Information processor, information processing method, and program

Abstract

To predict a radiation dosage from used nuclear fuel stored in a fuel pool when a water level in the fuel pool is at a predetermined water level.SOLUTION: An information processor having a processor and a memory device stores a learning model for performing learning based on learning data obtained by associating a feature quantity including data pertaining to a first fuel pool in which a first used nuclear fuel is stored in a flooded state in the past and a label including data indicating a radiation dosage due to the first used nuclear fuel at a first predetermined location inside or outside the first fuel pool, and predicts a radiation dosage from a second used nuclear fuel at a second predetermined location inside or outside a second fuel pool by inputting data pertaining to the second fuel pool when the second fuel pool in which the second used nuclear fuel is stored in a flooded state is at a predetermined water level at present into the learning model.SELECTED DRAWING: Figure 3

Description

[Technical field]

[0001] The present invention relates to an information processing device, an information processing method, and a program for predicting the amount of radiation released from spent nuclear fuel in a fuel pool. [Background technology]

[0002] In nuclear power plants, the thermal energy produced when nuclear fuel (such as uranium, plutonium, etc.) undergoes nuclear fission generates steam, which in turn rotates a steam turbine, and the rotating force of the steam turbine rotates a generator at the downstream stage, generating electricity.

[0003] Nuclear fuel that has been used for a certain period of time in a nuclear reactor is treated as spent nuclear fuel, for example, to be reprocessed. When the spent nuclear fuel is removed from the nuclear reactor, it is stored submerged in water stored in a fuel pool until a certain amount of decay heat that is generated when the spent nuclear fuel decays and emits radiation such as alpha rays, beta rays, and gamma rays is removed (see, for example, Patent Document 1). [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent Publication No. 2023-80605 Summary of the Invention [Problem to be solved by the invention]

[0005] The water stored in the fuel pool serves to shield the radiation emitted by the spent nuclear fuel and to cool the spent nuclear fuel. For example, if the water level in the fuel pool drops due to some kind of accident, the water's ability to shield the radiation from the spent nuclear fuel will decrease, and the radiation level inside and around the fuel pool will increase. This will prevent workers from approaching the fuel pool, and there is a risk that they will not be able to carry out the appropriate work of supplying water to the fuel pool.

[0006] In order to deal with such a problem, it is necessary to evaluate the relationship between the drop in the water level in the fuel pool and the rise in the radiation dose inside or around the fuel pool. For example, a method is known in which a numerical analysis is performed using the Monte Carlo method, with the structure (shape, dimensions) of the fuel pool, the radiation source strength of the spent nuclear fuel, the arrangement position of the spent nuclear fuel in the fuel pool, etc. as evaluation conditions. However, this method requires a long time for the numerical analysis, and therefore may not be practical to perform every time an accident occurs.

[0007] The present invention has been made in consideration of the above-mentioned problems, and one object of the present invention is to provide an information processing device, an information processing method, and a program that use a learning model to predict the amount of radiation released by spent nuclear fuel in a fuel pool when the water level in the fuel pool is at a predetermined water level. [Means for solving the problem]

[0008] One of the present inventions for achieving the above-mentioned object is an information processing device having a processor and a storage device, which stores a learning model that performs learning based on learning data that matches features including data relating to a first fuel pool in which a first spent nuclear fuel was stored in a flooded state in the past with labels including data indicating the radiation dose from the first spent nuclear fuel at a first predetermined position inside or outside the first fuel pool, and predicts the radiation dose from the second spent nuclear fuel at a second predetermined position inside or outside the second fuel pool by inputting data relating to the second fuel pool in which a second spent nuclear fuel is currently stored in a flooded state when the second fuel pool is at a predetermined water level into the learning model.

[0009] According to the information processing device of the present invention, when the water level of the second fuel pool drops from the normal water level that should be maintained to a specified water level, by inputting data related to the second fuel pool when the water level of the second fuel pool is at the specified water level into a learning model, it becomes possible to accurately predict the radiation dose from the second spent nuclear fuel at the desired location inside or outside the second fuel pool.

[0010] Another aspect of the present invention for achieving the above-mentioned object is, in an information processing device, the data relating to the first fuel pool includes information on the dimensions within the first fuel pool, the position of the first spent nuclear fuel in the first fuel pool, the water level within the first fuel pool, the source strength of the first spent nuclear fuel, and the first predetermined position, and is a plurality of data prepared for each water level within the first fuel pool, the data indicating the radiation dose from the first spent nuclear fuel at the first predetermined position is a plurality of data prepared for each water level within the first fuel pool, and the data relating to the second fuel pool is data including information on the dimensions within the second fuel pool, the position of the second spent nuclear fuel in the second fuel pool, the predetermined water level within the second fuel pool, the source strength of the second spent nuclear fuel, and the second predetermined position.

[0011] According to the information processing device of the present invention, multiple pieces of feature and label data constituting the learning data are prepared for each water level of the first fuel pool, and the learning model learns from this learning data as supervised data, making it possible to accurately predict the radiation dose from the second spent nuclear fuel.

[0012] In addition, another aspect of the present invention for achieving the above-mentioned object is that, in an information processing device, data indicating the radiation dose from the first spent nuclear fuel at the first predetermined position is obtained for each water level in the first fuel pool by performing a numerical analysis based on the Monte Carlo method on data related to the first fuel pool.

[0013] According to the information processing device of the present invention, multiple pieces of feature and label data constituting the learning data are prepared for each water level of the first fuel pool by numerical analysis using the Monte Carlo method, making it possible to accurately predict the radiation dose from the second spent nuclear fuel.

