Radioactive waste processing system and radioactive waste processing method

Abstract

To adjust, in immobilizing radioactive waste, the amount of blending to be optimal according to the properties of the radioactive waste to prevent the generation of a failure in mixing and kneading.SOLUTION: A radioactive waste processing system 1 acquires information on the properties of radioactive waste to be subjected to immobilization processing, acquires information on the properties of immobilization material that immobilizes the radioactive waste, inputs the information on the properties of radioactive waste and the information on the properties of immobilization material, as input data D5, to a blending prediction model M3 that has been machine-learned, and on the basis of output data D6 output from the blending prediction model M3, determines the amount of blending of the radioactive waste and the immobilization material to be put into a mixing and kneading container 14 that mixes and kneads the radioactive waste and the immobilization material.SELECTED DRAWING: Figure 5

Description

[Technical field]

[0001] FIELD OF THE DISCLOSURE Embodiments of the present invention relate to radioactive waste processing. [Background technology]

[0002] In nuclear power plants, granular or powdered ion exchange resins are used in large quantities to purify condensate (turbine cooling water) or wastewater from chemical analysis laboratories. Ion exchange resins are regenerated and reused, and the waste liquid generated during regeneration becomes radioactive waste. In addition, resins whose processing capacity has decreased due to repeated use also eventually become radioactive waste. Such low-level radioactive waste is treated by a solidification device remotely operated from outside the controlled area using hydraulic solidification materials such as ordinary Portland cement, and disposed of in a solidified state in drum containers. In addition, among low-level radioactive wastes with relatively low levels of radioactivity like used ion exchange resins, liquid wastes such as concentrated waste liquids, and solid wastes such as incineration ash after incineration of combustible radioactive waste are also treated in the same manner.

[0003] Government agencies set standards for the acceptance of solidified waste after processing, and the range of design specifications for the waste mix is ​​set to meet these standards. For example, when solidifying radioactive waste using cement, the unconfined compressive strength of the solidified radioactive waste must be 1470 kPa or more.

[0004] Radioactive waste generated from ion exchange resins used to purify cooling water in condensate systems or wastewater from chemical analysis laboratories may have ion concentrations in the waste liquid or ion components attached to the waste resin that may vary depending on the operating conditions. Despite this, the waste is transferred to the treatment equipment without any confirmation of changes in its properties. In addition, for concentrated waste liquid or incineration ash, the properties of the waste before concentration or incineration cannot be controlled, so the waste after treatment is not homogenous. Due to the nature of the waste handled in this way, the waste properties cannot be controlled to a constant state, but the treatment equipment cannot change the operating conditions or the mix depending on the mixed state because of the influence on the subsequent stages. The influence on the subsequent stages refers to the influence when the solidified body is buried.

[0005] Hydraulic solidification materials commonly used for solidifying low-level radioactive waste include ordinary Portland cement and blast-furnace cement. The properties of these cements are also affected by subtle differences in gypsum or small mixed ingredients depending on the manufacturer, manufacturing plant, lot, etc., and the mixing state may change even with the same mix. In addition, the mixing state may change even with the same mix due to the influence of weathering in the cement storage environment. If mixing failure occurs, it becomes difficult to stably operate the radioactive waste solidification equipment. In addition, it becomes necessary for workers to approach the solidification equipment in the controlled area to carry out recovery work or operate it, which takes time and makes it difficult to carry out the solidification process efficiently.

[0006] A method has also been proposed to estimate the fluidity of concrete by using images of the mixing state in a concrete mixer. However, this method requires the mix to be adjusted each time. [Prior art documents] [Patent documents]

[0007] [Patent Document 1] Patent No. 6941064 [Non-patent literature]

[0008] [Non-Patent Document 1] "Confirmation method for disposal of homogeneous / uniform solidified bodies and filled solidified bodies (partial revision)" (April 2008 JNES-SS-0801) Summary of the Invention [Problem to be solved by the invention]

[0009] The problem to be solved by the present invention is to adjust the optimum mixing amount according to the properties of the radioactive waste when solidifying the radioactive waste, and to suppress the occurrence of poor mixing. [Means for solving the problem]

[0010] A radioactive waste treatment system according to an embodiment of the present invention includes one or more computers configured to acquire property information of the radioactive waste to be solidified, acquire property information of the solidification material that solidifies the radioactive waste, input the property information of the radioactive waste and the property information of the solidification material as input data into a machine-learned blending prediction model, and determine the blending amounts of the radioactive waste and the solidification material to be put into a mixing vessel in which the radioactive waste and the solidification material are mixed, based on output data output from the blending prediction model. Effect of the Invention

