KR102274139B1 - technical specification monitoring system in nuclear power plant - Google Patents

Abstract

The present invention relates to a real-time monitoring system for operational technical guide compliance at a nuclear power plant. The present invention includes: an operation mode monitoring unit determining a nuclear power plant operation mode with respect to a nuclear power plant operation status from status information on process variables including reactivity conditions, rate heat output, and reactor coolant average temperature; a variable calculation unit calculating and outputting a calculation variable set from an input value with respect to the nuclear power plant process variables; a graph-related variable monitoring unit outputting temperature- and pressure-related graph area values in order to determine whether temperature and pressure values regarding a set process of the nuclear power plant are in a stable area; a measuring instrument monitoring unit monitoring whether the channel status of a measuring instrument measuring a measurement target parameter of the nuclear power plant is normal and outputting measuring instrument channel status result information; an operation limit condition monitoring unit monitoring operation limit conditions based on a set operational technical guide from monitoring target information including the process and calculation variables, operation mode information, graph area value, and measuring instrument channel status result information and outputting violated operation limit condition identification information, alarm, follow-up measure, and time limit in the event of a determination of operation limit condition violation; and a follow-up measure monitoring unit monitoring whether the follow-up measure is completed within the time limit.

Description

Real-time monitoring system for implementation of operational technical guidelines of nuclear power plant {technical specification monitoring system in nuclear power plant}

The present invention relates to a real-time monitoring system for the implementation of operating technical guidelines of a nuclear power plant, and in particular, it is possible to monitor whether a nuclear power plant violates the Limiting Condition for Operation (LCO) and whether follow-up measures are implemented based on the operating technical guidelines. It is related to the real-time monitoring system for the implementation of the operational technical guidelines of nuclear power plants, which was built to

Operational technical guidelines generally used in nuclear power plants are technical documents that prescribe the limiting conditions that operators must observe in order to safely operate a nuclear power plant. The ultimate purpose of this document is to protect workers and the public by requiring operators in charge of power plant operation to comply with minimum equipment and operating parameters to operate nuclear power plants safely.

It is one of the important tasks of the operator to monitor whether the nuclear power plant is being operated while complying with the provisions of the Operational Technical Guidelines. However, it is known that there are many difficulties in applying it as a real-time reference document during the operation of a power plant due to the vast amount of contents of the Operational Technical Guide, monitoring variables, variability of interpretation, time dependence, and workload.

Korean Patent Laid-Open No. 10-2015-0077941 discloses a computerized system of serious accident management guidelines so that these operating technical guidelines can be provided in a computerized form instead of printed materials for easy reference.

However, the system has a drawback in that it cannot provide a function of monitoring whether a follow-up action for a violation of the driving condition is properly made, but only can quickly check the contents of the guideline.

The present invention was devised to improve the above problems, and it is a nuclear power plant operation technology capable of monitoring whether the operation limit condition is violated based on the operation technical guideline, and presenting follow-up measures in case of violation and monitoring the implementation. The purpose is to provide a real-time monitoring system for implementing the guidelines.

In order to achieve the above object, the real-time monitoring system for the implementation of the operating technical guideline for a nuclear power plant according to the present invention is a nuclear power plant from state information of process variables including reactivity conditions, rated heat output, and average reactor coolant temperature for the operating state of a nuclear power plant. an operation mode monitoring unit for determining the operation mode of A variable calculation unit that calculates and outputs the calculated variables set from the input values for the process variables of the nuclear power plant: a graph of temperature and pressure to determine whether the temperature and pressure values for the set process of the nuclear power plant are located in the stable region a graph-related variable monitoring unit for outputting a region value; An instrument monitoring unit that monitors whether the channel state of the instrument measuring the measurement target parameter of the nuclear power plant is normal, and outputs the result information of the instrument channel state; and the process variable of the nuclear power plant, the operation mode monitoring unit based on the set operating technology guide Monitoring target information including the operation mode information determined by , the calculated variable calculated by the variable calculation unit, the graph area value output from the graph related variable monitoring unit, and the instrument channel state result information output from the measuring instrument monitoring unit a driving restriction condition monitoring unit that monitors the driving restriction condition from the controller and outputs the violation driving restriction condition identification information, an alarm, follow-up measures, and a time limit when it is determined that the driving restriction condition is violated; and a follow-up monitoring unit for monitoring whether the follow-up measures are completed within the time limit.