[0014] Another aspect of the present invention for achieving the above-mentioned object is an information processing device having a processor and a storage device, which stores a learning model that performs learning based on learning data that matches features including data relating to a first fuel pool in which a first spent nuclear fuel and a first spent control rod were stored in a flooded state in the past with labels including data indicating a radiation dose from the first spent nuclear fuel and the first spent control rod at a first predetermined position inside or outside the first fuel pool, and predicts the radiation dose from the second spent nuclear fuel and the second spent control rod at a second predetermined position inside or outside the second fuel pool by inputting data relating to the second fuel pool in which a second spent nuclear fuel and a second spent control rod are currently stored in a flooded state when the second fuel pool is at a predetermined water level into the learning model.

[0015] According to the information processing device of the present invention, when the water level of the second fuel pool drops from the water level that should normally be maintained to a specified water level, by inputting data related to the second fuel pool when the water level of the second fuel pool is at the specified water level into a learning model, it becomes possible to accurately predict the radiation dose from the second spent nuclear fuel and the second spent control rod at the desired position inside or outside the second fuel pool.

[0016] Another aspect of the present invention for achieving the above object is an information processing device, wherein the data relating to the first fuel pool includes information on dimensions within the first fuel pool, positions of the first spent nuclear fuel and the first spent control rod in the first fuel pool, a water level within the first fuel pool, radiation source strength of the first spent nuclear fuel and the first spent control rod, and the first predetermined position, and is a plurality of data prepared for each water level within the first fuel pool, data indicating the radiation dose from the first spent nuclear fuel and the first spent control rod at the first predetermined position is a plurality of data prepared for each water level within the first fuel pool, and the data relating to the second fuel pool is data including information on dimensions within the second fuel pool, positions of the second spent nuclear fuel and the second spent control rod in the second fuel pool, the predetermined water level within the second fuel pool, radiation source strength of the second spent nuclear fuel and the second spent control rod, and the second predetermined position.

[0017] According to the information processing device of the present invention, multiple pieces of feature and label data constituting the learning data are prepared for each water level of the first fuel pool, and the learning model learns from this learning data as supervised data, making it possible to accurately predict the radiation dose from the second spent nuclear fuel and the second spent control rod.

[0018] Another aspect of the present invention for achieving the above-mentioned object is an information processing device, in which data indicating the radiation dose from the first spent nuclear fuel and the first spent control rod at the first predetermined position is obtained for each water level in the first fuel pool by performing a numerical analysis based on a Monte Carlo calculation on data related to the first fuel pool.

[0019] According to the information processing device of the present invention, multiple pieces of feature and label data constituting the learning data are prepared for each water level of the first fuel pool by numerical analysis using the Monte Carlo method, making it possible to accurately predict the radiation dose from the second spent nuclear fuel and the second spent control rod.

[0020] Other problems and solutions disclosed in the present application will be made clear in the detailed description and drawings. Effect of the Invention

[0021] According to the present invention, it is possible to use a learning model to accurately predict the radiation dose from spent nuclear fuel stored in a fuel pool when the water level in the fuel pool is at a specified water level. [Brief description of the drawings]

[0022] [Figure 1] FIG. 2 is a side view showing an example of a state in which spent nuclear fuel 510 and spent control rods 520 are stored in a fuel pool 500 in a submerged state. [Diagram 2] 1 is a graph showing the relationship between the water level in the fuel pool 500 and the dose rate of radiation emitted by spent nuclear fuel 510 and spent control rods 520 at a predetermined position inside or outside the fuel pool 500. [Diagram 3] A block diagram showing the general configuration of a prediction system 100, including an information processing device 200 that predicts the radiation dose rate from spent nuclear fuel 510B and spent control rods 520B at a specified position inside or outside the fuel pool 500B when the current water level of the fuel pool 500B drops. [Figure 4] FIG. 2 is a diagram showing an example of features and labels constituting training data 242. [Diagram 5] A diagram showing the structure of a deep neural network, which is an example of a learning model 243. [Figure 6] 2 is a block diagram showing an example of hardware of an information processing device 200 used to realize the prediction system 100. FIG. [Figure 7] 13 is a flowchart showing a process performed by the information processing device 200 to learn a learning model 243 using learning data 242. [Figure 8]A flowchart showing the processing performed by the information processing device 200 using the learning model 243 to predict the radiation dose rate from spent nuclear fuel 510B and spent control rods 520B at the ceiling of the reactor building when the water level in the fuel pool 500B is at a predetermined water level. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0023] At least the following matters will become apparent from the description of this specification and the accompanying drawings. Hereinafter, the present invention will be described in detail based on an embodiment thereof with reference to the accompanying drawings.

[0024] 1 is a side view showing an example of a state in which spent nuclear fuel 510 and spent control rods 520 are stored in a submerged state in a fuel pool 500. For ease of explanation, the side surface of the fuel pool 500 on the near side of the page can be seen through to see a step 570 (a vertical step between a floor surface 530 and a floor surface 540) in the fuel pool 500, as well as the spent nuclear fuel 510 and spent control rods 520.

[0025] The fuel pool 500 is a storage facility for storing nuclear fuel (hereinafter referred to as "spent nuclear fuel 510") and control rods (hereinafter referred to as "spent control rods 520") that have been used for a certain period of time in a nuclear reactor and are now spent. Nuclear fuel refers to fuel containing uranium, plutonium, etc. that is used as fuel for nuclear power generation, and the spent nuclear fuel 510 refers to such nuclear fuel that has been used for a certain period of time. The control rod is composed of materials such as cadmium and boron carbide, and absorbs neutrons that are released when nuclear fuel undergoes nuclear fission in a nuclear reactor, thereby adjusting the thermal energy for generating steam and preventing the nuclear fuel from reaching a critical state in which nuclear fission continues spontaneously. The spent control rod 520 refers to such control rods that have been used for a certain period of time. Here, since the dose rate of radiation emitted by spent nuclear fuel 510 is greater than the dose rate of radiation emitted by spent control rods 520, spent nuclear fuel 510 is placed on the lowest floor surface 530 in fuel pool 500, and spent control rods 520 are placed on floor surface 540, which is higher than floor surface 530 in fuel pool 500, so that the distance from the top of spent nuclear fuel 510 to the water surface is longer than the distance from the top of spent control rods 520 to the water surface.

