Method for determining nuclear core loading pattern

Abstract

The present invention relates to a method for determining a core loading pattern defining an arrangement of fuel assemblies (40) in a nuclear core, the method comprising the following steps: defining at least one potential core loading pattern; calculating a predicted bending of the fuel assembly (40) at the end of an operating cycle for each potential core loading pattern, the calculation being performed by an automatic learning algorithm trained on a training data set comprising a plurality of other loading patterns and measured values ​​of the bending of the fuel assembly (40) at the end of the cycle for each of the other loading patterns; evaluating at least one core loading pattern based on the predicted bending calculation and at least one predetermined criterion; and selecting a potential core loading pattern.

Description

Technical Field

[0001] The invention relates to a method for determining a nuclear core loading mode. Background Art

[0002] A pressurized water reactor includes a vessel and a core located within the vessel. The core is composed of a plurality of nuclear fuel assemblies, each extending axially, preferably vertically, and arranged side by side. When the fuel assemblies are loaded into the vessel to form the core, they are arranged according to a core loading pattern that defines the position of each fuel assembly in the core for the next operating cycle of the nuclear power plant.

[0003] Each fuel assembly comprises a bundle of nuclear fuel rods supported by a support skeleton including a spacer grid designed to maintain the fuel rods in a spaced relationship transverse to the axial direction of the fuel assembly. Each rod comprises a cladding containing nuclear fuel pellets.

[0004] During operation, pressurized water flows within the vessel and through the core, for example from the bottom to the top of the core, and the coolant fluid recovers heat from the fuel assemblies. The coolant ensures cooling of the fuel assemblies and moderation of the nuclear reactions in the core.

[0005] During reactor operation, the fuel assemblies in the core may undergo mechanical deformation.

[0006] These deformations vary from one fuel assembly to another in the core, for example according to the position of the fuel assembly in the core and / or according to whether the fuel assembly is fresh or has been irradiated in at least one previous operating cycle, or contains fresh fuel rods and / or fuel rods that have been used in at least one previous operating cycle [to be confirmed].

[0007] These deformations of the fuel assemblies may undermine the operation and performance of the reactor: risk of incomplete insertion of the control cluster, which makes it possible to adjust the reactivity of the nuclear reactor core; or risk of an unacceptable increase in the descent time of the control cluster; or risk of local variations in the smoothness of the core, etc.

[0008] During maintenance, such as during unloading and reloading operations of the core, these deformations increase the risk of jamming between fuel assemblies, thereby increasing the risk of damaging the fuel assemblies.

[0009] Therefore, it is important to take these possible deformations into account when defining the core loading pattern.

[0010] To this end, it is known to use finite element mechanical models and coupled fluid-structure computer fluid dynamics (CFD) models to calculate the flow of the coolant fluid within the vessel and the mechanical deformations of the components.

[0011] However, these calculations are very time-consuming and require significant computing resources. For example, currently a CFD calculation for just one core loading pattern to be tested can take a week.

[0012] Therefore, it is known to define the core loading pattern based on some constraints.When designing the core loading pattern, the problem of potential fuel assembly deformation is only checked after the fact. Summary of the Invention

[0013] One object of the present invention is therefore to propose a method for determining the core loading pattern that better takes into account the constraints caused by the deformation of the fuel assemblies, thereby improving the safety of the reactor during operation or maintenance of the nuclear reactor. In particular, this will allow relying on a complete calculation process, rather than relying on a hybrid process based on calculations, but also relying on engineering experience input.

[0014] To this end, the present invention provides a method for determining a core loading pattern, which defines the arrangement of fuel assemblies in a nuclear core for an operating cycle of a nuclear power plant, the method comprising at least the following steps:

[0015] - define at least one potential core loading pattern,

[0016] - calculating the predicted bending of the fuel assembly at the end of the operating cycle for each potential core loading pattern, the calculation being performed by an automated learning algorithm trained on a training data set containing a plurality of other loading patterns and measured values ​​of the bending of the fuel assembly at the end of the cycle for each of these,

[0017] - evaluating at least one core loading mode based on the predicted bending calculation and at least one predetermined criterion, and

[0018] -Select a potential core loading mode.

