically, to a heavy water reactor vanadium detector failure prediction modeling system that accurately evaluates failure predictions and establishes a replacement plan accordingly.

In heavy water reactor nuclear power plants, vanadium detectors are used in flux mapping (FLX program). The FLX program calculates the output of 14 areas and corrects the output of the platinum detector measurement area (Pic: Calibrated Flux Power). In addition, the FLX program uses 15 mode shapes and 102 detector outputs to produce channel, bundle output, and channel output peak coefficients on-line (DCC) and off-line (core code) and checks whether operation limit conditions are exceeded. do.

The lifespan of the vanadium detector was predicted to be about 10 years after use, but the detector failure occurred too quickly (occurring 5 to 7 years after operation), causing internal confusion, such as questioning the core operation regarding the failure. If the number of vanadium detector failures exceeds 20, the function of the FLX program is frozen in the control computer, making area output, channel, bundle output, and channel output peak coefficient calculation impossible. Therefore, there is a need to accurately evaluate failure predictions and establish replacement plans accordingly.

Public Patent No. 10-2018-0079649 (2018.07.11)

The present invention was designed to solve the above-mentioned problems. It accurately calculates the lifespan of the vanadium detector and determines the replacement time to prevent unnecessary investment and maintain the FLX program stably to check operation limit conditions such as reactor control, channels, and bundle output. The purpose is to provide a heavy water reactor vanadium detector failure prediction modeling system that can take action to prevent problems.

The heavy water reactor vanadium detector failure prediction modeling system according to one aspect of the present invention includes an operating data input unit that inputs operating data and obtains post-simulation results; A mapping unit that receives the output of a plurality of vanadium detectors and maps the flux; a selection input unit that selects a fuel pipe to be replaced and inputs the characteristics of the selected fuel pipe into the core code diffusion equation provided by the designer to obtain pre-simulation results; and a failure prediction unit that calculates the performance results of the operation data input unit, mapping unit, and selection input unit, counts the number of vanadium detector failures, and calculates the channel prediction output, bundle prediction output, and prediction peak coefficient within the set value, wherein the failure The prediction unit predicts and evaluates vanadium detector failure using the standard normal “Z distribution” of probability from data collected based on the performance results of the operating data input unit, mapping unit, and selection input unit.

Preferably, the failure prediction unit predicts failure of the vanadium detector from the manufacturing date, uses an average of 13.2 years for failure of the vanadium detector, and evaluates the standard deviation as 2 years.

Preferably, the failure prediction unit performs failure prediction by determining the manufacturing date of the vanadium detector and includes a failure prediction condition module that predicts the first year of failure.

Preferably, the failure prediction unit includes a cumulative failure count module that evaluates the cumulative number of detector failures over time.

Preferably, the failure prediction unit evaluates the adequacy of the failure prediction modeling by comparing the result of predicting and evaluating the actual failure of the vanadium detector stored in the vanadium detector failure life evaluation table data DB.

Meanwhile, the heavy water reactor vanadium detector failure prediction modeling method includes the steps of (a) inputting operating data to obtain post-simulation results; (b) receiving the output of a plurality of vanadium detectors and mapping the flux; (c) selecting a fuel pipe to be replaced and obtaining pre-simulation results by inputting the characteristics of the selected fuel pipe into the core code diffusion equation provided by the designer; (d) calculating the results of steps (a) to (c), counting the number of vanadium detector failures, and calculating the channel prediction output, bundle prediction output, and prediction peak coefficient within a set value, Step (d) predicts and evaluates vanadium detector failure using the standard normal "Z distribution" of probability from the data collected based on the performance results of steps (a) to (c).

According to the heavy water reactor vanadium detector failure prediction modeling system according to an embodiment of the present invention, the lifespan of the vanadium detector is accurately calculated to determine replacement time to prevent unnecessary investment and to maintain the FLX program stably to control reactor control, channels, bundle output, etc. Measures can be taken to ensure that there are no problems in checking operation limit conditions, and there is an effect of accurately evaluating failure predictions and establishing a replacement plan accordingly.

Figure 1 is a diagram showing a method for predicting core output of a heavy water reactor.
Figure 2 is a diagram showing a heavy water reactor vanadium detector failure prediction modeling system according to an embodiment of the present invention.
Figure 3 is a diagram showing the prediction results of the heavy water reactor vanadium detector failure prediction modeling system according to an embodiment of the present invention.
Figure 4 is a diagram showing the installation location of the vanadium detector in the heavy water reactor vanadium detector failure prediction modeling system according to an embodiment of the present invention, and Figure 5 is a diagram showing the reactor area control of the heavy water reactor vanadium detector failure prediction modeling system according to an embodiment of the present invention. This is a drawing showing (Calibrated Flux Power).
Figure 6 is a flowchart showing a heavy water reactor vanadium detector failure prediction modeling method according to an embodiment of the present invention.