[0026] Water such as cooling water is supplied to the fuel pool 500 through a water supply device 550 so that the water level in the fuel pool 500 is at a certain level that submerges both the spent nuclear fuel 510 and the spent control rods 520. This water plays a role of shielding radiation emitted from the spent nuclear fuel 510 and the spent control rods 520, and a role of cooling water that mainly removes decay heat generated when the spent nuclear fuel 510 emits radiation and decays. The water supply device 550 supplies water to the fuel pool 500 so as to maintain a water level that can shield radiation emitted from the spent nuclear fuel 510 and the spent control rods 520. The water supply device 550 also includes a heat exchanger 560 that removes decay heat from the spent nuclear fuel 510 as one of the means for realizing the cooling function.

[0027] Fuel pool 500 has a substantially rectangular parallelepiped shape, although there is a step between floor surface 530 and floor surface 540. In addition, various distances in the vertical direction within fuel pool 500 are set to, for example, the following values ​​(1) to (7).

[0028] (1) Distance of the step between floor surface 530 and floor surface 540: 3.8 m (2) Distance between floor 530 and the top of spent nuclear fuel 510: 4.5 m (3) Distance between floor surface 540 and the top of spent control rod 520: 4.3 m (4) Distance between the top of the spent nuclear fuel 510 and the normal water level (solid line) required to shield radiation to a target dose rate (e.g., 10 mSv / h) or less: 7.0 m (5) Distance between the top of the spent nuclear fuel 510 and the minimum water level (dotted line) required to shield radiation so that it is below the target dose rate: 4.9 m (6) Distance between the top of the spent control rod 520 and the normal water level: 3.4 m (7) Distance between normal water level and minimum water level: 2.1m The normal water level is a water level that is set with a margin relative to the minimum water level when shielding the amount of radiation emitted by the spent nuclear fuel 510 and the spent control rods 520. In addition, when the operation of the water supply device 550 stops for a predetermined period due to the occurrence of a specified accident, the water level that is at the normal water level in the fuel pool 500 drops due to leakage or evaporation to a water level between the normal water level and the minimum water level, and it is confirmed that there are no problems with the shape and dimensions of the fuel pool 500 and the arrangement of the spent nuclear fuel 510 and the spent control rods 520.

[0029] FIG. 2 is a graph showing the relationship between the water level (horizontal axis: unit (m)) in the fuel pool 500 and the dose rate (vertical axis: unit (mSv / h)) of radiation emitted by the spent nuclear fuel 510 and the spent control rods 520 at a predetermined position inside or outside the fuel pool 500. The water level of 0 m on the horizontal axis is the water level at the top position of the spent nuclear fuel 510. In this embodiment, the predetermined position is, for example, the ceiling of the reactor building in which the fuel pool 500 is housed. The graph in FIG. 2 is an example of a simulation result obtained by performing a numerical analysis using the Monte Carlo method, using the above (1) to (7) and information indicating the radiation source strength of the spent nuclear fuel 510 and the spent control rods 520 as evaluation conditions.

[0030] The graph in FIG. 2 will be described below while the water level in the fuel pool 500 shown in FIG. 1 is lowered from the normal water level.

[0031] First, in the section where the water level in the fuel pool 500 drops from the normal water level to the minimum water level, the radiation dose rate at the ceiling of the reactor building increases linearly, and when the water level in the fuel pool 500 drops to the minimum water level, the radiation dose rate at the ceiling of the reactor building becomes 10 (mSv / h).

[0032] Next, in the section where the water level in the fuel pool 500 drops from the lowest water level to the water level at which the tops of the spent control rods 520 are exposed, the radiation dose rate at the ceiling of the reactor building continues to increase linearly with the same slope.

[0033] Next, in the section where the water level in the fuel pool 500 drops from the water level at which the tops of the spent control rods 520 are exposed to a water level just under 2 m higher than the tops of the spent nuclear fuel 510, the radiation dose rate at the ceiling of the reactor building becomes almost constant.

[0034] In other words, in the section where the water level in the fuel pool 500 is between the normal water level and a water level just under 2 m higher than the top of the spent nuclear fuel 510, the radiation dose rate at the ceiling of the reactor building is more affected by the dose rate of radiation emitted by the spent control rods 520 than by the dose rate of radiation emitted by the spent nuclear fuel 510.

[0035] Next, in the section where the water level in the fuel pool 500 drops from a water level slightly less than 2 m higher than the top of the spent nuclear fuel 510 to the top of the spent nuclear fuel 510, the radiation dose rate at the ceiling of the reactor building increases linearly again. In other words, in the section where the water level in the fuel pool 500 is between a water level slightly less than 2 m higher than the top of the spent nuclear fuel 510 and the top of the spent nuclear fuel 510, the radiation dose rate at the ceiling of the reactor building is more significantly affected by the dose rate of radiation emitted by the spent nuclear fuel 510 than by the dose rate of radiation emitted by the spent control rods 520.

[0036] 2 shows a one-to-one relationship between the change in water level and the change in dose rate in the fuel pool 500. In other words, since the graph in Fig. 2 shows the unique characteristics of only the fuel pool 500, the one-to-one corresponding values ​​of the water level and the dose rate can be read, and multiple data sets consisting of the water level and the dose rate can be extracted, and these multiple data sets can be used as the learning data 242 described later.