According to one aspect of the present invention, the variable calculation unit calculates the stop margin, which is the amount of reactivity that must be added instantaneously to reach the subcritical state of the nuclear reactor, from the total output according to the total output, current output, and initial burnout of the nuclear reactor. Output loss, output loss at all outputs according to the final burnout, current burnout, initial burnout, end burnout, bubble count, maximum controllability of control rods, number of control rods that cannot be operated, control bank controllability, abnormal It is calculated and provided from the values for the number of control banks, the number of abnormal control rods, and the total control rod control ability.

In addition, the graph-related variable monitoring unit may be configured to monitor whether the pressure and temperature of the reactor coolant system are within a normal region on the graph, and to calculate and output graph region values for the temperature and pressure.

Preferably, the graph-related variable monitoring unit is configured to monitor using a support vector machine.

In addition, the instrument monitoring unit is constructed to determine whether an abnormality is determined according to whether a residual between an output value and an input value reconstructed by applying an autoencoder to a value measured from the instrument exceeds a set threshold value.

In addition, the operation mode monitoring unit outputs operation, start, high temperature standby, high temperature stop, low temperature stop, nuclear fuel reload from state information of process variables including reactivity conditions, rated heat output, and average temperature of reactor coolant for the operating state of the nuclear power plant. It is constructed to determine which one of the driving modes is.

According to the real-time monitoring system for the implementation of the operating technical guidelines of a nuclear power plant according to the present invention, it is possible to significantly reduce the number of cases of violation of the operating technical guidelines by guiding the follow-up measures in case of a violation of the operating restriction conditions and monitoring the implementation, It is possible to minimize unnecessary shutdown or damage to the power plant by ensuring safety and the soundness of safety-related equipment.

1 is a block diagram showing a real-time monitoring system for implementing operating technical guidelines of a nuclear power plant according to the present invention;
FIG. 2 is a flowchart illustrating an example of a process of calculating a stop margin in the variable calculation unit of FIG. 1 .
3 is a diagram for explaining a process of calculating a graph area of temperature-pressure by applying a support vector machine in the graph-related variable monitoring unit of FIG. 1;
4 is a flowchart for explaining a process of monitoring a channel of an instrument by applying an autoencoder in the instrument monitoring unit of FIG. 1 .

Hereinafter, a real-time monitoring system for operating technical guidelines for a nuclear power plant according to a preferred embodiment of the present invention will be described in more detail with reference to the accompanying drawings.

1 is a block diagram showing a real-time monitoring system for implementing operating technical guidelines of a nuclear power plant according to the present invention.

Referring to FIG. 1 , the real-time monitoring system 100 for implementing the operating technical guidelines of a nuclear power plant includes an operation mode monitoring unit 110 , a variable calculation unit 120 , a graph related variable monitoring unit 130 , and a measuring instrument monitoring unit 140 . , an operating technical guide database (DB) 150 , a driving limit condition monitoring unit 160 , and a follow-up action monitoring unit 170 .

The operation mode monitoring unit 110 determines the operation mode from operation state information of the nuclear power plant. The operation mode monitoring unit 110 determines the operation mode of the nuclear power plant 10 from the state information of the process variables including the reactivity condition, the rated heat output, and the average temperature of the reactor coolant with respect to the operation state of the nuclear power plant 10, The driving mode information is provided to the driving limit condition monitoring unit 160 .