The specific structural or functional descriptions presented in the embodiments of the present invention are merely illustrative for the purpose of explaining the embodiments according to the concept of the present invention, and the embodiments according to the concept of the present invention may be implemented in various forms. In addition, it should not be construed as being limited to the embodiments described in this specification, and should be understood to include all changes, equivalents, and substitutes included in the spirit and technical scope of the present invention.

Meanwhile, in the present invention, terms such as first and/or second may be used to describe various components, but the components are not limited to the above terms. The above terms are used only for the purpose of distinguishing one component from other components, for example, without departing from the scope of rights according to the concept of the present invention, a first component may be named a second component, Similarly, the second component may also be referred to as the first component.

Hereinafter, embodiments of the present invention will be described with reference to the attached drawings. In describing embodiments of the present invention, if a description of a related known function or configuration is judged to unnecessarily obscure the gist of the present invention, the description is omitted.

Figure 1 is a diagram showing a method for predicting core output of a heavy water reactor.

First, the method for predicting core output of the heavy water reactor shown in FIG. 1 will be described. A step of obtaining simulation results by inputting operating data (S11), a step of selecting a fuel pipe to be replaced (S12), and a step of inputting the characteristics of the selected fuel pipe into the core code diffusion equation provided by the designer (S13). Wow, the peak coefficient channel and bundle output are calculated as a result of the analysis of the core code diffusion equation, the step of checking whether the calculation has been completed for all fuel pipes and returning to step S12 if not completed (S14), and the judgment of step S14 As a result, once the peak coefficient channel power and bundle power have been calculated for all fuel pipes, the reactor power is determined based on this (S15), the power plant power is adjusted according to the predicted reactor power (S16), and the fuel is replaced. It is configured to include a step (S17).

To explain the core output prediction method of this heavy water reactor in more detail, first, in step S11, operating data such as fuel replacement history and water level in the liquid area control system are input into the core code (RFSP 2GROUP) diffusion equation to perform post-simulation. Perform.

Next, in step S12, the fuel pipe to be replaced is selected, and the results of the post simulation and the characteristics of the selected fuel pipe are input into the core code diffusion equation provided by the designer in step S13 to predict the output. At this time, the core code diffusion equation provided by the designer is an equation that meets the conditions when designing each heavy water reactor, and is an equation that can calculate the output of the heavy water reactor by inputting information about a specific fuel pipe.

Next, in step S14, the output values of the peak coefficient channel output and bundle output for the specific fuel pipe are obtained, and it is confirmed whether the output values of the peak coefficient channel output and bundle output are obtained for all fuel pipes to be replaced. As a result of this confirmation, if there are more fuel pipes to be analyzed, the process returns to step S12 and inputs the characteristics of other fuel pipes to be replaced.

Next, in step S15, the output of the heavy water reactor is predicted and determined based on the peak coefficient channel output and bundle output for each fuel pipe obtained through the above process. After adjusting the power plant's output as in step S16, Fuel replacement is performed as in step S17.

As explained earlier, after replacing the fuel, the core output of the heavy water reactor increases by the input of new fuel. If the output of the power plant is in a normal state, the rapid increase in output due to the input of the new fuel is added to the output of the power plant, and the reactor Since the output exceeding the trip set value related to the operation stop may occur, the heavy water reactor core output is predicted, and then the fuel is replaced while the output of the power plant is lowered according to the predicted value.

Based on this prediction method, the heavy water reactor vanadium detector failure prediction modeling system according to an embodiment of the present invention reads the output of the vanadium detector, maps the flux, counts the number of vanadium detector failures, and predicts the heavy water reactor core output more accurately within the set value. By doing so, the amount of reduction that lowers the power plant output when replacing the fuel pipe can be appropriately set, thereby preventing operation interruption when replacing the fuel pipe and minimizing the amount of reduction in power generation. In other words, by counting the number of vanadium detector failures and predicting failures within the set value, when the number of vanadium detector failures exceeds the set value of 20, the function of the FLX program is frozen in the control computer to calculate area output, channel, bundle output, and channel output peak coefficient. This has the effect of solving problems that have become impossible, thereby accurately evaluating failure predictions and establishing replacement plans accordingly.