[0037] FIG. 3 is a block diagram showing a schematic configuration of a prediction system 100 including an information processing device 200 for predicting a radiation dose rate from spent nuclear fuel 510B and spent control rods 520B at a predetermined position inside or outside the fuel pool 500B when the water level of the current fuel pool 500B drops. In this embodiment, the past fuel pool (first fuel pool) is indicated by fuel pool 500A, and the current fuel pool (second fuel pool) is indicated by fuel pool 500B. Also, the past spent nuclear fuel (first spent nuclear fuel) and spent control rods (first spent control rods) are indicated by spent nuclear fuel 510A and spent control rods 520A, respectively, and the current spent nuclear fuel (second spent nuclear fuel) and spent control rods (second spent control rods) are indicated by spent nuclear fuel 510B and spent control rods 520B, respectively. Also, the above-mentioned predetermined position is, for example, the ceiling of a reactor building.

[0038] The prediction system 100 is a system that predicts the radiation dose rate from the spent nuclear fuel 510B and the spent control rods 520B on the ceiling of the reactor building by AI (Artificial Intelligence), when the water level in the fuel pool 500B drops from the normal water level to a predetermined water level, and outputs the prediction result to the outside. Note that since the radiation dose rate is the radiation dose per unit time, the prediction system 100 may predict the radiation dose taking into account the storage time of the spent nuclear fuel 510B and the spent control rods 520B in the fuel pool 500B.

[0039] The prediction system 100 includes an information processing device 200 and an external computer 300 as means for realizing the above functions. The information processing device 200 and the external computer 300 are connected in a state in which two-way communication is possible via a communication network 400. The communication network 400 is, for example, a local area network (LAN), a wide area network (WAN), a dedicated line, a power line communication network, various public communication networks, etc.

[0040] The information processing device 200 includes an input unit 210 , a prediction unit 220 , an output unit 230 , and a storage unit 240 .

[0041] The memory unit 240 stores a control program 241, learning data 242, and a learning model 243.

[0042] The control program 241 is a program executed by the information processing device 200 when the prediction unit 220 predicts the radiation dose rate from the spent nuclear fuel 510B and the spent control rods 520B at the ceiling of the reactor building when the water level in the fuel pool 500B drops from the normal water level to a predetermined water level.

[0043] The learning data 242 is supervised data in which feature values ​​and labels, which are actual results of the fuel pool 500A, are associated with each other and are prepared in advance to allow the prediction unit 220 to predict the radiation dose rate from the spent nuclear fuel 510B and the spent control rods 520B at the ceiling of the reactor building when the water level in the fuel pool 500B drops from the normal water level to a predetermined water level. In order to improve the learning accuracy of the learning model 243, the learning data 242 is a plurality of data prepared in advance for fuel pools 500B of various shapes and dimensions in which only the spent nuclear fuel 510B with various dose rates or the spent nuclear fuel 510B with various dose rates and the spent control rods 520B are stored. In this embodiment, the predetermined water level is the water level at the top of the spent nuclear fuel 510.

[0044] FIG. 4 is a diagram showing an example of features and labels constituting the training data 242.

[0045] The feature amount includes, for example, the following data (A) to (P).

[0046] (A) Distance between floor 530 and normal water level in fuel pool 500A (11.5 m) (B) Distance between floor 530 and floor 540 in fuel pool 500A (3.8 m) (C) Distance between floor 530 of fuel pool 500A and the ceiling of the reactor building (22.4 m) (D) Distance between the top of spent fuel 510A and the lowest water level (4.9m) (E) Water level in fuel pool 500A (0~7.0m) (F) Type of spent nuclear fuel 510A (G) Burnup of spent nuclear fuel 510A (H) Specific power of spent nuclear fuel 510A (I) Source strength of spent nuclear fuel 510A (J) Cooling period of spent nuclear fuel 510A (K) Number of spent nuclear fuel 510A stored (L) Type of used control rod 520A (M) Irradiation period of spent control rod 520A (N) Cooling period of spent control rod 520A (O) Source strength of spent control rod 520A (P) Number of used control rods 520A stored Here, (A) indicates the highest water level when the fuel pool 500A is operating normally, and is adopted as one of the feature quantities.

[0047] (B) is the distance of step 570 between different floor surfaces 530, 540 in fuel pool 500A. If the amount of water in fuel pool 500A decreases at a constant flow rate, the rate at which the water level drops near step 570 in fuel pool 500A will change. In addition, if spent control rods 520A, which are a different radiation source from spent nuclear fuel 510A, are stored on floor surface 540, which is higher than floor surface 530, the amount of radiation in the air will increase when spent control rods 520A are exposed from the water. In order to take these effects into account, (B) is adopted as one of the feature quantities.

[0048] (C) is a feature required as the distance from the radiation source, with the ceiling of the reactor building set as the measurement point for radiation emitted from the spent nuclear fuel 510A and the spent control rods 520A. It is also adopted as one of the feature amounts when considering the effect of radiation emitted from the spent nuclear fuel 510A and the spent control rods 520A being reflected by the ceiling of the reactor building.

[0049] (D) is adopted as one of the characteristic quantities in order to take into account the effect of radiation emitted by spent nuclear fuel 510A when the water level of fuel pool 500A drops below the minimum water level.

[0050] Since the learning data 242 is data that associates the radiation dose rate from the spent nuclear fuel 510A and the spent control rods 520A at the ceiling of the reactor building with each of multiple water levels in the fuel pool 500A between the normal water level and the top of the spent nuclear fuel 510A, (E) is adopted as one of the features.

