JP7440340B2 - Prediction model building device and prediction device - Google Patents

Description

The present invention relates to a prediction model construction device and a prediction device for reactor water radioactivity concentration and water discharge information of a nuclear power plant.

As nuclear power plants (also simply referred to as plants), for example, boiling water nuclear power plants and pressurized water nuclear power plants are known. In these plants, stainless steel, nickel-based alloys, and the like are used in the water-contact parts of major structural members such as reactor pressure vessels to suppress corrosion. Furthermore, in these plants, a part of the cooling water (hereinafter also referred to as reactor water) existing in the reactor pressure vessel is purified by a reactor water purification system, and the metal impurities present in the reactor water are actively removed. It is being removed.

Even if the above-mentioned corrosion prevention measures are taken, the presence of extremely small amounts of metal impurities remaining in the reactor water cannot be avoided. Adheres to external surfaces. The metal elements contained in the metal impurities attached to the outer surface of the fuel rod undergo a nuclear reaction when irradiated with neutrons emitted from the nuclear fuel material inside the fuel rod, and become radioactive nuclides such as cobalt-60, cobalt-58, chromium-51, and manganese-54. Become. Some radionuclides attached to the outer surface of the fuel rods in the form of oxides are eluted into the reactor water as ions, depending on the solubility of the incorporated oxides. Additionally, radionuclides are released into reactor water as an insoluble solid called cladding.

Some of the radionuclides in the reactor water are removed by the reactor water purification system. However, the radionuclides that are not removed accumulate on the surfaces of structural members that come into contact with the reactor water while circulating in the recirculation system and the like along with the reactor water. As a result, radiation is emitted from the surface of the structural member, causing radiation exposure to workers performing periodic inspections. The exposure doses of workers are controlled to ensure that each worker does not exceed the prescribed values. However, in recent years, this standard value has been lowered, and it has become necessary to reduce the exposure dose of each worker to the lowest possible level economically.

Under these circumstances, it is effective to predict the exposure dose at the next periodic inspection, create shielding plans and worker plans, and determine the necessity of decontamination in order to reduce the total exposure dose. This is a countermeasure. Prediction of exposure dose requires prediction of piping dose, and since piping dose strongly depends on reactor water radioactivity concentration during operation, it is difficult to predict changes in reactor water radioactivity concentration during plant operation. becomes important. In addition, since the exposure dose for the next periodic inspection is predicted and used in the periodic inspection plan, the prediction needs to be made as quickly as possible.

As a prediction device for reactor water radioactivity concentration, there is a water quality diagnosis system for reducing the dose rate of the primary reactor system using a physical and chemical simulation model as described in Patent Document 1, for example. This water quality diagnosis system uses a simulation model (mass balance model) that uses current water quality conditions as input to estimate changes in radioactivity in cooling water, predicts future plant dose rates, and uses the results to predict current water quality conditions. Diagnose whether it is good or bad.
The technique disclosed in Patent Document 1 requires that a simulation model and model parameters exist and be optimized. However, in reality, model parameters change over time, equipment and materials are replaced as operation continues, and the difference between model calculation values and actual measurements may increase.

The self-learning diagnosis and prediction device described in Patent Document 2 automatically repairs model parameters in response to temporal changes in plant specifications, characteristics, etc., prevents deterioration of prediction accuracy, and improves the model through self-learning. It has functions.

Japanese Patent Application Publication No. 64-063894 Japanese Patent Application Publication No. 06-289179

The technique described in Patent Document 2 optimizes model parameters of a model set based on a physical model or a chemical model and adjusts their contribution, and the model needs to be prepared in advance. Therefore, the correlation between the state quantity of the prediction target and the state quantity used as an input must be described as a mathematical expression.
The correlation between the state quantities expressed by the mass balance model is optimized by optimizing the model parameters. However, although correlations are possible, if the correlations are complex and difficult to express mathematically, an appropriate model cannot be constructed. For this reason, there is a need to accurately predict plant state quantities even when complex correlations that cannot be expressed mathematically are included.

In addition to predicting the radioactivity concentration in reactor water, plants also need to control the temperature of heated wastewater (water discharge). Specifically, the plant uses seawater to cool the steam generated from the heat generated by nuclear fission that spins the turbines. In plants, it is necessary to reduce the radioactivity concentration in reactor water and comply with regulations regarding the temperature rise range of seawater (heated wastewater and discharged water). Since the temperature depends on the seawater temperature at the time of water intake and the thermal output of the reactor, it is necessary to predict the temperature at the time of water discharge (or the temperature difference from the time of water intake, also referred to as water discharge information) as well as the radioactivity concentration in the reactor water.

The present invention has been made in view of this background, and provides a predictive model construction device and prediction model for constructing a predictive model that enables highly accurate prediction of reactor water radioactivity concentration and water discharge information in a nuclear power plant. The task is to provide equipment.

In order to solve the above-mentioned problems, the predictive model construction device according to the present invention is a predictive model that constructs a predictive model for predicting the reactor water radioactive metal corrosion product concentration and seawater discharge information in a nuclear power plant. The construction device includes a simulation unit that calculates a predicted plant state quantity calculated by a physical model that describes the plant state quantity of the nuclear power plant, and as input data, measurable plant state quantities and seawater information. , the predicted value of the plant state quantity, and includes, as output data, the actual measured value of the radioactive metal corrosion product concentration in the reactor water, and the water discharge information, which is trained by a machine learning model, to generate the prediction model. Equipped with a learning section to build.

Further, in order to solve the above-mentioned problems, a prediction device according to the present invention is a prediction device that predicts the concentration of radioactive metal corrosion products in reactor water and seawater discharge information in a nuclear power plant, and which uses input data as input data. , includes measurable plant state quantities, seawater information, and predicted plant state quantities calculated by a physical model that describes the plant state quantities of the nuclear power plant, and as output data, the reactor water, which is the actual measured value. A prediction model constructed by making a machine learning model learn training data including radioactive metal corrosion product concentration and the water discharge information is stored, and the predicted value of the plant state quantity is calculated from the plant state quantity using the physical model. A simulation unit for calculating, inputting the measurable plant state quantities, the seawater information, and the calculated predicted plant state quantities as input data to the prediction model, and calculating the reactor water radioactive metal corrosion product concentration. , and a prediction unit that calculates the water discharge information as an output.

According to the present invention, it is possible to provide a predictive model construction device and a prediction device that construct a predictive model that enables highly accurate prediction of reactor water radioactivity concentration and water discharge information of a nuclear power plant.