[0011] According to an embodiment of the present invention, when solidifying radioactive waste, it is possible to adjust the optimum blending amount according to the properties of the radioactive waste, and to suppress the occurrence of poor mixing. [Brief description of the drawings]

[0012] [Figure 1] FIG. 1 is a configuration diagram showing a radioactive waste treatment system. [Diagram 2] 1 is a flow chart showing radioactive waste treatment. [Diagram 3] FIG. 2 is an explanatory diagram showing the flow of input and output data of the waste prediction model. [Figure 4] FIG. 4 is an explanatory diagram showing the flow of input and output data of a material prediction model. [Diagram 5] An explanatory diagram showing the input / output data and judgment flow of the blend prediction model. [Figure 6] 11 is a flowchart showing a kneading state determination process. [Figure 7] FIG. 13 is an explanatory diagram showing the input / output data and judgment flow of the addition prediction model. [Figure 8] 1 is a graph showing the relationship between the solid content concentration and the viscosity of the kneaded product. [Figure 9] 1 is a graph showing the relationship between the specific surface area and the viscosity of a kneaded material. [Figure 10] Graph showing the relationship between the ettringite content and the viscosity of the kneaded product. [Figure 11] 4 is a graph showing the relationship between the current value of the kneading device and the viscosity of the kneaded material. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] Hereinafter, embodiments of a radioactive waste treatment system and a radioactive waste treatment method will be described in detail with reference to the drawings.

[0014] Reference numeral 1 in Fig. 1 denotes a radioactive waste treatment system according to this embodiment. A radioactive waste treatment method is carried out using this radioactive waste treatment system 1. The radioactive waste treatment system 1 mixes radioactive waste with a solidification material such as cement, and forms a solidified body suitable for burial disposal. Here, the radioactive waste treatment system 1 determines the mixing ratio of the radioactive waste and the solidification material when mixing them, using a trained model in which parameters have been machine-learned in advance.

[0015] The radioactive waste treatment system 1 comprises a waste storage tank 2, a material storage tank 3, a water storage tank 4, an admixture storage tank 5, a waste measuring tank 6, a material measuring tank 7, a water metering tank 8, an admixture metering tank 9, a waste information acquisition device 10, a kneading device 11, a status acquisition device 12, and a control computer 13.

[0016] The radioactive waste treatment system 1 may include components other than those shown in FIG. 1, and some of the components shown in FIG. 1 may be omitted.

[0017] The waste storage tank 2 is a tank for storing radioactive waste to be fed into the kneading device 11. The material storage tank 3 is a tank for storing solidification material to be fed into the kneading device 11. The water storage tank 4 is a tank for storing water to be fed into the kneading device 11. The admixture storage tank 5 is a tank for storing admixture to be fed into the kneading device 11.

[0018] The kneading device 11 kneads the radioactive waste, the solidification material, and other materials fed from the respective tanks. The kneading device 11 includes a kneading container 14, an agitation motor 15, an agitation shaft 16, and an agitation blade 17.

[0019] Radioactive waste, solidification material, water, and admixture are charged into the kneading vessel 14. These charges are then kneaded inside the kneading vessel 14. The stirring motor 15 is connected to a stirring shaft 16, and a stirring blade 17 is fixed to the lower end of this stirring shaft 16. When the stirring motor 15 is driven, the stirring blade 17 rotates via the stirring shaft 16. The rotation of this stirring blade 17 kneads the radioactive waste and solidification material.

[0020] The kneading device 11 has a function of kneading while feeding materials into the kneading container 14. In other words, feeding of materials into the kneading container 14 continues while kneading continues. For example, when feeding materials into the kneading container 14 in two batches, one at the front stage and one at the rear stage, the blending amount of the materials fed into the front stage may be different from the blending amount of the materials fed into the rear stage. When the blending amount of the materials fed into the front stage is not appropriate, the blending amount of the materials fed into the rear stage can be adjusted to an appropriate total blending amount.

[0021] The state acquisition device 12 acquires kneading state information indicating the state of the mixture, such as the radioactive waste and the solidified material, being kneaded in the kneading container 14. The kneading state information includes at least one of the current value or torque value of the stirring motor 15 provided in the kneading device 11, the image of the inside of the kneading container 14, the temperature in the kneading container 14, and the kneading viscosity measured by a viscometer attached to the kneading container 14. In this manner, the viscosity of the radioactive waste and the solidified material being kneaded in the kneading container 14 can be acquired from the current value or torque value of the stirring motor 15, the image of the inside of the kneading container 14, or the kneading viscosity measured by a viscometer. In addition, the temperature of the radioactive waste and the solidified material being kneaded in the kneading container 14 can be acquired from the temperature in the kneading container 14.