[0019] According to further advantageous aspects of the invention, the method comprises one or more of the following features, considered individually or in all technically possible combinations:

[0020] - the method comprises: after the evaluating step, if the evaluation fails, repeating the previous steps using at least one different potential core loading pattern;

[0021] - a plurality of potential core loading patterns are defined, the step of selecting a potential core loading pattern being based at least on a predicted bending calculation and at least one predetermined selection criterion;

[0022] - Defining at least one potential core loading mode is performed based on characteristics of the operating cycle and characteristics of the fuel assemblies and core design safety constraints.

[0023] - at least one predetermined criterion is selected from the group consisting of: minimum and / or maximum gap between two fuel assemblies, maximum bending amplitude of each fuel assembly, and average bending degree of each fuel assembly;

[0024] - the plurality of potential core loading modes comprises at least four core loading modes, in particular more than ten core loading modes;

[0025] - the training dataset includes data derived from fuel assemblies burned during operation of a nuclear power plant;

[0026] - the training data set includes data derived from bending calculations performed on the fuel assembly (e.g., finite element mechanical calculations and coupled fluid-structure computer fluid dynamics calculations); and

[0027] -Automatic learning algorithm based on neural networks.

[0028] The invention also provides a computer program product comprising software instructions which, when executed by a computer, implement the method as defined above.

[0029] The present invention also provides a method for loading a fuel assembly into a nuclear reactor core, comprising at least the following steps:

[0030] - selecting the core loading mode by the determination method defined above, and

[0031] - Loading the fuel assemblies into the nuclear core according to the selected core loading mode. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The characteristics and advantages of the present invention will become apparent on reading the following description, which is given purely as a non-limiting example and is made with reference to the accompanying drawings, in which:

[0033] - Figure 1 It is a schematic diagram of a pressurized water nuclear reactor, which includes a container and a core located within the container. The core includes nuclear fuel assemblies.

[0034] - Figure 2 yes Figure 1 A horizontal cross-sectional view of the reactor at the core.

[0035] - Figure 3 yes Figure 2 A schematic diagram of the two components of the core, and

[0036] - Figure 4 is a flow chart of a determination method according to the present invention. DETAILED DESCRIPTION

[0037] exist Figure 1 In the embodiment, the nuclear reactor 10 includes a container 11 and a core 12 located in the container 11 .

[0038] The nuclear reactor 10 also includes one or more steam generators 14, one or more turbines 16 (each coupled to a generator 18), and one or more condensers 20, each of which has only one Figure 1 Shown in.

[0039] The nuclear reactor 10 also includes a primary circuit 22 equipped with a pump 24 and a fluid such as Figure 1 The arrows in FIG. 2 show the flow in the primary circuit 22 .

[0040] This fluid flows in particular upwards inside the vessel 11 and through the core 12 to be heated therein, while ensuring cooling of the core 12 and moderation of the nuclear reaction in the core 12 .

[0041] The primary circuit 22 also includes a pressurizer 26 making it possible to adjust the pressure of the fluid flowing in the primary circuit 22 .

[0042] The nuclear reactor 10 is, for example, a pressurized water reactor (PWR), and the fluid flowing in the primary loop 22 is pressurized water.

[0043] Alternatively, the nuclear reactor 10 is a boiling water reactor (BWR), and the fluid flowing in the primary circuit is pressurized water, in the form of steam in the upper portion of the core, at a pressure generally lower than the pressure of the water flowing in the primary circuit of a pressurized water reactor. Alternatively, the nuclear reactor 10 is a reactor cooled with sodium, molten salt, or gas.

[0044] The nuclear reactor 10 includes a secondary circuit 34 connected to the steam generator 14. The fluid of the primary circuit 22 is supplied to the steam generator 14, where it is cooled by evaporating water in the secondary circuit 34. The steam generated by the steam generator 14 in the secondary circuit 34 is directed by the secondary circuit 34 to the turbine 16 and then to the condenser 20, where the steam is condensed by indirect heat exchange with cooling water flowing in the condenser 20. The secondary circuit 34 includes a pump 35 and a heater 36 downstream of the condenser 20.