[0051] As the nuclear fuel, any of a number of types of nuclear fuel is used, such as plutonium, plutonium compounds, materials containing plutonium and plutonium compounds, uranium 233, uranium 233 compounds, materials containing uranium 233 and uranium 233 compounds, etc. In order to take into account the effects according to the type of nuclear fuel, (F) is adopted as one of the characteristic quantities.

[0052] Burnup is the thermal energy generated per unit weight of fuel by the nuclear fission reaction of spent nuclear fuel 510A. The composition of the radioactive material contained in the nuclear fuel changes during the burning and cooling processes, and the type and intensity distribution of radiation emitted from the fuel also change, so (G) is used as one of the characteristic quantities.

[0053] The specific power is the thermal output per unit mass of the spent nuclear fuel 510A, and indicates the level of thermal burden of the spent nuclear fuel 510A. Since it is possible to calculate the irradiation period of the spent nuclear fuel 510A in combination with the burnup, (H) is adopted as one of the characteristic quantities.

[0054] The radiation source intensity of the spent nuclear fuel 510A varies depending on the type, burning, and cooling of the spent nuclear fuel 510A. Therefore, (I) is adopted as one of the characteristic quantities.

[0055] (J) is adopted as one of the features to take into account that the composition of the radioactive material in the spent nuclear fuel 510A changes over the cooling period, resulting in changes in the type and intensity distribution of radiation.

[0056] (K) is adopted as one of the feature quantities in order to take into consideration the planar spread of spent nuclear fuel 510A on floor surface 530 in fuel pool 500A.

[0057] Neutron absorbing materials such as hafnium and boron carbide are used in the spent control rods 520A, and (L) is adopted as one of the characteristic quantities to take into account the energy distribution of radiation emitted from the spent control rods 520A as a result of the activation of these materials.

[0058] (M) is adopted as one of the characteristic quantities in order to take into account the energy distribution of radiation emitted from the spent control rod 520A due to activation of the spent control rod 520A.

[0059] (N) is adopted as one of the characteristic quantities in order to take into account the energy distribution of radiation emitted from the spent control rod 520A due to activation of the spent control rod 520A.

[0060] (O) varies depending on the degree and type of activation of the spent control rod 520A. Therefore, (N) is adopted as one of the characteristic quantities.

[0061] (P) is adopted as one of the feature quantities in order to take into consideration the planar spread of the spent control rods 520A on the floor surface 540 in the fuel pool 500A.

[0062] On the other hand, the label contains the data shown in (Q) below.

[0063] (Q) The radiation dose rate from the spent nuclear fuel 510A and the spent control rods 520A at the ceiling of the reactor building when the water level in the fuel pool 500A is somewhere between the normal water level and the top of the spent nuclear fuel 510A. The data (E)(Q) are multiple data sets consisting of combinations of water levels and dose rates obtained for multiple water levels in fuel pool 500A by, for example, numerical analysis based on the Monte Carlo method.

[0064] The learning model 243 is a machine learning model formed to predict the radiation dose rate of the spent nuclear fuel 510B and the spent control rod 520B at the ceiling of the reactor building when the water level in the fuel pool 500B is between the normal water level and the top of the spent nuclear fuel 510B, by learning using the learning data 242 and inputting the same type of data as the above data (A) to (P) which are the feature amounts of the learning data 242, for the fuel pool 500B. Here, the water level in the fuel pool 500B corresponding to the feature amount (E) is one water level. The more the learning model 243 learns using the learning data 242, the more accurate the prediction becomes. In this embodiment, the learning model 243 is, for example, a deep neural network (DNN), but may be another type of model such as a gradient boosting (GBDT).

[0065] FIG. 5 is a diagram showing the structure of a deep neural network, which is an example of the learning model 243.

[0066] The learning model 243 includes three layers: an input layer 2431, an intermediate layer 2432, and an output layer 2433. The input layer 2431 is a layer to which the same type of data as the feature quantities (A) to (P) constituting the learning data 242 for the fuel pool 500B is input. The intermediate layer 2432 is a layer including one or more hidden layers consisting of one or more nodes including parameters adjusted by performing learning using the learning data 242. Based on the data input to the input layer 2431, the intermediate layer 2432 predicts the radiation dose rate from the spent nuclear fuel 510B and the spent control rods 520B at the ceiling of the reactor building when the water level in the fuel pool 500B is any water level between the normal water level and the top of the spent nuclear fuel 510B. The output layer 2433 is a layer that outputs the prediction result by the intermediate layer 2432.

[0067] When data relating to a fuel pool 500B of the same type as features (A) to (P) is input, the learning model 243 refers to learning data 242 relating to a fuel pool 500A similar to these data, and predicts the radiation dose rate from the spent nuclear fuel 510B and spent control rods 520B at the ceiling of the reactor building when the water level in the fuel pool 500B is anywhere between the normal water level and the top of the spent nuclear fuel 510B.

[0068] 3, when the information processing device 200 predicts the dose rate of radiation due to the spent nuclear fuel 510B and the spent control rods 520B at the ceiling of the reactor building when the water level in the fuel pool 500B is any water level between the normal water level and the top of the spent nuclear fuel 510B, the input unit 210 receives data related to the same type of fuel pool 500B as the features (A) to (P). The data input to the input unit 210 is then input to the prediction unit 220.

[0069] The prediction unit 220 inputs the above data input to the input unit 210 into the input layer 2431 of the learning model 243, and instructs the learning model 243 to predict the radiation dose rate from the spent nuclear fuel 510B and the spent control rods 520B at the ceiling of the reactor building.