The operation mode monitoring unit 110 determines the operation mode by combining the state information of the process variables including the reactivity condition, the rated heat output, and the average temperature of the reactor coolant with respect to the operation state of the nuclear power plant 10 . In general, operation mode is divided into reactivity condition and rated heat output, and there are output operation (operation mode 1), start (operation mode 2), and high temperature standby (operation mode 3). It consists of stop (operation mode 4), low temperature shutdown (operation mode 5), and nuclear fuel reload (operation mode 6), and is constructed to determine which one of these operation modes is.

The variable calculation unit 120 calculates and outputs a calculated variable set from an input value for the process variable of the nuclear power plant 10 .

The variable calculator 120 calculates and provides a shutdown margin (SDM), which is an amount of reactivity that must be instantaneously added to reach the subcritical state of the nuclear reactor. As shown in FIG. 2 , the variable calculation unit 120 calculates the total output of the nuclear reactor, the current output, the amount of output loss in the previous output according to the initial burnout, the output loss in the total output according to the end burnout, and the current combustion. Degree, initial burnout, end burnout, bubble coefficient meaning change in reactivity according to change in unit amount of bubbles, maximum controllability of control rods, number of control rods that cannot be operated, control bank control ability, number of abnormal control banks, number of abnormal control rods It can be constructed to calculate and provide a stopping margin from values for the number and total control rod controllability. Here, the control ability means the neutron control ability of the control rod.

When calculating the stopping margin, first, the amount of output loss is calculated from the current output of the initial burnout. For this, a value corresponding to the amount of power loss in the total power according to the total power of the reactor, the current output, and the initial burnout is used. The following calculates the amount of power loss from the current output of the end burner. For this, a value corresponding to the amount of power loss in the total power according to the total power of the reactor, the current power, and the burn-up at the end of the reactor is used. Thereafter, the output loss amount is calculated from the current output of the current burnup. For this, the output loss amount at the current output of the initial burnout calculated above, the output loss amount at the current output of the end burnup, and values corresponding to the current burnup, the initial burnout, and the end burnout are used. Subsequently, the total reactivity deficit is calculated. For this, the value corresponding to the output loss amount and the bubble coefficient in the current output of the current combustion calculated in the previous process is used. Then, the controllability of the unoperable or abnormal control rod is calculated. For this, the values of the control rod control ability that cannot be operated and the control rod control ability that are abnormal are considered. Through this process, the stopping margin is finally calculated. At this time, the total control rod control ability, the previously calculated total reactivity deficit, and the control ability of the inoperable or abnormal control rod are considered.

The graph related variable monitoring unit 130 receives the temperature and pressure for the set process of the nuclear power plant 10, and the graph for temperature and pressure to determine whether the received temperature and pressure values are located in a stable region on the graph Calculate and output the area value.

As an example, the graph-related variable monitoring unit 130 receives the pressure and temperature of the reactor coolant system, monitors whether the received temperature and pressure are within a normal region on the graph, and calculates the graph region values for the temperature and pressure. can be built to output.

Here, the graph-related variable monitoring unit 130 is preferably constructed to monitor using a support vector machine (SVM).

Here, the support vector machine is an engine that performs a methodology based on the idea of finding a hyperplane that can classify data into two different data as shown in FIG. 3 . Therefore, if two different data can be easily distinguished by a line, a support vector machine can be used as an optimal classification method. In order to apply the support vector machine to the graph-related variable monitoring unit 130, data preparation is first required. Data representing the pressure and temperature of the reactor coolant system are extracted from the pressure-temperature curve (PT curve), and the desired region (Z) is determined according to which region on the pressure-temperature curve the extracted pressure and temperature values are located on the PT curve. (1)) or undesirable regions (Z(0)) to construct the data set. The data set constructed in this way is divided into training data and test data. The second stage is the training stage of the support vector machine model. Train a support vector machine using the training data set. At this time, the performance of the support vector machine depends on the parameter C and the gamma value, and the support vector machine is trained with the goal of finding the hyperplane with the maximum margin. The third stage is the testing stage. The trained support vector machine model is tested on the test data set. When the pressure and temperature of the reactor coolant system of the test data set are input to the support vector machine, the region on the pressure-temperature curve (P-T curve) is output. At this time, as the output value matches the output value predefined in the test data set, it is considered as a classification model with higher accuracy.