[0022] The state acquisition device 12 has sensors such as an ammeter, a torque meter, a camera, a thermometer, and a viscometer. The kneading state information acquired by the state acquisition device 12 is sent to the control computer 13. When an image showing the inside of the kneading container 14 is acquired, the user may check the image once, and the confirmation result may be input to the control computer 13. The confirmation result includes the result of the user's judgment of whether the kneading state is good or bad when the user visually checks the image.

[0023] 11 is a graph showing the relationship between the current value of the stirring motor 15 of the kneading device 11 and the viscosity of the kneaded material. The current value and the torque value of the stirring motor 15 of the kneading device 11 are correlated with the viscosity of the kneaded material. Therefore, the viscosity of the kneaded material can be estimated from the current value of the stirring motor 15 of the kneading device 11.

[0024] Figure 8 is a graph showing the relationship between the particle shape of the solidification material and the viscosity of the mixture. For example, when the solid content in cement paste increases, the apparent viscosity also increases. Furthermore, based on the relationship between the particle shape and the viscosity of the mixture, it can be said that these physical properties are correlated with viscosity. The relationship between the particle shape and the viscosity of the mixture is expressed by the following relationship.

[0025]

number

[0026]

number

[0027] Here, η m and η c is the apparent viscosity of mortar and concrete. s m , φ g m is the actual volume ratio of fine aggregate and coarse aggregate. β s and β g is a shape coefficient that depends on the particle shape and particle size of the fine and coarse aggregates. The coefficients a and b are called solidification coefficients, and indicate the degree to which the paste and mortar become solid (suspended matter) that does not flow due to compaction or adhesion to the aggregate surface.

[0028] 9 is a graph showing the relationship between the specific surface area and the viscosity of the kneaded product. The viscosity of cement paste is correlated with the type of cement and the specific surface area.

[0029] Figure 10 is a graph showing the relationship between the ettringite content and the viscosity of the kneaded material. When the amount of sulfur trioxide in the solidification material and the concentration of sulfate ions in the radioactive waste increase, the calcium component (aluminate phase) in the solidification material reacts as shown in the following reaction formula to produce ettringite, a needle-shaped crystal mineral. Ettringite is formed faster than general cement hydrates. Ettringite exerts strength by itself and has the property of expanding the hardened cement body, so the amount produced affects the strength and soundness of the solidified body.

[0030] 3CaO・Al2O3+3(CaSO4・2H2O)+nH2O→3CaO・Al2O3・3CaSO4・32H2O

[0031] In conventional technology, the properties of radioactive waste cannot be controlled to a constant level, and the operating conditions and mixing amount cannot be significantly changed depending on the mixing state because of the influence on the subsequent processing steps. Therefore, the radioactive waste treatment system 1 of this embodiment aims to suppress the occurrence of poor mixing by adjusting the optimal mixing amount within the design specifications using machine learning depending on the properties of the radioactive waste, the properties of the solidification material, and the mixing state.

[0032] 1, the waste measuring tank 6 measures the amount of radioactive waste that is guided from the waste storage tank 2 to the kneading vessel 14. The waste measuring tank 6 also adjusts the flow rate of the radioactive waste that flows toward the kneading vessel 14.

[0033] The material measuring tank 7 measures the amount of the solidification material that is introduced from the material storage tank 3 to the kneading vessel 14. The material measuring tank 7 also adjusts the flow rate of the solidification material that flows toward the kneading vessel 14.

[0034] The water metering tank 8 measures the amount of water introduced from the water storage tank 4 to the kneading vessel 14. The water metering tank 8 also adjusts the flow rate of water flowing toward the kneading vessel 14.

[0035] The admixture metering tank 9 measures the amount of admixture introduced from the admixture storage tank 5 to the kneading vessel 14. The admixture metering tank 9 also adjusts the flow rate of the admixture flowing toward the kneading vessel 14.

[0036] The additives of this embodiment include at least one of an admixture, water, and an additional solidification material. The material metering tank 7, the water metering tank 8, and the admixture metering tank 9 serve as additive metering tanks of this embodiment. The additive metering tanks not only measure the fluid, but also adjust the flow rate. The flow rate adjustment includes switching between a flowing state and a non-flowing state. The additive metering tanks are controlled by a control computer 13. Various data acquired by the additive metering tanks are sent to the control computer 13.

[0037] In this embodiment, an in-drum mixing system is illustrated. Note that this embodiment is not limited to the in-drum mixing system, and an out-drum mixing system may also be applied.