[0070] The output unit 230 outputs the prediction result by the learning model 243 to the outside in accordance with an instruction from the prediction unit 220 triggered by the output of the prediction result from the intermediate layer 2432 to the output layer 2433. In addition, when the predicted radiation dose rate exceeds a predetermined dose rate, the output unit 230 may also output a message such as a warning. Possible output methods for the output unit 230 to output the prediction result or message include, for example, a method of displaying and outputting the prediction result or message on a display panel, a method of outputting the prediction result or message as sound from a speaker, a method of printing and outputting the prediction result or message on a paper medium, and the like.

[0071] The external computer 300 has a memory unit 310, and stores data indicating prediction results and messages output from the learning model 243 in the memory unit 310, and reads out these data from the memory unit 310 and outputs them to the information processing device 200 as necessary.

[0072] FIG. 6 is a block diagram showing an example of hardware of an information processing device 200 used to realize the prediction system 100. As shown in FIG.

[0073] The information processing device 200 includes a processor 2010, a main memory device 2020, an auxiliary memory device 2030, an input device 2040, an output device 2050, and a communication device 2060. The information processing device 200 is, for example, a personal computer, an office computer, various server devices, a general-purpose machine, or the like. The information processing device 200 may be realized, in whole or in part, by using virtual information processing resources provided by using a virtualization technology, such as a virtual server provided by a cloud system. The prediction system 1 may be realized by using a plurality of information processing devices 200 connected to each other so as to be able to communicate with each other.

[0074] The processor 2010 is configured using, for example, a Central Processing Unit (CPU), a Micro Processing Unit (MPU), a Graphics Processing Unit (GPU), a Field Programmable Gate Array (FPGA), an Application Specific Integrated Circuit (ASIC), an Artificial Intelligence (AI) chip, etc.

[0075] The main memory device 2020 is a device that stores programs and data, and is, for example, a Read Only Memory (ROM), a Random Access Memory (RAM), or a non-volatile memory (NVRAM (Non Volatile RAM)).

[0076] The auxiliary storage device 2030 is, for example, a solid state drive (SSD), a hard disk drive, an optical storage device (CD (Compact Disc), DVD (Digital Versatile Disc), etc.), a storage system, an IC card, an SD card, a read / write device for a recording medium such as an optical recording medium, a storage area of ​​a cloud server, etc. Programs and data can be read into the auxiliary storage device 2030 via a recording medium reader or a communication device 2060. The programs and data stored in the auxiliary storage device 2030 are read into the main storage device 2020 as needed.

[0077] The input device 2040 is an interface that accepts input from the outside, and is, for example, a keyboard, a mouse, a touch panel, a card reader, a pen-input tablet, a voice input device, or the like.

[0078] The output device 2050 is an interface that outputs various information such as the progress of processing and processing results. The output device 2050 is, for example, a display device (LCD (Liquid Crystal Display), graphic card, etc.) that visualizes the various information described above, a device that converts the various information described above into voice (voice output device (speaker, etc.)), or a device that converts the various information described above into text (printer, etc.). The information processing device 200 may be configured to input and output information to and from other devices via the communication device 2060.

[0079] The input device 2040 and the output device 2050 constitute a user interface that receives and presents information to and from the user.

[0080] The communication device 2060 is a device that realizes communication (wired communication or wireless communication) with other devices via a communication infrastructure such as a communication network 400, and is configured using, for example, a NIC (Network Interface Card), a wireless communication module, a USB module, etc.

[0081] In addition, the information processing device 200 may be implemented with, for example, an operating system, a file system, a DBMS (Data Base Management System) (relational database, NoSQL, etc.), a KVS (Key-Value Store), etc.

[0082] The functions of the prediction system 100 are realized by the processor 2010 of the information processing device 200 executing a program (control program) loaded into the main memory device 2020, or by the functions of the hardware (FPGA, ASIC, AI chip, etc.) constituting the prediction system 100. For example, the memory unit 240 in FIG. 3 is realized by the main memory device 2020 and the auxiliary memory device 2030, the input unit 210 in FIG. 3 is realized by the input device 2040, the prediction unit 220 in FIG. 3 is realized by the processor 2010, and the output unit 330 in FIG. 3 is realized by the output device 2050 and the communication device 2060.

[0083] FIG. 7 is a flowchart showing a process performed by the information processing device 200 to learn the learning model 243 using the learning data 242.

[0084] First, the information processing device 200 generates learning data 242 in which the feature amounts shown in (A) to (P) are associated with the labels shown in (Q) (S1000).

[0085] Next, the information processing device 200 uses the learning data 242 to train the learning model 243 (S1010). Note that the information processing device 200 may verify the prediction accuracy of the trained learning model 243. In this case, the information processing device 200 prepares learning data and verification data as the learning data 242, trains the learning model 243 using the learning data, and verifies the learning model 243 using the verification data.

[0086] Figure 8 is a flowchart showing the processing performed by the information processing device 200 using the learning model 243 to predict the radiation dose rate from the spent nuclear fuel 510B and spent control rods 520B at the ceiling of the reactor building when the water level in the fuel pool 500B is any level between the normal water level and the top of the spent nuclear fuel 510B.

[0087] First, the information processing device 200 determines whether data related to the same type of fuel pool 500B as the features (A) to (P) has been input to the input unit 210 (S2000). When the information processing device 200 determines that the above data has not been input to the input unit 210 (S2000: NO), it executes the process of step S2000 again.

[0088] Next, when the information processing device 200 determines that the above data has been input to the input section 210 (S2000: YES), it uses the learning model 243 to predict the radiation dose rate from the spent nuclear fuel 510B and the spent control rods 520B at the ceiling of the reactor building when the water level in the fuel pool 500B is at a specified water level (S2010).

[0089] Next, the information processing device 200 outputs the radiation dose rate prediction result in step S2010 and, if necessary, a message from the output unit 230 (S2020).