[0045] The nuclear reactor 10 includes Figure 2 Shown are a heat shield 37 , a core housing 38 , and a reflector 28 located within the vessel 11 .

[0046] The container 11 comprises an inlet orifice 39A for the fluid and an outlet orifice 39B for the fluid, which are connected to the primary circuit 22 .

[0047] The core 12 is formed by a plurality of nuclear fuel assemblies 40 loaded in a container 11. Each fuel assembly 40 is elongated along the axial direction. The fuel assemblies 40 are arranged side by side in the core with their axial directions being parallel therebetween and generally parallel to the vertical direction Z.

[0048] The core 12 typically includes more than one hundred fuel assemblies 40. Figure 2 In the example of a 900 MWe reactor shown, the core 12 includes one hundred fifty-seven (157) fuel assemblies 40 .

[0049] Figure 2 A top view of an example of the distribution of these various assemblies 40 within the core 12 is shown. Figure 2 Each square in FIG. 4 represents a corresponding fuel assembly 40 .

[0050] This distribution of fuel assemblies 40 defines a core loading pattern. Each fuel assembly 40 of the core 12 is individually identified and the core loading pattern defines the corresponding position of each fuel assembly 40 within the core 12. The core loading pattern specifically defines the appropriate configuration of fresh fuel assemblies and already irradiated fuel assemblies.

[0051] The reactor 10 further includes a lower plate 41A and an upper plate 41B. When the reactor 10 is in operation, the lower plate 41A and the upper plate 41B are located on both sides of the assembly 40 in the axial direction. The assembly 40 is arranged on the lower plate 41A, and the upper plate 41B is located above the assembly 40 and contacts their upper ends. Figure 1 shown.

[0052] Reactor 10 includes Figure 1 Control rod clusters 42 are shown positioned above certain components 40 within container 11 . Figure 1 A single rod cluster 42 is shown in FIG. Rod clusters 42 can be moved by mechanism 44 to be inserted into and removed from the assembly 40 from which they are suspended. Conventionally, each control rod cluster 42 includes absorber rods comprising one or more neutron-absorbing materials and, optionally, inert rods (i.e., rods having no specific neutron-absorbing properties). Vertical movement of rod clusters 42 allows for adjustment of reactivity in core 12 and allows the total power P provided by core 12 to be varied from zero power to a nominal power PN based on the insertion of control rod clusters 42 into fuel assemblies 40.

[0053] like Figure 3 As shown, each fuel assembly 40 includes a bundle of nuclear fuel rods 46 and a support skeleton 48 for supporting the fuel rods 46. The skeleton 48 includes, for example, a lower end piece 50, an upper end piece 52, a guide tube 54 connecting the two end pieces 50 and 52, and a spacer grid 56 distributed along the guide tube 54 for holding the rods 46. Each guide tube 54 is used to receive a corresponding rod of the control rod cluster 42 when the control rod cluster 42 is inserted into the core 12.

[0054] Figure 3 Two adjacent components 40 are shown, namely Figure 2 In the illustration of , two assemblies 40 correspond to adjacent squares in a horizontal plane perpendicular to the axial direction Z. The two adjacent assemblies 40 are positioned continuously and are spaced apart from each other in the transverse direction by a first gap BP1 between the corresponding grids 56 of the two assemblies 40 and a second gap BP2 between the corresponding bars 46 of the two assemblies 40, as shown in FIG. Figure 3 and as shown in Figure 5.

[0055] exist Figure 2 , the assemblies 40 are substantially aligned in two corresponding directions X and Y in a horizontal plane perpendicular to the vertical direction Z, and those skilled in the art will understand that when two consecutive assemblies 40 are aligned in direction X, then the lateral direction associated with the first gap BP1 and the second gap BP2 corresponds to the direction X. Similarly, when two consecutive assemblies 40 are aligned in direction Y, then the lateral direction associated with the first gap BP1 and the second gap BP2 corresponds to the direction Y.

[0056] For a given position in the axial direction Z, the size of the first gap BP1 is equal to the distance in the transverse direction between the outer surfaces of the perspective grids 56 of the two assemblies 40. By convention, for a given position in the axial direction Z, the size of the second gap BP2 is equal to the distance between the centers of two corresponding peripheral rods 46 of the two fuel assemblies 40.