FIG. 1 is an overall system configuration diagram of a nuclear power plant according to the present embodiment. FIG. 3 is a diagram for explaining a mass balance model of transfer behavior of metal corrosion products to reactor water according to the present embodiment. FIG. 2 is a functional configuration diagram of a reactor water radioactivity concentration/discharge water temperature prediction device according to the present embodiment. FIG. 3 is a diagram for explaining the configuration of data that is the source of training data for a predictive model in learning processing according to the present embodiment. FIG. 3 is a diagram for explaining input data and output data of a predictive model, which is teacher data according to the present embodiment. It is a flowchart of the learning process which the learning part concerning this embodiment performs. FIG. 3 is a diagram for explaining the configuration of data that is the source of input data and output data of a prediction model in prediction processing according to the present embodiment. FIG. 3 is a diagram for explaining input data and output data of a prediction model in prediction processing according to the present embodiment. It is a flowchart of the prediction process which the prediction part concerning this embodiment performs. FIG. 7 is a diagram for explaining the configuration of data that is the source of training data for a prediction model in a learning process according to a modification of the present embodiment. FIG. 7 is a diagram for explaining the configuration of data that is the source of input data and output data of a prediction model in prediction processing according to a modification of the present embodiment. FIG. 2 is a diagram for explaining changes in predicted cobalt-60 concentration data and discharge water temperature due to changes in plant state quantities and seawater information according to the present embodiment. FIG. 3 is a functional configuration diagram of a reactor water radioactivity concentration/discharge water temperature prediction device according to a modification of the present embodiment. It is a flowchart of the prediction process using the planning pattern based on the modification of this embodiment.

Before explaining the reactor water radioactivity concentration/discharge temperature prediction device (prediction model construction method and prediction method) in the mode for carrying out the present invention (embodiment), we will explain the nuclear power plant to be predicted and the reactor water temperature prediction device (prediction model construction method and prediction method). A simulation model (mass balance model) for estimating changes in radioactivity will be explained.

≪Overview of nuclear power plant≫
FIG. 1 is an overall system configuration diagram of a nuclear power plant P100 according to this embodiment. A schematic configuration of a nuclear power plant P100 (for example, a boiling water nuclear power plant) to which the reactor water radioactivity concentration/discharge water temperature prediction device in this embodiment is applied will be described with reference to FIG. 1.
The nuclear power plant P100 includes a nuclear reactor P1, a turbine P3, a condenser P4, a reactor purification system, a water supply system, and the like. The nuclear reactor P1 installed in the reactor containment vessel P11 has a reactor pressure vessel P12 containing a reactor core P13. A cylindrical core shroud P15 installed within the reactor pressure vessel P12 surrounds the reactor core P13. A plurality of fuel assemblies (not shown) are loaded in the core P13. Each fuel assembly includes a plurality of fuel rods filled with a plurality of fuel pellets made of nuclear fuel material. An annular downcomer P17 is formed between the inner surface of the reactor pressure vessel P12 and the outer surface of the reactor core shroud P15. A plurality of internal pumps P21 are installed at the bottom of the reactor pressure vessel P12. The impeller of the internal pump P21 is arranged at the lower part of the downcomer P17.

The water supply system includes a water supply pipe P10 connecting the condenser P4 and the reactor pressure vessel P12, a condensate pump P5, a condensate purification device P6, a water supply pump P7, a low pressure feed water heater P8, and a high pressure feed water heater P9. are installed in this order from the condenser P4 toward the reactor pressure vessel P12. A hydrogen injection device P16 is connected to the water supply pipe P10 between the condensate purification device P6 and the water supply pump P7 by a hydrogen injection pipe P18. An on-off valve P19 is provided in the hydrogen injection pipe P18.

The reactor purification system includes a purification system piping (stainless steel member) P20 made of stainless steel that connects the reactor pressure vessel P12 and the water supply piping P10, a purification system isolation valve P23, a purification system pump P24, and a regenerative heat exchanger P25. , non-regenerative heat exchanger P26, and reactor water purification device P27 are installed and configured in this order. The residual heat removal system installed in the nuclear power plant P100 has one end connected to the reactor pressure vessel P12 and connected to the downcomer P17, and the other end connected to the inside of the reactor pressure vessel P12 above the reactor core P13. It has a residual heat removal system piping P28. A residual heat removal system pump P29 and a heat exchanger (cooling device) P30 are installed in this residual heat removal system piping P28. One end of the purification system piping P20 is connected to the residual heat removal system piping P28 upstream of the residual heat removal system pump P29.

Cooling water (reactor water) present in the downcomer P17 in the reactor pressure vessel P12 is pressurized by the internal pump P21 and guided to the lower plenum below the reactor core P13. Reactor water is supplied from the lower plenum to the reactor core P13, and is heated by heat generated by fission of nuclear fuel material contained in the fuel rods of the fuel assembly. Some of the heated reactor water turns into steam. This steam is guided from the reactor pressure vessel P12 through the main steam pipe P2 to the turbine P3, and rotates the turbine P3. A generator (not shown) connected to turbine P3 is rotated to generate electric power. Steam discharged from turbine P3 is condensed into water in condenser P4.

The condenser P4 is a type of heat exchanger, which cools the steam with seawater taken in at the water intake and returns it to water. The heated waste water, whose temperature has increased by cooling the steam, is discharged from the water outlet. The difference between the temperature at the time of water intake (water intake temperature) and the temperature at the time of water discharge (water discharge temperature) is specified to be 7 to 8°C. For this reason, the plant is operated while adjusting the flow rate of seawater and heat output.

The water condensed in the condenser P4 is supplied as water supply into the reactor pressure vessel P12 through the water supply pipe P10. The feed water flowing through the water supply pipe P10 is pressurized by the condensate pump P5, impurities are removed by the condensate purifier P6, further pressurized by the feed water pump P7, and heated by the low pressure feed water heater P8 and the high pressure feed water heater P9. Ru. The extracted steam extracted from the main steam pipe P2 and the turbine P3 through the extraction pipe P14 is supplied to the low pressure feed water heater P8 and the high pressure feed water heater P9, respectively, and serves as a heating source for the feed water.