[0090] As described above, the information processing device 200 having the processor 2010 and the memory device (main memory device 2020, auxiliary memory device 2030) stores the learning model 243 that performs learning based on the learning data 242 that associates the feature amount including the data related to the fuel pool 500A in which the spent nuclear fuel 510A and the spent control rods 520A were stored in a flooded state in the past with the label including the data indicating the dose rate of radiation due to the spent nuclear fuel 510A and the spent control rods 520A at the ceiling of the reactor building in which the fuel pool 500A is contained, and predicts the dose rate of radiation due to the spent nuclear fuel 510B and the spent control rods 520B at the ceiling of the reactor building in which the fuel pool 500B is contained by inputting the data related to the fuel pool 500B in which the spent nuclear fuel 510B and the spent control rods 520B are currently stored in a flooded state into the learning model 243 when the fuel pool 500B is at a predetermined water level.

[0091] According to the information processing device 200, when the water level of the fuel pool 500B drops from the water level that should be maintained to a specified water level, by inputting data related to the fuel pool 500B when the water level of the fuel pool 500B is at the specified water level into the learning model 243, it becomes possible to accurately predict the radiation dose rate from the spent nuclear fuel 510B and the spent control rods 520B at the ceiling of the reactor building.

[0092] In the information processing device 200, the data related to the fuel pool 500A includes information on the dimensions of the fuel pool 500, the arrangement positions of the spent nuclear fuel 510A and the spent control rods 520A in the fuel pool 500A, multiple water levels in the fuel pool 500A, the radiation source strength of the spent nuclear fuel 510A and the spent control rods 520A, and the position of the ceiling of the reactor building, and is multiple data prepared for each of multiple water levels in the fuel pool 500A. The data indicating the dose rate of radiation from spent nuclear fuel 510B and spent control rods 520A is a plurality of data prepared for each of a plurality of water levels in fuel pool 500A, and the data relating to fuel pool 500B is data including information on the dimensions within fuel pool 500B, the positions of spent nuclear fuel 510B and spent control rods 520B in fuel pool 500B, the specified water level within fuel pool 500B, the radiation source strength of spent nuclear fuel 510B and spent control rods 520B, and the position of the ceiling of the reactor building.

[0093] According to the information processing device 200, multiple pieces of feature and label data constituting the learning data 242 are prepared for each water level of the fuel pool 500A, and the learning model 243 learns using this learning data 242 as supervised data, making it possible to accurately predict the radiation dose from the spent nuclear fuel 510B and the spent control rods 520B.

[0094] In addition, in the information processing device 200, data indicating the radiation dose rate from the spent nuclear fuel 510A and the spent control rods 520A at the ceiling of the reactor building is obtained for each water level in the fuel pool 500A by performing numerical analysis based on Monte Carlo calculations on the data related to the fuel pool 500A.

[0095] According to the information processing device 200, the feature and label data constituting the learning data 242 are prepared in multiple pieces for each water level of the fuel pool 500A by numerical analysis using the Monte Carlo method, making it possible to accurately predict the radiation dose rate from the spent nuclear fuel 510B and the spent control rods 520B.

[0096] The above-described embodiment is for the purpose of facilitating understanding of the present invention, and is not intended to limit the present invention. The present invention may be modified or improved without departing from the spirit of the present invention, and equivalents thereof are also included in the present invention.

[0097] In this embodiment, a fuel pool 500B has a step 570, and both spent nuclear fuel 510B and spent control rods 520B are stored in the fuel pool 500B. The present embodiment describes predicting the dose rate of radiation from spent nuclear fuel 510B and spent control rods 520B when the water level in the fuel pool 500B drops to a predetermined water level, but is not limited to this.

[0098] For example, in a case where the spent nuclear fuel 510B and the spent nuclear fuel 520B are stored in the fuel pool 500B not having the step 570, the dose rate of radiation from the spent nuclear fuel 510B and the spent control rods 520B when the water level in the fuel pool 500B drops to a predetermined water level may be predicted. Also, in a case where only the spent nuclear fuel 510B is stored in the fuel pool 500B not having the step 570, the dose rate of radiation from the spent nuclear fuel 510B when the water level in the fuel pool 500B drops to a predetermined water level may be predicted. As the learning data 242, it is necessary to prepare feature amounts corresponding to both cases.

[0099] In addition, in the example shown as the present embodiment, the measurement point is the ceiling of the reactor building, but the parameters of the learning data 242 may be set to take into account the horizontal spread, such as on the plane of the floor of the reactor building in which the fuel pool 500 is installed, and the evaluation is not limited to evaluating the dose rate in the vertical direction from the fuel pool 500.

[0100] In addition, while the example shown as the present embodiment is intended to evaluate the dose rate, this embodiment may be applied to evaluate the water level that should be maintained from the perspective of dose management, or to evaluate other parameters, and further, optimization may be performed by fixing some parameters and randomly varying other parameters to find the optimal solution. [Explanation of symbols]

[0101] 100 Prediction System 200 Information processing device 210 Input section 220 Prediction Department 230 Output section 240,310 storage section 241 Control Program 242 Training Data 243 Learning Model 300 External Computer 400 Communication Network 500,500A,500B Fuel Pool 510,510A,510B Spent nuclear fuel 520, 520A, 520B Used control rods 530,540 Floor 550 Water supply equipment 560 heat exchanger 570 Steps 2010 Processors 2020 Main storage 2030 Auxiliary storage 2040 Input Device 2050 Output Device 2060 Communication Equipment 2431 Input Layer 2432 Middle Class 2433 Output Layer

Claims

1. A processor and a storage device are included. storing a learning model that performs learning based on learning data that associates a feature amount including data related to a first fuel pool in which a first spent nuclear fuel was stored in a flooded state in the past with a label including data indicating a radiation dose from the first spent nuclear fuel at a first predetermined position inside or outside the first fuel pool; Data relating to the second fuel pool in which the second spent nuclear fuel is currently stored in a flooded state when the second fuel pool is at a predetermined water level is input to the learning model, thereby predicting the radiation dose from the second spent nuclear fuel at a second predetermined position inside or outside the second fuel pool. Information processing device.