The instrument monitoring unit 140 monitors whether the instrument channel state for measuring the measurement target parameter of the nuclear power plant 10 is normal, and outputs instrument channel state result information.

The instrument monitoring unit 140 applies an auto-encoder, which is one of the artificial intelligence engines, to the value measured from the instrument, and is constructed to determine whether the residual between the reconstructed output value and the input value exceeds a set threshold value to determine whether it is abnormal. It is preferable to be

Autoencoder is one of the artificial neural network models and is an engine that reconstructs the input value as it is composed of an encoder and a decoder. Since the values measured from the instruments of the nuclear power plant 10 are varied and have characteristics of time series data, a well-known autoencoder is applied to deal with them and will be described with reference to FIG. 4 together.

In order to apply the autoencoder to the instrument monitoring function, data preparation is required first. Data are constructed by extracting sensor values from the instrument channel being monitored and divided into a training data set and a test data set. At this time, the training data set consists of normal instrument sensor data to train the autoencoder model. The test data set consists of normal and abnormal sensor data to evaluate the anomaly detection capability of the autoencoder model. Using the training data set, the autoencoder model is trained. At this time, the goal of learning is to update the weights in the direction of reducing the difference between the value output by the autoencoder and the input value. The error value of the residual is the squared value of the difference between the output value reconstructed by the autoencoder model and the value input to the autoencoder. Changes in residual values are important because they imply the occurrence of an anomalous sensor. For a normal instrument sensor, the autoencoder model will show a low residual value, whereas for an anomalous instrument sensor, the autoencoder model will show a high residual value. The distinction between normal and abnormal sensors is determined by determining a specific threshold. This threshold is determined using the residual value of the autoencoder model trained with normal sensor data.

Therefore, when the autoencoder model is trained and the threshold is determined, the autoencoder model is tested using the test data set. When the sensor value of the instrument channel of the test data set is input to the learned autoencoder model, this value is reconstructed and output as it is. The residual between the reconstructed output value and the input value is obtained. If the threshold value is not exceeded, it is considered a normal sensor, and if the threshold value is exceeded, it is considered an abnormal sensor.

The operation limit condition monitoring unit 160 determines the process variables of the nuclear power plant 10 and the operation mode monitoring unit 110 based on the operation technology guide database (DB) 150 in which the contents of the set operation technology guidelines are recorded. The operation mode information, the stop margin as a calculated variable calculated by the variable calculation unit 120 , the graph area value in which the temperature-pressure is located in the graph related variable monitoring unit 130 , and the measuring instrument output from the measuring instrument monitoring unit 140 . The driving restriction condition is monitored from the monitoring target information including the channel state result information and the information input from the driver input unit 20 and the current time information, and when it is determined that the driving restriction condition is violated, the violation driving restriction condition identification information and an alarm , follow-up measures, and the time limit are output to the main output unit 180 and the follow-up monitoring unit 170 .

Here, the main output unit 180 may be a driver terminal that displays and provides driving state information to the driver. Here, the operator terminal may be a display device of a main control room of a nuclear power plant, a display device of an operation technology room, a display device of a general situation room, and the like.

The follow-up monitoring unit 170 monitors whether the follow-up measures are completed within the time limit, and provides the monitoring result to the main output unit 180 .

The follow-up monitoring unit 170 includes the process variables of the nuclear power plant 10 , the operation mode information determined by the operation mode monitoring unit 110 , the stop margin that is the calculated variable calculated by the variable calculation unit 120 , and graph related variables. Monitoring target information including the graph area value in which the temperature-pressure is located in the monitoring unit 130 , the instrument channel state result information output from the measuring instrument monitoring unit 140 , information input from the driver input unit 20 , and current time information is received, it is determined whether the follow-up measures have been implemented, and the result information on whether the implementation has been completed is provided to the main output unit 180 .