Since the reactor water in the reactor pressure vessel P12 contains metal corrosion products contained in the feed water and products generated by corrosion of the structural materials in the reactor pressure vessel P12, a certain proportion of the reactor water is It is purified by the purification system. The reactor water in the reactor pressure vessel P12 is supplied to the regenerative heat exchanger P25 and the non-regenerative heat exchanger P26 through the purification system piping P20 branched from the residual heat removal system piping P28 by driving the purification system pump P24. It is cooled down to about 50°C by a heat exchanger. The cooled reactor water passes through the reactor water purification device P27 to remove metal corrosion products contained in the reactor water, is heated in the regenerative heat exchanger P25, and then merges with the feed water flowing in the water supply pipe P10. and is supplied to the reactor pressure vessel P12.

When stopping the operation of the nuclear reactor P1, all control rods (not shown) are inserted into the reactor core. By inserting all the control rods, the fission reaction of the nuclear fuel material is stopped, and the operation of the nuclear reactor P1 is stopped. The heat remaining in the equipment in the reactor core P13 and reactor pressure vessel P12 is removed by evaporation of the reactor water, but if the temperature drops to a certain extent, the heat removal efficiency due to evaporation of the reactor water decreases, so the reactor water temperature is reduced to 150°C. When the residual heat removal system is used to cool down the equipment in the reactor core P13 and reactor pressure vessel P12. That is, by driving the residual heat removal system pump P29, the reactor water in the reactor pressure vessel P12 is supplied to the heat exchanger P30 through the residual heat removal system piping P28, and the reactor water is cooled by the heat exchanger P30. It is returned to the reactor pressure vessel P12.

Reactor water during nuclear reactor operation contains radioactive metal corrosion products such as cobalt-60, which adhere to structural materials depending on their concentration. Radiation exposure occurs to personnel performing routine inspections.
The above is an overview of the nuclear power plants targeted for prediction. Next, a simulation model (mass balance model) used in this embodiment will be explained.

≪Overview of mass balance model≫
FIG. 2 is a diagram for explaining a mass balance model 140 (see FIG. 3 described later) of the transfer behavior of metal corrosion products to reactor water according to the present embodiment. The mass balance model 140 assumes that metal corrosion products contained in the feed water and metal corrosion products generated as a result of corrosion of structural materials inside and outside the reactor that are in contact with the reactor water are interposed in the reactor water. This is a physical model that uses the macroscopic law of conservation of mass to describe the dynamic behavior of fuel rods, redeposition to the surfaces of fuel rods and structural materials inside and outside the reactor, and removal from the system by the reactor water purification system. Note that the solid line and broken line arrows in FIG. 2 indicate the transition between the cladding and the ions.
The mass balance model 140 of metal corrosion products is described by a group of simultaneous differential equations shown below (Equation 1) to (Equation 8).

In the above formula, the meanings of variables and parameters are as follows.
C: Concentration of metal corrosion products in reactor water (concentration of metal corrosion products in reactor water, for example, concentration of iron, nickel, cobalt, etc.)
t: Time V: Amount of reactor water held F _f : Feed water flow rate C _f : Concentration of metal corrosion products in the feed water (concentration of metal corrosion products in the feed water)
X: Generation rate of metal corrosion products caused by corrosion of reactor internal structural materials ζ: Elution or peeling constant of fuel rod deposits ζ _p ¹ : Elution or peeling constant of reactor internal structural materials deposits ζ _p ² : Outside the reactor Elution or peeling constant of structural material deposits

M: Amount of metal corrosion products that accumulate on the fuel rod _m1 : Amount of metal corrosion product that adheres and accumulates on the surface of structural materials inside the reactor _m2 : Amount of metal corrosion product that adheres and accumulates on the surface of structural materials outside the reactor δ: Fuel rod Adhesion constant β to: Removal rate in the reactor purification system δ _p ¹ : Adhesion constant to internal structural materials δ _p ² : Adhesion constant to external structural materials S ₁ : Surface area of structural materials inside the reactor S ₂ : Surface area of structural materials outside the furnace

R: Concentration of radioactive metal corrosion products in reactor water (concentration of radioactive metal corrosion products in reactor water, for example, concentration of cobalt-60, cobalt-58, manganese-54, etc.)
Y: Generation rate of radioactive metal corrosion products caused by corrosion of structural materials inside the reactor A: Amount of radioactive metal corrosion products accumulated on fuel rods Γ ₁ : Amount of radioactive metal corrosion products deposited and accumulated on the surface of structural materials inside the reactor Γ ₂ : Amount of radioactive metal corrosion products deposited and accumulated on the surface of structural materials outside the reactor λ: Decay constant of radioactive metal corrosion products G: Production rate of radionuclides on fuel rods G ₁ : Radioactive nuclides on structural materials inside the reactor generation rate of

Among the above variables, C, C _f , F _f , R, and Γ ₂ are state quantities that can be measured during operation, and M, A, and Γ 1 are state quantities that can be measured during operation, and M, A, and Γ ₁ are state quantities that can be measured when the fuel rods are removed from the reactor during shutdown such as periodic inspections. It is a state quantity that can be measured when Further, V, S ₁ and S ₂ are plant parameters unique to the plant. λ, G, G ₁ are physical constants determined depending on the nuclide of the radioactive metal corrosion product, and X, Y, ζ, ζ _p ¹ , ζ _p ² , δ, δ _p ¹ , δ _p ² , β are Basically, it is a model parameter. Note that m ₁ and m ₂ are state quantities that are usually difficult to measure because it is not possible to distinguish between what has adhered from the water side and what has occurred due to corrosion of the structural material.

In the conventional technology, model parameters are adjusted so that the calculated values of state quantities such as C, M, and R on the left side match the actual values, and the future estimated value of C _f is given as input using the adjusted model parameters. The prediction is made by calculating each state quantity on the left side.

≪Reactor water radioactivity concentration/discharge water temperature prediction device: Overall configuration≫
Below, a description will be given of a reactor water radioactivity concentration/discharge temperature prediction device that predicts the reactor water radioactivity concentration and discharge water temperature in the nuclear power plant P100 according to the present embodiment. The reactor water radioactivity concentration/water discharge temperature prediction device predicts the concentrations of cobalt-60, cobalt-58, chromium-51, manganese-54, etc. in the reactor water, and the discharge water temperature. For prediction, simulation using the mass balance model 140 described by (Formula 1) to (Formula 8) and machine learning technology are used. Specifically, the results of the simulation are used as part of the input to the machine learning model.

According to the reactor water radioactivity concentration/discharge temperature prediction device according to this embodiment, the concentration of radioactive metal corrosion products and the discharge water temperature can be predicted with high accuracy. By using this prediction, plant operators can lower the radioactivity concentration of reactor water during operation, install shielding, and carry out chemical decontamination while adhering to the temperature control regulations for water discharge (heated wastewater). Students will be able to formulate and implement plans to reduce exposure.