Figure 2 is a diagram showing a heavy water reactor vanadium detector failure prediction modeling system 10 according to an embodiment of the present invention. As shown in FIG. 2, it includes a driving data input unit 110, a mapping unit 120, a selection input unit 130, and a failure prediction unit 140.

The operating data input unit 110 inputs operating data and obtains post-simulation results. This operating data input unit inputs operating data into the core code diffusion equation to obtain post simulation. Operating data include fuel replacement history and water level in the liquid area control system. At this time, post simulation can check the current channel output, bundle output, and peak coefficient.

The mapping unit 120 receives the outputs of multiple vanadium detectors and maps the flux. This mapping unit, separate from the operating data input unit, receives the output of 102 vanadium detectors and maps the flux. Flux mapping is also called POWERMAP, and can obtain accurate output values that reflect the aging deterioration of the current heavy water reactor. The vanadium detector is located within the nuclear reactor assembly and detects neutron flux, and maps the detected neutron flux using a defined flux shape. At this time, channel power, bunch power, and peak coefficient can be obtained through neutron flux mapping.

The selection input unit 130 selects a fuel pipe to be replaced, inputs the characteristics of the selected fuel pipe into the core code diffusion equation provided by the designer, and obtains a pre-simulation result. The selection input unit selects the fuel pipe to be replaced and inputs it into the core code diffusion equation provided by the designer to obtain a free simulation. At this time, the pre-simulation predicts the output value after a certain period of time after replacing the fuel pipe. When the new fuel pipe is used for the first time, the core output increases rapidly and stabilizes at the original output value after 3 days. This pre-simulation includes channel output, bundle output, and peak coefficient information after a certain period of time.

The failure prediction unit 140 calculates the results of the operation data input unit, mapping unit, and selection input unit, counts the number of vanadium detector failures, and calculates the channel prediction output, bundle prediction output, and prediction peak coefficient within the set value. The failure prediction unit predicts channel output, bundle output, and peak coefficient using pre-simulation, post-simulation, and flux mapping results.

The calculation of the predicted channel output is shown in Equation 1 below.

[Formula 1]

Predicted channel output = channel output of flux mapping + channel output of pre-simulation - channel output of post-simulation

As shown above, the output predicted value is obtained using the diffusion equation presented by the designer, the difference from the current output value obtained using the diffusion equation is obtained, and then this is added to the currently detected neutron flux mapping result to obtain the predicted value, which is The same can be applied to the prediction of not only channel output but also bundle output and peak coefficient.

Equation 2 below is the predicted bundle output, and Equation 3 is the equation that calculates the predicted peak coefficient.

[Formula 2]

Prediction bundle output = bundle output of flux mapping + bundle output of pre-simulation - bundle output of post-simulation

[Formula 3]

Predicted peak coefficient = Peak coefficient of flux mapping + Peak coefficient of pre-simulation - Peak coefficient of post-simulation

Through this process, it is possible to accurately calculate channel output, bundle output, and peak coefficient considering aging deterioration.

The failure prediction unit according to this embodiment counts the number of vanadium detector failures and calculates the channel prediction output, bundle prediction output, and prediction peak coefficient within the set value. If the number of vanadium detector failures exceeds 20, the control computer freezes the function of the FLX program. It solves the problem of making it impossible to calculate area output, channel, bundle output, and channel output peak coefficients, and enables accurate evaluation of failure prediction and establishment of a replacement plan accordingly.

The failure prediction unit 140 of the heavy water reactor vanadium detector failure prediction modeling system according to an embodiment of the present invention includes a regular analysis module 141, a failure prediction condition module 142, a cumulative failure count module 143, and a comparative evaluation module ( 144), a regression module 145, and a control module 146.

The normal analysis module 141 predicts and evaluates detector failure using the standard normal "Z distribution" of probability from data collected based on the performance results of the operating data input unit, mapping unit, and selection input unit. At this time, an average of 13.2 years for detector failure is used and the standard deviation is evaluated as 2 years.

[Formula 4]

This standard normal distribution is used to approximate the distribution of the collected data, with the two parameters averaging and standard deviation The shape is determined, and the distribution at this time is It is written as The probability density based on vanadium detector failure prediction modeling is as follows:

[Formula 5]

Figure 3 is a diagram showing the prediction results of the heavy water reactor vanadium detector failure prediction modeling system according to an embodiment of the present invention.