[0057] Now we will use Figure 4 The method for determining the nuclear core loading mode of the nuclear reactor 10 according to the present invention is described with reference to a flowchart of FIG.

[0058] In a first step 101, at least one potential core loading pattern is defined.

[0059] In particular, the potential core loading mode or each potential core loading mode is defined based on characteristics of the next operating cycle and characteristics of the fuel assemblies 40 and design safety limitations of the core 12 .

[0060] Characteristics of the next operating cycle are, for example, the cycle length, the expected power output of the reactor 10 during the cycle, the temperature program in the core 12 during the operating cycle.

[0061] The characteristic of the fuel assembly 40 is, for example, the fuel composition of the fuel assembly 40 during the previous irradiation period(s) or the strength of the fuel assembly 40 during the previous irradiation period(s).

[0062] The core 12 design safety limits are, for example, limits on local core power, primary fluid temperature, or reactivity of the fuel assemblies 40 .

[0063] Advantageously, a plurality of core loading modes are defined. The plurality of core loading modes preferably comprises at least four core loading modes, in particular more than ten core loading modes.

[0064] Then, in a second step 102 , the calculation of the bending of each fuel assembly 40 at the end of the operating cycle is performed for each core loading pattern defined in the core loading pattern definition step.

[0065] In particular, the bending is calculated along the transverse directions X and Y.

[0066] This calculation is performed by an automated learning algorithm trained on an existing training data set that includes a variety of loading patterns and, for each of them, bending measurements of the fuel assembly 40 at the end of the cycle.

[0067] In particular, automatic learning algorithms are based on neural networks comprising neurons or neural nodes which are usually organized into multiple layers.

[0068] For example, automatic learning methods are based on models that use statistical methods to improve the performance of the method in solving tasks without having to be explicitly programmed for each of these tasks.

[0069] Automatic learning has two phases. The first phase involves defining the model from the data present in the dataset, which is called training. This so-called training phase is usually carried out before the actual use of the model.

[0070] The second phase corresponds to the use of the model: each potential core loading pattern defined in the core loading pattern definition step is submitted to the model as a new input, and the model calculates at least the predicted bending of the fuel assembly 40 associated with the corresponding core loading pattern.

[0071] The training data set includes data derived from fuel assemblies 40 burned during operation of nuclear reactor 10 or another nuclear reactor, preferably another reactor having the same or similar characteristics as reactor 12, preferably another reactor of the same concept.

[0072] As a variant or in addition, data are derived from the fuel assembly 40 during laboratory testing.

[0073] Alternatively or additionally, the data is derived from bending calculations performed on the fuel assembly 40, such as finite element mechanical calculations and coupled fluid-structure computer fluid dynamics (CFD) calculations.

[0074] Then, in step 103, the method includes the step of evaluating the predicted bending of the fuel assembly 40 according to at least one predetermined criterion.

[0075] Criteria may be defined based on optimal thermal hydraulics of cooling fluid into the core during operation and / or based on maintenance capabilities during outages and / or based on nuclear safety requirements and / or mechanical integrity of components, etc.

[0076] Assessments are based on a single criterion or a combination of criteria.

[0077] In particular, the at least one predetermined criterion is selected from the following group:

[0078] - the maximum gap BP1, BP2 between two fuel assemblies 40, for example to take into account the optimal thermohydraulic forces (water gap) for the primary coolant to enter the core during operation

[0079] - A minimum gap BP1 , BP2 between two fuel assemblies 40 , for example to allow for maintenance capabilities during power outages (grid interlocking).

[0080] - the maximum bending amplitude of each fuel assembly 40, and

[0081] - The average degree of curvature of each fuel assembly 40.

[0082] The evaluation may be based on a comparison of the predicted bending characteristics to an acceptable threshold value of at least one predetermined criterion.

[0083] Then in the next step 104, the potential core loading pattern corresponding to the predicted bending with the best evaluation is selected as the core loading pattern to be implemented in the next operating cycle.