[0038] The waste information acquisition device 10 acquires various data on the radioactive waste stored in the waste measuring tank 6 in order to acquire information on the properties of the radioactive waste. The various data acquired by the waste information acquisition device 10 is sent to a control computer 13.

[0039] The control computer 13 comprehensively controls the radioactive waste treatment system 1. The radioactive waste treatment method of this embodiment is realized by executing various programs on the control computer 13. The control computer 13 includes at least a processing circuit and a storage unit.

[0040] The processing circuit is, for example, a circuit equipped with a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), or a dedicated or general-purpose processor. This processor realizes various functions by executing various programs stored in a storage unit. The processing circuit may also be configured with hardware such as an FPGA (Field Programmable Gate Array) or an ASIC (Application Specific Integrated Circuit). Various functions can also be realized by such hardware. The processing circuit can also realize various functions by combining software processing by a processor and a program with hardware processing.

[0041] The storage unit stores various information required for performing radioactive waste treatment. For example, the storage unit stores a predetermined program executed by the processing circuit. The storage unit also stores a trained model.

[0042] The trained model includes an input layer, an intermediate layer, and an output layer. Input data is input to the input layer. The intermediate layer has its parameters trained in advance by machine learning using training data. The output layer outputs output data indicating the results of processing in the intermediate layer in response to input data input to the input layer.

[0043] The trained model is machine-trained using training data in which at least one of the input data or data simulating the input data is input, and at least one of the output data or data simulating the input data is output.

[0044] The machine learning of this embodiment includes at least one of a neural network, XGBoost (a method for enhancing decision trees and random forests), a k-nearest neighbor method, a support vector machine, a random forest, a decision tree, and a linear regression. In the following description, a neural network is exemplified.

[0045] The control computer 13 includes artificial intelligence (AI). The control computer 13 may also include a deep learning unit that extracts a specific pattern from a plurality of patterns based on deep learning.

[0046] A mathematical algorithm such as a deep learning algorithm or regression analysis can be used for the analysis using the control computer 13. In addition, the form of machine learning includes forms such as clustering and deep learning.

[0047] For example, a neural network is provided with an intermediate layer having multiple layers. Each intermediate layer is composed of multiple units. In addition, by having the multi-layered neural network learn in advance using learning data (teacher data), it is possible to automatically extract feature quantities in patterns of changes in the state of the system. Note that the number of intermediate layers, the number of units, the learning rate, the number of learning times, and the activation function of the multi-layered neural network can be set on the user interface.

[0048] In addition, deep reinforcement learning may be used in the neural network, in which a reward function is set for each information item to be learned, and the information item with the highest value is extracted based on the reward function.

[0049] The trained models of this embodiment include a waste prediction model M1 (FIG. 3), a material prediction model M2 (FIG. 4), a blend prediction model M3 (FIG. 5), and an addition prediction model M4 (FIG. 7). Although these trained models are generated individually, they may be generated together as a single trained model.

[0050] In this embodiment, the control computer 13 automatically controls various devices, but other configurations are also possible. For example, the control computer 13 may control various devices by receiving input operations from a user. In other words, the control computer 13 may be a remote control unit for controlling various devices through manual operations by a user.

[0051] Next, the radioactive waste treatment executed by the control computer 13 will be explained using the flowchart of Fig. 2. This flowchart simplifies the flow of the main process to facilitate understanding. The above-mentioned drawings will be used as references as appropriate.

[0052] Note that the arrows in FIG. 2 are one example showing the flow of processes, and there may be other process flows than those shown by the arrows. Furthermore, the order of each process is not necessarily fixed, and the order of some processes may be interchanged. Furthermore, some processes may be executed in parallel with other processes. Furthermore, the following processes are at least some of the processes included in the radioactive waste treatment, and other processes may be included in the radioactive waste treatment.

[0053] First, in step S1, the control computer 13 acquires information on the properties of the radioactive waste that is the target of solidification treatment.

[0054] In the next step S2, the control computer 13 acquires property information of the solidification material used to solidify the radioactive waste.

[0055] In the next step S3, the control computer 13 determines the mixing amounts of the radioactive waste and the solidification material to be charged into the kneading vessel 14 in which the radioactive waste and the solidification material are mixed.

[0056] In the next step S4, the kneading device 11 kneads the materials fed into the kneading vessel 14. Here, the kneading device 11 kneads the radioactive waste, solidification material, water, and admixtures introduced from the waste measuring tank 6, material measuring tank 7, water measuring tank 8, and admixture measuring tank 9 in the kneading vessel 14.

[0057] In the next step S5, the state acquisition device 12 acquires information on the kneading state of the material being kneaded in the kneading container .