As shown in FIG. 3, the failure prediction condition module 142 determines the manufacturing date of the vanadium detector, performs failure prediction, and displays the actual failed vanadium detector in a square shape. By determining the detector failure average (13.2 years), vanadium detector failure is predicted and evaluated from the time of manufacture, not from the period of use, as shown in Figure 3. At this time, predict the first year of detector failure.

The cumulative failure count module 143 evaluates the cumulative number of detector failures over time.

Figure 3 is a diagram showing the results of vanadium detector failure prediction evaluation from the time of production of the heavy water reactor vanadium detector failure prediction modeling system according to an embodiment of the present invention.

The failure prediction unit 140 compares the actual failure of the vanadium detector stored in the past vanadium detector failure life evaluation table data DB shown in Table 1 below with the previous prediction and evaluation results, and reflects this in FIG. 3 to evaluate the appropriateness of failure prediction modeling. It includes a comparative evaluation module 144 that does.

This comparative evaluation module 144 can evaluate the appropriateness of modeling by comparing actual failure and predicted failure.

[Table 1]

Figure 4 is a diagram showing the installation location of the vanadium detector in the heavy water reactor vanadium detector failure prediction modeling system according to an embodiment of the present invention, and Figure 5 is a diagram showing the reactor area control of the heavy water reactor vanadium detector failure prediction modeling system according to an embodiment of the present invention. This is a drawing showing (Calibrated Flux Power).

The regression module 145 of the failure prediction unit according to this embodiment checks whether prediction has been completed for all fuel pipes to be replaced. This regression module 145 checks whether the prediction has been completed and, if not, returns to select a fuel pipe to be replaced, thereby obtaining predicted values of channel output, bundle power, and peak coefficient for all fuel pipes to be replaced.

If the prediction has been completed for all fuel pipes as determined by the regression module 145, the control module 146 determines the reactor power using the predicted channel power, predicted bundle power, and predicted peak coefficient. At this time, the reactor output is the expected output of the heavy water reactor core when replacing the fuel pipe, and can be predicted more accurately than the conventional method of predicting only the diffusion equation provided by the designer. The control module then reduces the power plant output by considering the predicted reactor output value and the currently set trip settings. At this time, the amount of power reduction can be accurately calculated because the predicted reactor power value is more accurately predicted. In other words, by setting the output reduction amount to the minimum value within a range that does not exceed the trip setting value, the reduction in power generation can be minimized. After this process, replace the fuel pipe.

Figure 6 is a flowchart showing a heavy water reactor vanadium detector failure prediction modeling method according to an embodiment of the present invention. As shown in Figure 5, the heavy water reactor vanadium detector failure prediction modeling method according to an embodiment of the present invention includes the steps of (a) inputting operating data to obtain post-simulation results; (b) receiving the output of a plurality of vanadium detectors and mapping the flux; (c) selecting a fuel pipe to be replaced and obtaining pre-simulation results by inputting the characteristics of the selected fuel pipe into the core code diffusion equation provided by the designer; (d) calculating the results of steps (a) to (c), counting the number of vanadium detector failures, and calculating the channel prediction output, bundle prediction output, and prediction peak coefficient within the set value; Step d) predicts and evaluates vanadium detector failure using the standard normal "Z distribution" of probability from the data collected based on the performance results of steps (a) to (c).

Here, in step (d), the failure prediction of the vanadium detector is performed from the date of manufacture, and an average of 13.2 years for failure of the vanadium detector is used and the standard deviation is evaluated as 2 years. In addition, failure prediction is performed by determining the manufacturing date of the vanadium detector, and the first year of failure is predicted. Additionally, the cumulative number of detector failures over time is evaluated. In addition, the adequacy of failure prediction modeling is evaluated by comparing the results of predicting and evaluating the actual failure of the vanadium detector stored in the vanadium detector failure life evaluation table data DB.

After step (d), the regression module 145 of the failure prediction unit checks whether the prediction has been completed for all fuel pipes to be replaced. This regression module 145 checks whether the prediction has been completed and, if not, returns to select a fuel pipe to be replaced, thereby obtaining predicted values of channel output, bundle power, and peak coefficient for all fuel pipes to be replaced.

Next, if the prediction has been completed for all fuel pipes as determined by the regression module 145, the control module 146 determines the reactor power using the predicted channel power, predicted bundle power, and predicted peak coefficient. At this time, the reactor output is the expected output of the heavy water reactor core when replacing the fuel pipe, and can be predicted more accurately than the conventional method of predicting only the diffusion equation provided by the designer. The control module then reduces the power plant output by considering the predicted reactor output value and the currently set trip settings. At this time, the amount of power reduction can be accurately calculated because the predicted reactor power value is more accurately predicted. In other words, by setting the output reduction amount to the minimum value within a range that does not exceed the trip setting value, the reduction in power generation can be minimized. After this process, replace the fuel pipe.