FIG. 3 is a functional configuration diagram of the reactor water radioactivity concentration/discharge water temperature prediction device 100 according to this embodiment. The reactor water radioactivity concentration/discharge water temperature prediction device 100 is a computer, and includes a control section 110, a storage section 120, and an input/output section 160. The reactor water radioactivity concentration/discharge water temperature prediction device 100 receives operational data such as heat output from a process computer P110 used in the nuclear power plant P100, and output data from a measuring instrument P120 installed in the nuclear power plant P100. Receive.

The storage unit 120 is composed of a RAM (Random Access Memory), a ROM (Read Only Memory), an SSD (Solid State Drive), etc., and stores a prediction model 130, a mass balance model 140, and a plant state quantity/seawater information database 150. . The prediction model 130 is a learning model of machine learning, and is, for example, a neural network. The mass balance model 140 is a simulation model described by (Formula 1) to (Formula 8), and includes formulas and parameters that describe the model. Alternatively, the mass balance model 140 may be considered to be a program that executes a simulation.

The plant state quantity/seawater information database 150 stores plant state quantities including plant data, water supply data, and reactor water quality data of the nuclear power plant P100, which are the values of variables and parameters included in (Formula 1) to (Formula 8). Remember. The plant state quantity/seawater information database 150 also stores plant state quantities such as the stay period in the reactor of the fuel assembly loaded in the reactor core P13 (see FIG. 1), electrical output, and seawater information. The seawater information includes water intake temperature (seawater temperature at the water intake) and water discharge temperature (temperature of heated waste water at the water discharge). The seawater information may include information such as the temperature and position of ocean currents that affect the water intake temperature.

The input/output unit 160 receives data from the process computer P110 and the measuring instrument P120 and stores it in the plant state quantity/seawater information database 150. The input/output unit 160 also includes a display, keyboard, and mouse (not shown), and receives operations from users of the reactor water radioactivity concentration/discharge water temperature prediction device 100 and displays data such as prediction results. .

The control unit 110 is composed of a CPU (Central Processing Unit), and includes a learning unit 111, a prediction unit 112, and a simulation unit 113. The learning unit 111 performs a learning process (see FIG. 6 described later) using the data stored in the plant state quantity/seawater information database 150 as teacher data (learning data), and predicts the radioactivity concentration and discharge water temperature of the reactor water. A predictive model 130 is generated. The prediction unit 112 inputs data stored in the plant state quantity/seawater information database 150 into the generated prediction model 130, and performs a prediction process (see FIG. 9 described later) to predict the radioactivity concentration and discharge water temperature of the reactor water. (see). The simulation unit 113 executes a simulation using the mass balance model 140 described by (Formula 1) to (Formula 8). The simulation execution result becomes input data for the prediction model 130. Details of the processing by the learning unit 111 and the prediction unit 112 will be explained with reference to FIGS. 4 to 9, which will be described later.

≪Teacher data≫
FIG. 4 is a diagram for explaining the configuration of data that is the source of training data of the prediction model 130 in the learning process according to the present embodiment. FIG. 5 is a diagram for explaining input data and output data of the prediction model 130, which is teacher data according to this embodiment. In this embodiment, an example of predicting the future three years (36 months) from data of the past three years (36 months) will be described.

The data set 210 shown in FIG. 4 is a data set that is the source of one set of teacher data whose starting month is the first month. The data set 210 includes, as input data (original data of input data), plant state quantities and seawater information (water intake temperature and water discharge temperature) for the first to 36th months. Furthermore, the data set 210 includes the cobalt-60 reactor water radioactivity concentration (also referred to as cobalt-60 concentration) and discharge water temperature from the 37th month to the 72nd month as output data (original data of the output data). These data are stored in the plant state quantity/seawater information database 150.

Included in each of the plant state quantities from the 1st month to the 36th month are feed water flow rate, feed water metal corrosion product concentration, reactor water metal corrosion product concentration, reactor water radioactive metal corrosion product concentration, reactor water purification flow rate, and fuel. Measurable plant state quantities (actual measurement data), such as the period of stay of the aggregate in the reactor, electrical output, water intake temperature, and water discharge temperature, are included as input data of the prediction model 130. The teacher data serving as the input data of this prediction model 130 corresponds to the arrow pointing from the plant state quantity/seawater information 151 shown in FIG. 5 to the prediction model 130.

In addition, the mass balance model 140 is calculated from the plant state quantities from the 1st month to the 36th month, such as the feed water flow rate, the concentration of metal corrosion products in the feed water, the reactor water purification flow rate, the stay period of the fuel assembly in the reactor, and the electrical output. The amount of metal corrosion products (M) and the amount of radioactive metal corrosion products (A) accumulated on the fuel rods are included as input data (also referred to as plant state quantity predicted values) of the prediction model 130. The teacher data serving as input data for this prediction model 130 corresponds to the arrow pointing from the mass balance model 140 to the prediction model 130 shown in FIG. Note that although the reactor water purification flow rate, the stay period of the fuel assembly in the reactor, the electrical output, and the seawater information do not appear directly in (Equations 1) to (Equations 8), they influence the variables and parameters. For example, the adhesion constant (δ) to the fuel rods depends on electrical output and seawater information, and the removal rate (β) in the reactor purification system is calculated from the reactor water purification flow rate.

The cobalt-60 concentration and water discharge temperature (actual measured values) from the 37th month to the 72nd month are included as output data (reactor water radioactive metal corrosion product concentration and discharge water temperature) of the prediction model 130. The teacher data serving as the output data of this prediction model 130 corresponds to the arrow pointing from the cobalt-60 concentration/water discharge temperature 152 shown in FIG. 5 to the prediction model 130.

The input data and output data described above become one set of training data for the prediction model 130. Similarly, one set of teacher data is generated from the data set 220 whose start date is the second month. Thereafter, by repeating this process, teacher data (data sets of teacher data) for 30 sets of prediction models 130 are generated from the plant state quantities including cobalt-60 and seawater information for the 1st month to the 101st month. The teacher data that is input to the prediction model 130 includes measurable plant state quantities, seawater information (water intake temperature and water discharge temperature), and predicted plant state quantities. Further, the teacher data that is the output of the prediction model 130 includes the actually measured concentration of radioactive metal corrosion products in the reactor water (cobalt-60 concentration) and water discharge information (water discharge temperature).