On the other hand, the driving restriction condition monitoring unit 160 is constructed to record the driving restriction condition violation history information in a storage device (not shown) and to support reading.

According to the real-time monitoring system for the implementation of the operating technical guidelines of nuclear power plants described above, it is possible to significantly reduce the number of cases of violation of the operating technical guidelines by guiding follow-up measures in case of violation of the operating restriction conditions and monitoring the implementation. It is possible to minimize unnecessary shutdown or damage to the power plant by securing safety and ensuring the soundness of safety-related equipment.

110: operation mode monitoring unit 120: variable calculation unit
130: graph related variable monitoring unit 140: measuring instrument monitoring unit
150: Operational technical guide database (DB) 160: Operation limit condition monitoring unit
170: Follow-up monitoring department

Claims (6)

an operation mode monitoring unit for determining an operation mode of the nuclear power plant from state information of process variables including reactivity conditions, rated heat output, and average temperature of a reactor coolant with respect to the operation state of the nuclear power plant;
A variable calculation unit that calculates and outputs the calculated variable set from the input value of the process variable of the nuclear power plant;
a graph-related variable monitoring unit for outputting graph region values for temperature and pressure to determine whether temperature and pressure values for a set process of a nuclear power plant are located in a stable region;
A measuring instrument monitoring unit that monitors whether a channel state of a measuring instrument measuring a parameter to be measured in a nuclear power plant is normal, and outputs instrument channel state result information;
Based on the set operating technology guideline, the process variable of the nuclear power plant, the operation mode information determined by the operation mode monitoring unit, the calculated variable calculated by the variable calculation unit, the graph area value output from the graph related variable monitoring unit, and , the operation restriction condition is monitored from the monitoring target information including the instrument channel status result information output from the instrument monitoring unit, and when it is determined that the operation restriction condition is violated, the violation operation restriction condition identification information, alarm, follow-up measures, and restrictions a driving restriction condition monitoring unit for outputting time;
and a follow-up monitoring unit for monitoring whether the follow-up measures are completed within the time limit. The method of claim 1, wherein the variable calculation unit
The stop margin, which is the amount of reactivity that must be added instantaneously to reach the subcritical state of the nuclear reactor, is calculated from the total output of the reactor, the current output, the amount of power loss at the full output according to the initial burnout, and the total output according to the end burnout. output loss, current burnout, initial burnout, end burnout, bubble count, maximum controllability of control rods, number of control rods that cannot be operated, control bank control ability, number of abnormal control banks, number of abnormal control rods, total control rod control A real-time monitoring system for operating technical guidelines for nuclear power plants, characterized in that they are calculated and provided from values for performance. The nuclear power plant according to claim 2, wherein the graph-related variable monitoring unit monitors whether the pressure and temperature of the reactor coolant system are within the normal region on the graph, and calculates and outputs the graph region values for the temperature and pressure. A real-time monitoring system for the implementation of operating technical guidelines for power plants. [4] The real-time monitoring system according to claim 3, wherein the graph-related variable monitoring unit is configured to monitor using a support vector machine. 5. The method of claim 4, wherein the instrument monitoring unit
Real-time monitoring of implementation of operating technical guidelines for nuclear power plants, characterized in that by applying an auto-encoder to the values measured from the instrument and determining whether abnormalities are detected according to whether the residual between the reconstructed output value and the input value exceeds a set threshold value system. The method of claim 5, wherein the operation mode monitoring unit
From the state information of process variables including reactivity condition, rated heat output, and average temperature of reactor coolant for the operating state of a nuclear power plant, which one of operation mode is output operation, start, high temperature standby, high temperature stop, low temperature stop, and nuclear fuel reload? A real-time monitoring system for implementing technical guidelines for operation of nuclear power plants, characterized in that they are determined to be determined.

Images