[0058] In the next step S6, the control computer 13 judges whether the kneading is normal or not. If the kneading is normal (YES in step S6), the current state is maintained and continues until the predetermined kneading time has elapsed. On the other hand, if the kneading is not normal (NO in step S6), the process proceeds to step S7.

[0059] In the next step S7, the control computer 13 determines the amount of at least one of the admixture, water, and additional solidification material to be added as an additive to the kneading vessel 14.

[0060] In the next step S8, the control computer 13 adds the additives to the kneading vessel 14, and the process returns to step S4.

[0061] FIG. 3 shows an example of the process in which the control computer 13 acquires information on the properties of the radioactive waste in step S1 (FIG. 2) described above.

[0062] The control computer 13 inputs radioactive waste usage history information as input data D1 to the machine-learned waste prediction model M1. The control computer 13 then predicts radioactive waste property information based on the output data D2 output from this waste prediction model M1. In this way, it is possible to predict radioactive waste property information from radioactive waste usage history information.

[0063] The radioactive waste usage history information as input data D1 includes at least one of the solid content concentration of the radioactive waste, the particle shape of the radioactive waste, the solution pH of the radioactive waste, and the sulfate ion concentration of the radioactive waste. Note that data other than these may also be included in the input data D1. The input data D1 is acquired by the waste information acquisition device 10 and sent to the control computer 13. The input data D1 may also be input to the control computer 13 by a user.

[0064] The property information of the radioactive waste as the output data D2 includes at least one of the kneading viscosity in the basic mix and the additional kneading viscosity from the amount of ettringite produced. Note that data other than these may be included in the output data D2.

[0065] FIG. 4 shows an example of the process in which the control computer 13 acquires property information of the solidified material in step S2 (FIG. 2) described above.

[0066] The control computer 13 inputs the quality information of the solidified material as input data D3 to the machine-learned material prediction model M2. The control computer 13 then predicts the property information of the solidified material based on the output data D4 output from the material prediction model M2. In this way, the property information of the solidified material can be predicted from the quality information of the solidified material.

[0067] The quality information of the solidified material as the input data D3 includes at least one of the cement type of the solidified material, the specific surface area of ​​the solidified material, and the amount of sulfur trioxide of the solidified material. Note that data other than these may be included in the input data D3. The input data D3 is input to the control computer 13 by a user.

[0068] The property information of the solidification material as the output data D4 includes at least one of the kneading viscosity in the basic mixture and the additional kneading viscosity from the amount of ettringite produced. Note that data other than these may be included in the output data D4.

[0069] FIG. 5 shows an example of the process in which the control computer 13 determines the mixing ratio of the radioactive waste and the solidification material in step S3 (FIG. 2) described above.

[0070] The control computer 13 inputs the property information of the radioactive waste, the property information of the solidification material, and the design specification information as input data D5 to the machine-learned mixture prediction model M3. Then, the control computer 13 determines the mixture amount of the radioactive waste and the solidification material based on the output data D6 output from this mixture prediction model M3.

[0071] The input data D5 includes at least the property information of the radioactive waste and the property information of the solidification material. That is, the input data D5 includes the mixing viscosity in the basic mix and the additional mixing viscosity from the amount of ettringite produced. The input data D5 also includes design specification information. This design specification information is a design specification that has been approved in advance by a government agency. Therefore, it is assumed that the range of the upper and lower limits of the mix adjustment falls within the range of the measurement tolerance. Furthermore, data other than these may be included in the input data D5.

[0072] The output data D6 includes at least the blending amount. The output data D6 may also include a blending viscosity estimated when the blending amount is output. Furthermore, data other than these may also be included in the output data D6.

[0073] In step S11, which is performed after the output data D6 is output, the control computer 13 judges whether the blend amount based on the output data D6 output from the blend prediction model M3 is within the range of the predetermined design specifications. If the blend amount is not within the range of the design specifications (NO in step S11), the process proceeds to step S12. On the other hand, if the blend amount is within the range of the design specifications (YES in step S11), the process proceeds to step S13.

[0074] In step S12, the control computer 13 outputs an error notification, thereby making it possible to inform the user that there is a risk of poor kneading.

[0075] In step S13, the control computer 13 judges whether the strength of the solidified body estimated when the radioactive waste and the solidification material are solidified in the amount based on the output data D6 output from the mixture prediction model M3 is within the range of the design specifications. If the strength of the solidified body is not within the range of the design specifications (NO in step S13), the input data D5 is input again to the mixture prediction model M3, the output data D6 is output, and the process is repeated. On the other hand, if the strength of the solidified body is within the range of the design specifications (YES in step S13), the process proceeds to step S14.