≪Reactor water radioactivity concentration/discharge water temperature prediction device: Learning process≫
FIG. 6 is a flowchart of the learning process executed by the learning unit 111 according to the present embodiment. The learning process for constructing the predictive model 130 will be explained with reference to FIG.
In step S11, the learning unit 111 starts a process of repeating steps S12 to S13 every predetermined starting month (first month to 30th month).
In step S12, upon receiving the instruction from the learning unit 111, the simulation unit 113 executes a simulation using the mass balance model 140 using the plant state quantity/seawater information 151 as input data. Specifically, the simulation unit 113 calculates the feed water flow rate, feed water metal corrosion product concentration, reactor water purification flow rate, stay period of the fuel assembly in the reactor, which is included in the plant state quantity/sea water information 151 for 36 months from the start month The amount of metal corrosion products (M) and the amount of radioactive metal corrosion products (A) accumulated on the fuel rods are calculated using the mass balance model 140 ((Formula 1) to (Formula 8)) from the electrical output and the like.

In step S13, the learning unit 111 generates teacher data. In detail, the learning unit 111 calculates the feed water flow rate, feed water metal corrosion product concentration, reactor water metal corrosion product concentration, and reactor water radioactive metal corrosion product concentration, which are included in each of the plant state quantities and seawater information 151 for 36 months. , measurable plant state quantities such as reactor water purification flow rate, stay period of fuel assemblies in the reactor, electrical output, seawater information including water intake temperature and water discharge temperature, and metal corrosion formation accumulated on the fuel rods calculated in step S12. The amount of material (M) and the amount of radioactive metal corrosion products (A) are used as input data, and teacher data is generated that uses the cobalt-60 concentration and water discharge temperature 152 after 36 to 71 months from the start month as output data.

In step S14, if the learning unit 111 has executed steps S12 to S13 for every predetermined start month, the process proceeds to step S15, and if there is an unprocessed start month, the learning unit 111 returns to step S12 to determine the unprocessed start month. Process.
In step S15, the learning unit 111 trains the predictive model 130 using the teacher data generated in step S13 (makes the predictive model 130 learn the teacher data) to construct the predictive model 130.
Through the learning process described above, a prediction model 130 that predicts the cobalt-60 concentration and discharge water temperature for 36 months in the future can be constructed from the plant state quantities and seawater information for the past 36 months. Next, a process for predicting the cobalt-60 concentration and water discharge temperature using the prediction model 130 will be described.

≪Reactor water radioactivity concentration/discharge water temperature prediction device: Prediction processing≫
FIG. 7 is a diagram for explaining the configuration of data that is the source of input data and output data of the prediction model 130 in the prediction process according to the present embodiment. FIG. 8 is a diagram for explaining input data and output data of the prediction model 130 in the prediction process according to this embodiment.

In FIG. 7, the reference month is the month in which the prediction process is executed, and the prediction unit 112 calculates the amount of cobalt from January to 36 months after the reference month based on the plant state quantities and seawater information for the past 36 months including the reference month. 60 concentration and discharge temperature.
In detail, the water supply flow rate, water supply metal corrosion product concentration, reactor water metal corrosion product concentration, reactor water radioactive metal corrosion product concentration, and reactor water contained in the plant state quantities from 35 months before the reference month to the reference month Measurable plant state quantities such as water purification flow rate, period of stay of the fuel assembly in the reactor, and electrical output, and seawater information (actual measurement data) including water intake temperature and water discharge temperature are included as input data of the prediction model 130. . This input data corresponds to the arrow pointing from the plant state quantity/seawater information 153 to the prediction model 130 shown in FIG.

In addition, a mass balance model is created based on the plant state quantities from 35 months before the reference month to the reference month, such as feed water flow rate, feed water metal corrosion product concentration, reactor water purification flow rate, fuel assembly stay period in the reactor, and electrical output. The amount of metal corrosion products (M) and the amount of radioactive metal corrosion products (A) accumulated on the fuel rods calculated using 140 are included as input data (predicted plant state quantities) of the prediction model 130. This input data corresponds to the arrow pointing from the mass balance model 140 to the prediction model 130 shown in FIG.

The cobalt-60 concentration and water discharge temperature from January to 36 months after the reference date are included as output data (prediction results) of the prediction model 130. This output data corresponds to the arrow pointing from the prediction model 130 shown in FIG. 8 to the cobalt-60 concentration/water discharge temperature 154.

FIG. 9 is a flowchart of the prediction process executed by the prediction unit 112 according to this embodiment. A prediction process for predicting the cobalt-60 concentration and water discharge temperature for 36 months in the future (future) from actual measurement data for the past 36 months will be explained with reference to FIG.
In step S21, upon receiving the instruction from the prediction unit 112, the simulation unit 113 uses the plant state quantity/seawater information 153 as input data and executes a simulation using the mass balance model 140. In detail, the simulation unit 113 calculates the feed water flow rate, the concentration of metal corrosion products in the feed water, the reactor water purification flow rate, the in-reactor fuel assembly The amount of metal corrosion products (M) and the amount of radioactive metal corrosion products (A) accumulated on the fuel rods are calculated using the mass balance model 140 (see (Equations 1) to (Equations 8)) based on the stay period, electrical output, etc. calculate.

In step S22, the prediction unit 112 inputs the input data to the prediction model 130. In detail, the prediction unit 112 calculates the feed water flow rate, feed water metal corrosion product concentration, reactor water metal corrosion product concentration, reactor water Measurable plant state quantities such as radioactive metal corrosion product concentration, reactor water purification flow rate, period of stay of fuel assemblies in the reactor, electrical output, seawater information, and the amount of metal corrosion products accumulated on the fuel rods calculated in step S21 ( M) and the amount of radioactive metal corrosion products (A) are input into the prediction model 130.
In step S23, the prediction unit 112 predicts the cobalt-60 concentration and the water discharge temperature by executing the prediction model 130 (processing using the prediction model 130) and obtaining the outputs of the cobalt-60 concentration and the discharge water temperature 154.

When the reference point is the present, the cobalt-60 concentration and water discharge temperature 154 can be predicted from the plant state quantities and seawater information 153, which are past actual values (actual measurements). Also, by setting the reference point in the past, predictions can be verified. In detail, the accuracy of prediction is evaluated by calculating cobalt-60 concentration and discharge water temperature 154 from past actual values of plant state quantities and seawater information 153 and comparing them with the actual values of cobalt-60 concentration and discharge water temperature. be able to. For example, the accuracy of the prediction can be evaluated by comparing the cobalt-60 concentration and water discharge temperature when the reference point was set 36 months ago with this month's cobalt-60 concentration and water discharge temperature.