2. 2. The information processing device according to claim 1, The data relating to the first fuel pool is Dimensions within the first fuel pool; a position of the first spent nuclear fuel in the first fuel pool; the water level in the first fuel pool; a source strength of the first spent nuclear fuel; The first predetermined position, a plurality of data sets prepared for each water level in the first fuel pool, the data indicating the radiation dose from the first spent nuclear fuel at the first predetermined position is a plurality of data prepared for each water level in the first fuel pool, The data relating to the second fuel pool is Dimensions within the second fuel pool; a position of the second spent nuclear fuel in the second fuel pool; a predetermined water level in the second fuel pool; the source strength of the second spent nuclear fuel; The second predetermined position, It is data that contains information on Information processing device.

3. 3. The information processing device according to claim 2, The data indicating the amount of radiation from the first spent nuclear fuel at the first predetermined position includes: A numerical analysis based on the Monte Carlo method is performed on the data related to the first fuel pool to obtain a water level in the first fuel pool. Information processing device.

4. A processor and a storage device are included. storing a learning model that performs learning based on learning data that associates a feature amount including data related to a first fuel pool in which a first spent nuclear fuel and a first spent control rod have been stored in a flooded state in the past with a label including data indicating a radiation dose from the first spent nuclear fuel and the first spent control rod at a first predetermined position inside or outside the first fuel pool; Data relating to the second fuel pool, in which the second spent nuclear fuel and the second spent control rod are currently stored in a flooded state, when the second fuel pool is at a predetermined water level is input to the learning model, thereby predicting the radiation dose from the second spent nuclear fuel and the second spent control rod at a second predetermined position inside or outside the second fuel pool. Information processing device.

5. 5. The information processing device according to claim 4, The data relating to the first fuel pool is Dimensions within the first fuel pool; a position of the first spent nuclear fuel and the first spent control rod in the first fuel pool; the water level in the first fuel pool; a source strength of the first spent nuclear fuel and the first spent control rod; The first predetermined position, a plurality of data sets prepared for each water level in the first fuel pool, the data indicating the radiation dose from the first spent nuclear fuel and the first spent control rod at the first predetermined position is a plurality of data prepared for each water level in the first fuel pool, The data relating to the second fuel pool is Dimensions within the second fuel pool; a position of the second spent nuclear fuel and the second spent control rod in the second fuel pool; a predetermined water level in the second fuel pool; a source strength of the second spent nuclear fuel and the second spent control rod; The second predetermined position, It is data that contains information on Information processing device.

6. 6. The information processing device according to claim 5, Data indicating a radiation dose from the first spent nuclear fuel and the first spent control rod at the first predetermined position, A numerical analysis based on the Monte Carlo method is performed on the data related to the first fuel pool to obtain a water level in the first fuel pool. Information processing device.

7. An information processing method of an information processing device having a processor and a storage device, storing a learning model that performs learning based on learning data that associates a feature amount including data related to a first fuel pool in which a first spent nuclear fuel was stored in a flooded state in the past with a label including data indicating a radiation dose from the first spent nuclear fuel at a first predetermined position inside or outside the first fuel pool; Data relating to the second fuel pool in which the second spent nuclear fuel is currently stored in a flooded state when the second fuel pool is at a predetermined water level is input to the learning model, thereby predicting the radiation dose from the second spent nuclear fuel at a second predetermined position inside or outside the second fuel pool. Information processing methods.

8. An information processing method of an information processing device having a processor and a storage device, storing a learning model that performs learning based on learning data that associates a feature amount including data related to a first fuel pool in which a first spent nuclear fuel and a first spent control rod have been stored in a flooded state in the past with a label including data indicating a radiation dose from the first spent nuclear fuel and the first spent control rod at a first predetermined position inside or outside the first fuel pool; Data relating to the second fuel pool, in which the second spent nuclear fuel and the second spent control rod are currently stored in a flooded state, when the second fuel pool is at a predetermined water level is input to the learning model, thereby predicting the radiation dose from the second spent nuclear fuel and the second spent control rod at a second predetermined position inside or outside the second fuel pool. Information processing methods.

9. A computer having a processor and a storage device, a first process for storing a learning model that learns based on learning data that associates a feature amount including data related to a first fuel pool in which a first spent nuclear fuel was stored in a flooded state in the past with a label including data indicating a radiation dose from the first spent nuclear fuel at a first predetermined position inside or outside the first fuel pool; a second process for predicting a radiation dose from the second spent nuclear fuel at a second predetermined position inside or outside the second fuel pool by inputting data related to the second fuel pool in which the second spent nuclear fuel is currently stored in a flooded state at a predetermined water level into the learning model; A program that executes the following.

10. A computer having a processor and a storage device, a first process for storing a learning model that learns based on learning data that associates a feature amount including data related to a first fuel pool in which a first spent nuclear fuel and a first spent control rod have been stored in a flooded state in the past with a label including data indicating a radiation dose from the first spent nuclear fuel and the first spent control rod at a first predetermined position inside or outside the first fuel pool; a second process for predicting a radiation dose from the second spent nuclear fuel and the second spent control rod at a second predetermined position inside or outside the second fuel pool by inputting data related to the second fuel pool in which the second spent nuclear fuel and the second spent control rod are currently stored in a flooded state when the second fuel pool is at a predetermined water level into the learning model; A program that executes the following.

Images