[0076] In step S14, the control computer 13 determines the mixing ratio based on the output data D6. The control computer 13 controls the waste measuring tank 6 and the material measuring tank 7 based on the determined mixing ratio. In this way, the mixing ratio of the radioactive waste and the solidification material can be adjusted.

[0077] The control computer 13 measures the amount of radioactive waste and solidification material led to the kneading vessel 14 based on the measurements made by the waste metering tank 6 and the material metering tank 7. Furthermore, the control computer 13 adjusts the amount of radioactive waste and solidification material led to the kneading vessel 14 based on the kneading state information acquired by the state acquisition device 12. In this way, when the blending amount based on the blending prediction model M3 is not appropriate, the blending amount can be appropriately adjusted.

[0078] The control computer 13 may also train the blending prediction model M3 using the adjusted blending amount as learning data. In this way, the blending amount of the next batch is presented in advance as an optimal blending amount based on the contents of the previous process by machine learning, so that the processing time can be shortened. In addition, the frequency of occurrence of poor kneading can be reduced by accumulating learning data.

[0079] FIG. 6 shows a specific example of the process executed by the control computer 13 while mixing the radioactive waste and the solidification material in the above-mentioned steps S4 to S8 (FIG. 2).

[0080] First, in step S21, the control computer 13 acquires kneading state information from the state acquisition device 12. This kneading state information includes at least one of the viscosity and temperature of the radioactive waste and solidification material being kneaded in the kneading vessel 14.

[0081] In the next step S22, the control computer 13 judges whether the kneading state is within the range of the design specifications. For example, the control computer 13 judges whether the viscosity and temperature of the radioactive waste and solidification material being kneaded in the kneading container 14 are within the range of the predetermined design specifications. Here, if the kneading state is within the range of the design specifications (YES in step S22), the current state is maintained and continues until the predetermined kneading time has elapsed. On the other hand, if the kneading state is not within the range of the design specifications (NO in step S22), the process proceeds to step S23.

[0082] In step S23, the control computer 13 controls the additive metering tanks consisting of the material metering tank 7, the water metering tank 8, and the admixture metering tank 9, and adds at least one of the additives, the admixture, the water, and the additional solidification material, to the kneading vessel 14.

[0083] In the next step S24, the control computer 13 acquires additional kneading state information from the state acquisition device 12.

[0084] In the next step S25, the control computer 13 judges whether the kneading state is within the range of the design specifications. If the kneading state is within the range of the design specifications (YES in step S25), the current state is maintained and continues until the predetermined kneading time has elapsed. On the other hand, if the kneading state is not within the range of the design specifications (NO in step S25), the process proceeds to step S26.

[0085] In step S26, the control computer 13 outputs an error notification, thereby making it possible to inform the user that there is a risk of poor kneading.

[0086] FIG. 7 shows an example of the process in which the control computer 13 determines the amount of additive to be added in step S7 (FIG. 2) described above.

[0087] The control computer 13 inputs the property information of the radioactive waste, the property information of the solidification material, the design specification information, the compounding amount, and the kneading state information as input data D7 to the machine-learned additive prediction model M4. Then, the control computer 13 determines the additive amount based on the output data D8 output from this additive prediction model M4.

[0088] The input data D7 includes at least the property information of the radioactive waste and the property information of the solidification material. That is, the input data D7 includes the kneading viscosity in the basic mix and the additional kneading viscosity from the amount of ettringite produced. The input data D7 also includes design specification information. The input data D7 also includes at least one of the viscosity and temperature of the radioactive waste and the solidification material being kneaded in the kneading vessel 14 as kneading state information. The input data D7 may also include an admixture amount indicating the amount of the admixture. Furthermore, data other than these may also be included in the input data D7.

[0089] The output data D8 includes at least the additive amount. The output data D8 may also include a kneading viscosity estimated when the additive is added in the output additive amount and kneaded. Furthermore, data other than these may also be included in the output data D8.

[0090] In step S31, which is performed after the output data D8 is output, the control computer 13 judges whether the blending amount and the addition amount based on the output data D8 output from the addition prediction model M4 are within the range of the predetermined design specifications. If the blending amount and the addition amount are not within the range of the design specifications (NO in step S31), the process proceeds to step S32. On the other hand, if the blending amount and the addition amount are within the range of the design specifications (YES in step S31), the process proceeds to step S33.

[0091] In step S32, the control computer 13 outputs an error notification. That is, the control computer 13 judges whether the additive amount based on the output data output from the additive prediction model M4 is within the range of the predetermined design specifications, and outputs an error notification if the additive amount is not within the range of the design specifications. In this way, it is possible to inform the user that there is a risk of poor kneading.