According to the heavy water reactor vanadium detector failure prediction modeling system and method according to an embodiment of the present invention, it is possible to determine that the failure prediction of the vanadium detector is not related to the detector usage period by evaluating the vanadium detector failure prediction from the date of manufacture. In addition, by presenting a clear methodology for evaluating vanadium detector failure, the actual failure of the vanadium detector can be compared with the prediction, and by establishing a detector replacement strategy (no need for early replacement), unnecessary replacement costs of about 1.7 billion won/unit can be saved.

The present invention described above is not limited to the above-described embodiments and the accompanying drawings, and it will be clear to those skilled in the art that various substitutions, modifications, and changes can be made without departing from the technical spirit of the present invention.

110: Driving data input unit
120: mapping unit
130: selection input unit
140: Failure prediction unit
141: Regular analysis module
142: Failure prediction condition module
143: Fault cumulative count module
144: Comparison evaluation module
145: regression module
146: Control module

Claims (7)

A driving data input unit that inputs driving data and obtains post-simulation results;
A mapping unit that receives the output of a plurality of vanadium detectors and maps the flux;
a selection input unit that selects a fuel pipe to be replaced and inputs the characteristics of the selected fuel pipe into the core code diffusion equation provided by the designer to obtain pre-simulation results; and
It includes a failure prediction unit that calculates the performance results of the operating data input unit, mapping unit, and selection input unit, counts the number of vanadium detector failures, and calculates the channel prediction output, bundle prediction output, and prediction peak coefficient within the set value,
The failure prediction unit predicts and evaluates vanadium detector failure using the standard normal “Z distribution” of probability from data collected based on the performance results of the operating data input unit, mapping unit, and selection input unit,
The failure prediction unit includes a comparison evaluation module that compares actual failure data and prediction results to evaluate the appropriateness of failure prediction, and checks whether predictions have been completed for all fuel pipes to be replaced to select fuel pipes to be replaced based on whether the predictions have been completed. A heavy water reactor vanadium detector failure prediction modeling system that includes a regression module that allows regression to obtain predicted values of channel power, bundle power, and peak coefficient for all fuel pipes to be replaced. According to paragraph 1,
The failure prediction unit is a heavy water reactor vanadium detector failure prediction modeling system that predicts failure of the vanadium detector from the date of manufacture. According to paragraph 1,
The failure prediction unit uses an average of 13.2 years for failure of the vanadium detector and evaluates the standard deviation as 2 years. A heavy water reactor vanadium detector failure prediction modeling system. According to paragraph 1,
The failure prediction unit performs failure prediction by determining the manufacturing date of the vanadium detector, and includes a failure prediction condition module that predicts the first year of failure. A heavy water reactor vanadium detector failure prediction modeling system. According to paragraph 1,
The failure prediction unit is a heavy water reactor vanadium detector failure prediction modeling system including a failure cumulative count module that evaluates the cumulative number of detector failures over time. According to paragraph 1,
The failure prediction unit is a heavy water reactor vanadium detector failure prediction modeling system that evaluates the adequacy of failure prediction modeling by comparing the result of predicting and evaluating the actual failure of the vanadium detector stored in the vanadium detector failure life evaluation table data DB. (a) obtaining post-simulation results by inputting operating data;
(b) receiving the output of a plurality of vanadium detectors and mapping the flux;
(c) selecting a fuel pipe to be replaced and obtaining pre-simulation results by inputting the characteristics of the selected fuel pipe into the core code diffusion equation provided by the designer; and
(d) calculating the results of steps (a) to (c), counting the number of vanadium detector failures, and calculating the channel prediction output, bundle prediction output, and prediction peak coefficient within the set value;
In step (d), the failure of the vanadium detector is predicted and evaluated using the standard normal "Z distribution" of probability from the data collected based on the performance results of steps (a) to (c), and the failure prediction unit is used with the actual failure data and Comparative evaluation step to evaluate the appropriateness of failure prediction by comparing the prediction results, checking whether the prediction has been completed for all fuel pipes to be replaced, and regressing to select the fuel pipe to be replaced according to whether the prediction has been completed, and all fuel pipes to be replaced A heavy water reactor vanadium detector failure prediction modeling method including the step of obtaining predicted values of channel output, bundle output, and peak coefficient.