≪Modification: Period of input/output data to the prediction model≫
In the embodiment described above, the data input to the prediction model 130 (see FIGS. 5 and 8) was for 36 months, but the data is not limited to this, and for example, the plant state quantity for a longer period of time is It may be data based on. In addition, in the embodiment described above, the cobalt-60 concentration and water discharge temperature are predicted from the next month to 36 months after the reference month, but the cobalt-60 concentration and water discharge temperature are predicted in the future (long term), for example, from the next month to 48 months. It may also be predicted. By creating training data and constructing the prediction model 130 according to the desired period of input data and the month in which cobalt-60 concentration and water discharge temperature are predicted (prediction month), the period of input data corresponding to the training data and It becomes possible to predict the cobalt-60 concentration and water discharge temperature for the prediction month.

In the embodiment described above, the input data does not include the plant state quantities for the predicted months after the reference month. Plant state quantities and seawater information that are predetermined or predictable in the operation plan of the nuclear power plant may be added to the input data as planned values or predicted values.
FIG. 10 is a diagram for explaining the configuration of data that is the source of the training data of the prediction model 130 in the learning process according to the modification of the present embodiment. FIG. 11 is a diagram for explaining the configuration of data that is the source of input data and output data of the prediction model 130 in prediction processing according to a modification of the present embodiment. Below is an example of predicting cobalt-60 concentration and water discharge temperature for 36 months from the month following the reference month based on actual data for the past 36 months including the reference month, and planned and predicted values for the 36 months from the month following the reference month. explain.

In the learning process, in addition to the plant state quantities and seawater information for 36 months from the start month (see Figure 4), the plant state quantities and seawater information excluding the cobalt-60 concentration and discharge water temperature for 36 months are used as training data. This is the data that is the basis of the input data. In the prediction process, in addition to the plant state quantities and seawater information for the past 36 months including the base month (see Figure 7), cobalt information for the forecast month after the base month (from the time of input to the forecast time) is used. The plant state quantities and seawater information excluding the 60 concentration and discharge water temperature are used as the basis of input data. This input data includes planned values of plant state quantities and predicted values of seawater information.

For example, the feed water flow rate, the concentration of metal corrosion products in the feed water, the reactor water purification flow rate, the stay period of fuel assemblies in the reactor, and the electrical output are determined (predictable) in the operation plan or the operation of the nuclear power plant. Since it can be adjusted by, it is added to the input to the prediction model 130 as a planned value of the plant state quantity. Furthermore, using the mass balance model 140, the amount of metal corrosion products and the amount of radioactive metal corrosion products accumulated on the fuel rods that can be calculated from these plant state quantities are added to the input to the prediction model 130. Furthermore, regarding seawater information, in addition to the actual values up to the reference month, the expected (predicted) intake water temperature (seawater information excluding water discharge information) in the prediction month is added to the input to the prediction model 130. By adding the data of the predicted month, the reactor water radioactivity concentration/discharge water temperature prediction device 100 can predict the cobalt-60 concentration and discharge water temperature with higher accuracy.

≪Operation planning using reactor water radioactivity concentration and discharge water temperature prediction device≫
In order to formulate an operation plan for suppressing the cobalt-60 concentration to a desired value or less, the prediction process may be repeatedly executed while changing the plant state quantity/seawater information 153 (see FIG. 8), which is the planned value. For example, the plant state quantities/seawater information 153 that affect the cobalt-60 concentration and water discharge temperature include the thermal output of the nuclear reactor and the water intake temperature. The amount of cobalt-60 generated and the water discharge temperature (temperature difference from the water intake temperature) depend on the nuclear reaction of the reactor and are correlated with the thermal output. Moreover, the lower the intake water temperature, the more efficiently the steam can be cooled in the condenser P4 (see FIG. 1), which has an effect on the water discharge temperature.
Heat output can be adjusted in a planned manner. In addition, although the intake water temperature cannot be adjusted, it is possible to predict it. By setting the heat output relative to the predicted water intake temperature, it is possible to effect a change in the cobalt-60 concentration.

FIG. 12 is a diagram for explaining changes in predicted cobalt-60 concentration data and discharge water temperature 154 due to changes in plant state quantities and seawater information 153 (see FIG. 8), according to the present embodiment. A graph 310 in FIG. 12 is a graph showing the thermal output (electrical output) included in the plant state quantity/seawater information 153, and before time t is actual data (actual measurement data), and after time t is the three plans. The heat output input as a value is shown by a dashed line, a dashed-dotted line, and a dashed-double-dotted line. The graph 320 is a graph showing the intake water temperature included in the plant state quantity/seawater information 153, and the data before time t is actual data (actually measured data), and the dotted line after time t shows predicted values.
Graphs 330 and 340 are prediction results of the plant state quantities including the thermal output shown in graph 310, the cobalt-60 concentration corresponding to the intake water temperature shown in graph 320, and the water discharge temperature. The dashed line, one-dot chain line, and two-dot chain line of graphs 330 and 340 are prediction results corresponding to the dashed line, one-dot chain line, and two-dot chain line of graph 310, respectively. It can be seen that as the heat output increases, the cobalt-60 concentration and the water discharge temperature increase.

In this way, by repeatedly executing the prediction process while changing the plant state quantity of the planned value of thermal output, the planned value of the plant state quantity for achieving the desired cobalt-60 concentration can be obtained, and the nuclear power plant P100 It is possible to draw up an operation plan. The prediction may be repeated in the same way using a plant state quantity that affects the cobalt-60 concentration other than the thermal output (for example, the iron concentration in the water supply), or the prediction may be made by combining a plurality of plant state quantities. Alternatively, a plurality of predicted water intake temperatures may be prepared, and the prediction may be made while changing the heat output and iron concentration for each water intake temperature.

≪Characteristics of learning processing and prediction processing≫
The source of cobalt-60 in reactor water is cobalt attached to fuel rods, which becomes cobalt-60 when irradiated with neutrons from the fuel rods. This cobalt-60 is incorporated into metal oxides whose main component is iron oxide on the surface of the fuel rods, and the cobalt-60 in the reactor water is eluted from there. Therefore, changes over time in the cobalt-60 concentration in the reactor water are strongly influenced by the amount of metal oxides formed on the surfaces of the fuel rods. This plant state quantity cannot be measured while the nuclear power plant is in operation, but can only be measured after the fuel rods are removed from the core P13 (see FIG. 1) during the shutdown of the nuclear power plant, such as during periodic inspections.