[0092] In step S33, the control computer 13 judges whether the strength of the solidified body estimated when the radioactive waste and the solidification material are solidified with the blending amount and the addition amount based on the output data D8 output from the addition prediction model M4 is within the range of the design specifications. If the strength of the solidified body is not within the range of the design specifications (NO in step S33), the input data D7 is input again to the addition prediction model M4, the output data D8 is output, and the process is repeated. On the other hand, if the strength of the solidified body is within the range of the design specifications (YES in step S33), the process proceeds to step S34.

[0093] In step S34, the control computer 13 determines the amount of additive based on the output data D8. Based on the determined amount of additive, the control computer 13 controls the material metering tank 7, the water metering tank 8, and the admixture metering tank 9. In this way, the kneading state can be adjusted by the additive.

[0094] In addition, the control computer 13 measures the additives to be introduced into the kneading vessel 14 based on measurements in the material metering tank 7, the water metering tank 8, and the admixture metering tank 9. Furthermore, the control computer 13 adjusts the amount of additives to be introduced into the kneading vessel 14 based on the kneading state information acquired by the state acquisition device 12. In this way, when the addition amount output from the addition prediction model M4 is not appropriate, the addition amount can be appropriately adjusted.

[0095] The control computer 13 may also train the addition prediction model M4 using the adjusted addition amount as learning data. In this way, the next amount of additive to be added is presented in advance as an optimal blend based on the contents of the previous processing by machine learning, thereby shortening the processing time.

[0096] The radioactive waste treatment system 1 described above includes a control device, a storage device, an output device, an input device, and a communication interface. Here, the control device includes a highly integrated processor such as a central processing unit (CPU), a graphics processing unit (GPU), a field programmable gate array (FPGA), or a dedicated chip. The storage device includes a read only memory (ROM), a random access memory (RAM), a hard disk drive (HDD), a solid state drive (SSD), etc. The output device includes a display panel, a head mounted display, a projector, a printer, etc. The input device includes a mouse, a keyboard, a touch panel, etc. This radioactive waste treatment system 1 can be realized by a hardware configuration using a normal computer.

[0097] The program or trained model executed in the radioactive waste treatment system 1 is provided by being pre-installed in a ROM or the like. Additionally or alternatively, the program or trained model is provided by being stored in a computer-readable non-transitory storage medium as a file in an installable or executable format. The storage medium includes a CD-ROM, a CD-R, a memory card, a DVD, a flexible disk (FD), and the like.

[0098] The program or trained model executed in the radioactive waste treatment system 1 may be stored in a computer connected to a network such as the Internet and downloaded via the network to be provided. That is, the program or trained model may be provided from a cloud computing resource. A server on the cloud may execute the program or trained model, and only the processing results may be provided via the cloud. The radioactive waste treatment system 1 may also be configured by combining separate modules that independently perform the functions of the components, which are mutually connected via a network or a dedicated line.

[0099] Each component of the radioactive waste treatment system 1 does not necessarily have to be provided in one control computer 13. For example, one radioactive waste treatment system 1 may be realized by a plurality of computers connected to each other via a network. For example, the waste prediction model M1, the material prediction model M2, the blend prediction model M3, and the addition prediction model M4 may each be installed in an individual computer.

[0100] In the above example, the control computer 13 constituting the radioactive waste treatment system 1 performs various processes (including various processes such as judgment and setting), but other modes are also possible. For example, a user may perform some of the various processes described above, and the computer may receive input of the processing results and use them in its own processing.

[0101] According to the embodiment described above, by determining the blending amount based on the output data D6 output from the blending prediction model M3, when solidifying radioactive waste, the blending amount can be adjusted to an optimum amount according to the properties of the radioactive waste, thereby suppressing the occurrence of poor mixing.

[0102] Although some embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These embodiments can be implemented in various other forms, and various omissions, substitutions, modifications, and combinations can be made without departing from the spirit of the invention. These embodiments or modifications thereof are within the scope of the invention and its equivalents as described in the claims, as well as within the scope and spirit of the invention. [Explanation of symbols]

[0103] 1...radioactive waste treatment system, 2...waste storage tank, 3...material storage tank, 4...water storage tank, 5...admixture storage tank, 6...waste metering tank, 7...material metering tank, 8...water metering tank, 9...admixture metering tank, 10...waste information acquisition device, 11...kneading device, 12...status acquisition device, 13...control computer, 14...kneading vessel, 15...mixing motor, 16...mixing shaft, 17...mixing blade, D1, D3, D5, D7...input data, D2, D4, D6, D8...output data, M1...waste prediction model, M2...material prediction model, M3...mixing prediction model, M4...addition prediction model.