[0084] In one embodiment, the evaluation may be based on additional calculations, such as control rod cluster descent time tests, using as input the predicted bending of one or all fuel assemblies 40 of the nuclear reactor 10 .

[0085] As a variation in the global step 105, the standards, predicted bending of the fuel assemblies 40 and their associated core loading patterns are part of the input to an optimizer model for evaluation / selection, which has its own calculation algorithm to propose the optimal core loading pattern as output.

[0086] The selection of the optimal loading pattern is based on the predicted bending calculation and at least one predetermined criterion. Advantageously, the selection is also based on other technical criteria, such as the characteristics of the next operating cycle, the characteristics of the fuel assembly 40 or the design safety limits of the core 12, and other economic criteria, such as the manufacturing cost of each potential loading pattern.

[0087] In the event that no core loading pattern meets the predetermined criteria, the method comprises performing a new iteration of steps 101 to 103 using at least one different potential core loading pattern.

[0088] Finally, after a core loading mode is selected, the fuel assemblies 40 are loaded into the nuclear core 12 according to the selected core loading mode.

[0089] The nuclear reactor 10 may then generate electricity during the next operating cycle.

[0090] It can be seen that the determination method according to the present invention can better consider the constraints caused by the deformation of the fuel assembly 40.

[0091] In particular, the present invention enables selection of a core loading pattern that minimizes deformation of the fuel assemblies 40 , thereby avoiding grid damage or incomplete rod insertion events and disturbances in power and flow distribution in the core 12 .

[0092] Therefore, the present invention can improve the safety of the reactor 10 during operation.

[0093] Furthermore, the determination method is performed rapidly and does not require extensive computational effort.

Claims

1. A method for determining a nuclear core (12) loading pattern, wherein the nuclear core (12) loading pattern defines the arrangement of fuel assemblies (40) in the nuclear core (12) for an operating cycle of a nuclear power plant (10), the method comprising at least the following steps: - define multiple potential core loading modes, - calculating the predicted bending of the fuel assembly (40) at the end of the operating cycle for each potential core loading pattern, said calculation being performed by an automated learning algorithm trained on a training data set comprising a plurality of other loading patterns and measured values ​​of the bending of the fuel assembly (40) at the end of the cycle for each of these, - evaluating said potential core loading pattern based on a predicted bending calculation and at least one predetermined criterion, and - selecting one of said potential core loading modes, Wherein the at least one predetermined criterion is selected from the following group: - the minimum and / or maximum gap between two fuel assemblies (40), - the maximum bending amplitude of each fuel assembly (40), and - the average degree of curvature of each fuel assembly (40).

2. The method according to claim 1, comprising: After the evaluation step, if the evaluation fails, the previous steps are repeated using at least one different potential core loading pattern.

3. The method of claim 1 , wherein the step of selecting one of the potential core loading patterns is based at least on the predicted bending calculation and the at least one predetermined selection criterion.

4. A determination method according to claim 1, wherein the definition of multiple potential core loading modes is carried out based on the characteristics of the operating cycle and the characteristics of the fuel assembly (40) and based on the design safety limitations of the core (12).

5. The determination method according to claim 1, wherein the plurality of potential core loading modes includes at least four core loading modes.

6. The determination method according to claim 5, wherein the plurality of potential core loading modes comprises more than ten core loading modes.

7. The determination method of claim 1, wherein the training data set comprises data derived from fuel assemblies (40) burned during operation of the nuclear power plant (10).

8. The determination method of claim 1, wherein the training data set comprises data derived from bending calculations performed on a fuel assembly (40).

9. The determination method according to claim 8, wherein the bending calculation comprises a finite element mechanical calculation and a coupled fluid structure computer fluid dynamics calculation.

10. The determination method according to claim 1, wherein the automatic learning algorithm is based on a neural network.

11. A computer program product comprising software instructions which, when executed by a computer, implement the determination method according to claim 1.

12. A method for loading a fuel assembly (40) in a nuclear reactor core (12), comprising at least the following steps: - selecting a core loading mode using the determination method according to claim 1, and - loading the fuel assembly (40) into the nuclear core (12) according to a selected core loading mode.

Images