The input data of the prediction model 130 in the embodiment described above includes the amount of metal corrosion products (M) and the amount of radioactive metal corrosion products (A) accumulated on the fuel rods, which are calculated using a mass balance model. By adding plant state quantities that cannot be measured during operation but have a strong influence on the cobalt-60 concentration to the input data of the prediction model 130, it is possible to compare the plant state quantities with only the actual values (measured values) of the plant state quantities as input. As a result, it becomes possible to improve the prediction accuracy of cobalt-60 concentration and water discharge temperature. Furthermore, compared to predictions using a mass balance model, the predictions take into account the correlation of plant state quantities that are not included in the mass balance model, improving accuracy.

≪Modification: Prediction processing using planning pattern≫
In FIG. 12, the reactor water radioactivity concentration and water discharge temperature prediction device 100 predicts the cobalt-60 concentration and the water discharge temperature by manually inputting the planned value of thermal output and the predicted value of water intake temperature. Planned values for thermal output and predicted values for water intake temperature may be prepared in advance, and the cobalt-60 concentration and water discharge temperature may be predicted for each planned value and predicted value, and the predicted results may be output.

FIG. 13 is a functional configuration diagram of a reactor water radioactivity concentration/discharge water temperature prediction device 100A according to a modification of the present embodiment. Compared to the reactor water radioactivity concentration/discharge water temperature prediction device 100 (see FIG. 3), the reactor water radioactivity concentration/discharge water temperature prediction device 100A further stores a plan pattern 121 in the storage unit 120, and Instead, a prediction unit 112A is provided. The planned pattern 121 is a pattern of the above-described planned value of thermal output and a predicted value of intake water temperature. The prediction unit 112A executes a prediction process using a planning pattern shown in FIG. 14, which will be described later.

FIG. 14 is a flowchart of prediction processing using a planning pattern according to a modification of this embodiment.
In step S31, the prediction unit 112A starts processing to execute step S32 for each planned value of thermal output and predicted value of intake water temperature included in the planned pattern 121.
In step S32, the prediction unit 112A executes a prediction process using the plant state quantity including the thermal output shown in the plan pattern 121 and the predicted value of the intake water temperature included in the plan pattern 121 as source data of input data.
In step S33, the prediction unit 112A executes step S32 for every planned value of thermal output and predicted value of intake water temperature, and then proceeds to step S34, where the predicted value of unprocessed planned value of thermal output and predicted value of water intake temperature is If there is, the process returns to step S32 and the prediction process is executed.
In step S34, the prediction unit 112A outputs the result of the prediction process in step S32 (displays it on the display of the input/output unit 160).

According to the reactor water radioactivity concentration/discharge water temperature prediction device 100A described above, the user (plant operator) of the reactor water radioactivity concentration/discharge water temperature prediction device 100A can calculate It will be possible to understand whether the predicted values of cobalt-60 concentration and water discharge temperature will change. Note that the planned values included in the planned pattern 121 are not limited to the thermal output, but may be other plant state quantities that affect the cobalt-60 concentration (for example, the iron concentration in the water supply).

≪Modified example: Output data of the mass balance model input to the prediction model≫
In the embodiment described above, what is calculated by the mass balance model 140 and becomes input data to the prediction model 130 is the amount of metal corrosion products (M) and the amount of radioactive metal corrosion products (A) that accumulate on the fuel rods. The reactor water radioactivity concentration/discharge water temperature prediction device 100 calculates the metal corrosion product concentration (C) and radioactive metal corrosion product concentration (R) in the reactor water using a mass balance model 140 and uses the calculated metal corrosion product concentration (C) and radioactive metal corrosion product concentration (R) as input data for the prediction model 130. Good too.

Specifically, the metal corrosion product concentration (C) and radioactive metal corrosion product concentration (R) in the reactor water calculated in step S12 of the learning process (see FIG. 6) are used as training data for the prediction model 130 in step S13. Add to the input data. This input data corresponds to the arrow pointing from the mass balance model 140 to the prediction model 130 shown in FIG.
Similarly, in the prediction process (see FIG. 9), the concentration of metal corrosion products (C) and the concentration of radioactive metal corrosion products (R) in the reactor water calculated in step S21 are calculated using the prediction model 130 in step S22. Add to the input data for . This input data corresponds to the arrow pointing from the mass balance model 140 to the prediction model 130 shown in FIG.

By increasing the data predicted by the mass balance model 140 as input data to the prediction model 130, the prediction accuracy of the cobalt-60 concentration by the prediction model 130 can be improved.

≪Modification: Metal whose concentration is to be predicted≫
In the embodiment described above, the reactor water radioactivity concentration/discharge water temperature prediction device 100 learns and predicts the cobalt-60 concentration. In addition to cobalt 60, cobalt 58 and manganese 54 can be similarly learned and predicted.

≪Modification: Separation of devices that execute learning processing and prediction processing≫
Note that the present invention is not limited to the embodiments described above, and can be modified without departing from the spirit thereof. For example, the learning process and the prediction process executed by the reactor water radioactivity concentration/discharge water temperature prediction apparatus 100 may be executed by another apparatus. A predictive model construction device may execute a learning process to construct a predictive model, and the predictive device that has acquired the constructed predictive model may execute the predictive process.

≪Other variations≫
In the embodiment described above, the input data is monthly data, but the input data is not limited to this, and may be data for one day or data every six hours, for example. Furthermore, the interval is not limited to a fixed interval, and for example, the interval for input data far from the predicted date may be set to be longer. By lengthening the interval and reducing the number of input data, the reactor water radioactivity concentration/discharge water temperature prediction device 100 can speed up the learning process and prediction process.
In the embodiments described above, a mass balance model is used as a model for the migration behavior of metal corrosion products. Alternatively, physical or chemical models suitable for pressurized water nuclear power plants or other reactor types (eg, heavy water reactors) may be used.

In the embodiment described above, the input data of the prediction model 130 includes seawater information (water intake temperature and discharge water temperature), and the output data (prediction result) includes discharge water temperature. In the input data, the temperature difference between the intake water temperature and the discharge water temperature may be used instead of the seawater information, and in the output data, the temperature difference between the intake water temperature and the discharge water temperature may be used instead of the discharge water temperature. By using such a prediction model, it becomes possible to construct a prediction model that is independent of the intake water temperature and predict the temperature difference between the intake water temperature and the water discharge temperature.