Claims

1. Obtain information on the properties of radioactive waste that is to be solidified, Acquire information on the properties of a solidification material for solidifying the radioactive waste; The property information of the radioactive waste and the property information of the solidification material are input as input data into a machine learning-based mixture prediction model; determining a blending amount of the radioactive waste and the solidification material to be put into a kneading vessel in which the radioactive waste and the solidification material are mixed based on output data output from the blending prediction model; one or more computers configured to Radioactive waste treatment system.

2. The computer includes: The radioactive waste usage history information is input as input data into a machine learning-based waste prediction model; predicting property information of the radioactive waste based on output data output from the waste prediction model; It is configured as follows: The radioactive waste treatment system according to claim 1.

3. The usage history information includes at least one of a solid content concentration, a particle shape, a solution pH, and a sulfate ion concentration. The radioactive waste treatment system according to claim 2.

4. The computer includes: The quality information of the solidified material is input as input data into a machine learning-processed material prediction model; predicting property information of the solidified material based on output data output from the material prediction model; It is configured as follows: The radioactive waste treatment system according to any one of claims 1 to 3.

5. The quality information includes at least one of cement type, specific surface area, and sulfur trioxide amount; The radioactive waste treatment system according to claim 4.

6. The computer includes: Determining whether the blend amount based on the output data output from the blend prediction model is within a predetermined design specification range; outputting an error notification when the blending amount is not within the range of the design specification; It is configured as follows: The radioactive waste treatment system according to any one of claims 1 to 3.

7. The computer includes: Acquire mixing state information indicating a state of the radioactive waste and the solidification material being mixed in the mixing vessel; Weighing the radioactive waste and the solidification material introduced into the kneading vessel; adjusting the amounts of the radioactive waste and the solidification material to be introduced into the kneading vessel based on the kneading state information; It is configured as follows: The radioactive waste treatment system according to any one of claims 1 to 3.

8. The kneading state information includes at least one of a current value or a torque value of a stirring motor provided in the kneading vessel, an image of the inside of the kneading vessel, a temperature in the kneading vessel, and a viscosity measured by a viscometer attached to the kneading vessel. The radioactive waste treatment system according to claim 7.

9. The kneading state information includes at least one of a viscosity and a temperature of the radioactive waste and the solidification material being kneaded in the kneading vessel, The computer includes: determining whether the viscosity and the temperature are within predetermined design specifications; outputting an error notification if at least one of the viscosity or the temperature is not within the design specification range; It is configured as follows: The radioactive waste treatment system according to claim 7.

10. The computer includes: The property information of the solidification material, the property information of the radioactive waste, the blending amount, and the kneading state information are input as input data into a machine learning-processed addition prediction model; Based on the output data output from the addition prediction model, an addition amount of at least one of an admixture, water, and the additional solidification material is determined when the admixture is added as an additive to the kneading vessel. It is configured as follows: The radioactive waste treatment system according to claim 7.

11. The computer includes: Metering the additives introduced into the kneading vessel; adjusting the amount of the additive to be introduced into the kneading vessel based on the kneading state information; It is configured as follows: The radioactive waste treatment system according to claim 10.

12. The computer includes: determining whether the addition amount of the additive based on the output data output from the addition prediction model is within a predetermined design specification range; outputting an error notification if the amount of addition is not within the range of the design specification; It is configured as follows: The radioactive waste treatment system according to claim 10.

13. A waste information acquisition device for acquiring information on the properties of the radioactive waste; A waste measuring tank for measuring the amount of the radioactive waste introduced into the kneading vessel; A material measuring tank for measuring the solidification material introduced into the kneading vessel; An additive measuring tank for measuring the additives to be introduced into the kneading vessel; a kneading device for kneading the radioactive waste, the solidification material, and the additives introduced from the waste measuring tank, the material measuring tank, and the additive measuring tank in the kneading vessel; A state acquisition device for acquiring the kneading state information; Equipped with The radioactive waste treatment system according to claim 10.

14. Obtain information on the properties of radioactive waste that is to be solidified, Acquire property information of a solidification material for solidifying the radioactive waste; The property information of the radioactive waste and the property information of the solidification material are input as input data into a machine learning-based mixture prediction model; determining a blending amount of the radioactive waste and the solidification material to be put into a kneading vessel in which the radioactive waste and the solidification material are mixed based on output data output from the blending prediction model; The processing is performed by one or more computers; Radioactive waste disposal methods.

Images