In the embodiment described above, the condenser P4 cools the steam discharged from the turbine P3 with seawater, but the invention is not limited to this. It may be river water or lake water instead of seawater, and the reactor water radioactivity concentration/discharge water temperature prediction device 100 is a prediction device for predicting the temperature of discharged river water or lake water (or temperature difference with intake water). You can build models and make predictions.

Several embodiments of the present invention (prediction model construction device, prediction device, and method corresponding to these devices) have been described above, but these embodiments are merely illustrative and do not fall within the technical scope of the present invention. It is not limited to. The present invention can take various other embodiments, and various changes such as omissions and substitutions can be made without departing from the gist of the present invention. These embodiments and their modifications are included within the scope and gist of the invention described in this specification and the like, as well as within the scope of the invention described in the claims and its equivalents.

100 Reactor water radioactivity concentration/discharge water temperature prediction device (prediction model construction device, prediction device)
111 Learning unit 112, 112A Prediction unit 113 Simulation unit 121 Planning pattern 130 Prediction model 140 Mass balance model (physical model, physical model that describes plant state quantities)
151,153 Plant state quantities/seawater information 152,154 Cobalt-60 concentration/water discharge temperature (reactor water radioactive metal corrosion product concentration/water discharge information)
P1 Reactor P100 Nuclear power plant

Claims (15)

A predictive model construction device for constructing a predictive model for predicting reactor water radioactive metal corrosion product concentration and seawater discharge information in a nuclear power plant, comprising:
a simulation unit that calculates a predicted value of a plant state quantity calculated by a physical model that describes a plant state quantity of the nuclear power plant;
Input data includes measurable plant state quantities, seawater information, and the predicted plant state quantities, and output data includes the reactor water radioactive metal corrosion product concentration, which is an actual measurement value, and the water discharge information. A predictive model construction device comprising: a learning unit that constructs the predictive model by causing a machine learning model to learn training data included therein. The measurable plant state quantities include the reactor water supply flow rate, feedwater metal corrosion product concentration, reactor water metal corrosion product concentration, reactor water radioactive metal corrosion product concentration, reactor water purification flow rate, and fuel assembly. The predictive model building device according to claim 1, wherein the predictive model building device includes at least one of a body's stay period in the furnace and electrical output. The output data is an actual value measured after the input data is actually measured,
The seawater information included in the input data includes an intake temperature and a discharge temperature of the seawater,
The predictive model construction device according to claim 1, wherein the water discharge information included in the output data includes the discharge temperature of the seawater. The physical model is a mass balance model regarding the transfer behavior of metal corrosion products and radioactive metal corrosion products to the reactor water using the reactor water feed flow rate and feed water metal corrosion product concentration,
The predicted plant state quantity value includes at least one of the amount of metal corrosion products accumulated on the fuel rods of the nuclear reactor and the amount of radioactive metal corrosion products calculated by the mass balance model. The predictive model construction device according to claim 1. The predicted plant state quantity further includes at least one of a metal corrosion product concentration and a radioactive metal corrosion product concentration in the reactor water calculated by the mass balance model. The predictive model construction device according to claim 4. The predictive model construction according to claim 1, wherein the reactor water radioactive metal corrosion product concentration is a reactor water radioactive metal corrosion product concentration of any one of cobalt-60, cobalt-58, and manganese-54. Device. A prediction device for predicting reactor water radioactive metal corrosion product concentration and seawater discharge information in a nuclear power plant, comprising:
The input data includes measurable plant state quantities, seawater information, and predicted plant state quantities calculated by a physical model that describes the plant state quantities of the nuclear power plant, and the output data includes actually measured values. storing a prediction model constructed by causing a machine learning model to learn training data including the concentration of radioactive metal corrosion products in the reactor water and the water discharge information;
a simulation unit that calculates the predicted value of the plant state quantity from the plant state quantity using the physical model;
The measurable plant state quantities, the seawater information, and the calculated plant state quantity predicted values are input into the prediction model as input data, and the reactor water radioactive metal corrosion product concentration and the water discharge information are calculated. A prediction device comprising: a prediction unit that calculates as an output; The measurable plant state quantities include the reactor water supply flow rate, feedwater metal corrosion product concentration, reactor water metal corrosion product concentration, reactor water radioactive metal corrosion product concentration, reactor water purification flow rate, and fuel assembly. The prediction device according to claim 7, wherein the prediction device includes at least one of a period of time the body stays in the furnace and an electric output. The output data is an actual value measured after the input data is actually measured,
The seawater information included in the input data includes an intake temperature and a discharge temperature of the seawater,
The prediction device according to claim 7, wherein the water discharge information included in the output data includes a discharge temperature of the seawater. The prediction device according to claim 7, wherein the plant state quantities and the seawater information input to the prediction model are actually measured data at the time of input. The plant state quantity input to the prediction model is a plant state that is different from the reactor water radioactive metal corrosion product concentration from the time of input to the prediction time of the reactor water radioactive metal corrosion product concentration and the water discharge information. including the planned value of the quantity;
The seawater information input to the prediction model further includes a predicted value of seawater information different from the water discharge information from the input time to the prediction time of the reactor water radioactive metal corrosion product concentration and the water discharge information. The prediction device according to claim 10, characterized in that: The prediction device calculates a planned pattern of planned values of plant state quantities different from the reactor water radioactive metal corrosion product concentration from the time of input to the prediction time of the reactor water radioactive metal corrosion product concentration and the water discharge information. and a predicted value of seawater information different from the water discharge information from the input time to the predicted time of the reactor water radioactive metal corrosion product concentration and the water discharge information,
The prediction according to claim 10, wherein the plant state quantity input to the prediction model includes a planned value of the planning pattern, and the seawater information input to the prediction model further includes the predicted value. Device. The physical model is a mass balance model regarding the transfer behavior of metal corrosion products and radioactive metal corrosion products to the reactor water using the reactor water feed flow rate and feed water metal corrosion product concentration,
The predicted plant state quantity value includes at least one of the amount of metal corrosion products accumulated on the fuel rods of the nuclear reactor and the amount of radioactive metal corrosion products calculated by the mass balance model. The prediction device according to any one of claims 7 to 12. The predicted plant state quantity further includes at least one of a metal corrosion product concentration and a radioactive metal corrosion product concentration in the reactor water calculated by the mass balance model. The prediction device according to claim 13. The prediction device according to claim 7, wherein the reactor water radioactive metal corrosion product concentration is a reactor water radioactive metal corrosion product concentration of any one of cobalt-60, cobalt-58, and manganese-54.

Images