Method for characterizing a loss of primary coolant in a nuclear reactor, based on a non-destructive measurement carried out on the primary circuit.

Abstract

Method for characterizing a loss of coolant (3) in a primary circuit (2) of a nuclear reactor cooled by the coolant, the primary circuit comprising at least one cooling loop having a pipe, around which is arranged a non-destructive measuring device (1, 11, 12, 13, 14) for measuring the concentration of a neutron absorber in the primary circuit, comprising: detection of the occurrence of a loss of coolant following a breach formed in a cooling loop; estimation of a concentration of neutron absorber, in each cooling loop, step b) being carried out in several consecutive instants so as to obtain a temporal evolution of the concentration of neutron absorber; comparison of each temporal evolution resulting from step b), for each cooling loop, to different modeled functions;based on comparisons made in step c), identification of the damaged cooling loop and / or estimation of the breach size.;

Description

Title of the invention: Method for characterizing a loss of primary coolant in a nuclear reactor, based on a non-destructive measurement carried out on the primary circuit. technical field

[0001] The technical field of the invention is the non-destructive measurement of a neutron absorber in the primary circuit of a nuclear reactor, in order to characterize the occurrence of a primary coolant loss. PRIOR ART

[0002] Figure 1 schematically illustrates the main elements of the primary circuit of a pressurized water reactor. In this example, the reactor comprises a vessel V connected to four cooling loops, two of which are shown, forming the primary circuit through which a heat transfer fluid circulates. In this example, the heat transfer fluid is pressurized water: the water is maintained under pressure (155 bar) so as to remain liquid up to an average temperature of 300 °C. Each cooling loop comprises a hot section H and an intermediate section I on either side of a steam generator V, as well as a cold section C, extending between the intermediate section and the vessel. The intermediate section I extends to a pump P. The intermediate section includes a U-shaped branch, extending between the steam generator and the pump P. The cold section C extends between the pump P and the vessel V.The primary circuit consists of 2 pipes, extending between several components of the GV type, pump, some components not being shown (pressurizer, injection circuits...).

[0003] The number of cooling loops can be between one and four depending on the reactor design.

[0004] In a water-cooled nuclear reactor, reactivity must be controlled to prevent the occurrence of a criticality situation. Reactivity in the primary circuit is adjusted by adding an isotope with high neutron absorption. Such an absorbing isotope can be 10B, added to the water in the form of boric acid. In a pressurized water reactor, the concentration of 10B generally varies between 1000 and 3000 ppm. In this presentation, a concentration of 10B varying between 0 and 5000 ppm is considered.

[0005] The occurrence of a breach in the primary circuit is an accidental situation, resulting in a drop in pressure and a loss of water from this circuit. This accident is usually referred to as a "Loss of Coolant Accident" (LOCA). This results in a A drop in pressure and a loss of water in the primary circuit can occur. The consequences include heating of the fuel rods, which must be kept to a minimum to maintain cooling of the reactor core and preserve the integrity of the fuel cladding. An auxiliary circuit, usually designated RIS (Safety Injection Circuit), is configured to inject borated water into the primary circuit. Several RIS safety injection circuits can be installed, which activate based on the pressure in the primary circuit. This keeps the fuel elements submerged in the event of a coolant loss and ensures the removal of residual heat generated by the reactor core after automatic shutdown. The injection of borated water by each safety injection circuit is triggered when the pressure in the primary circuit falls below certain thresholds.

[0006] The course and consequences of an APRP-type accident are directly related to the location and size of the breach. For example, a breach affecting the cold branch is considered more detrimental to core cooling, as it leads to the loss of water injected by the RIS. A breach affecting the hot branch is more detrimental to the containment building, due to the high temperature of the heat transfer fluid and the resulting pressure increase.

[0007] After a heat transfer fluid leak has been detected, it is useful to identify the affected loop as quickly as possible. It can also be useful to estimate characteristics of the leak, such as its size or location.

[0008] The publication Tabti N. et al., "Exploring Boron monitoring solutions in pressurized water reactors for enhanced loss of coolant accidents mitigation," hereafter referenced as [TABTI 2024], describes the possibility of installing a boron measurement system around a pipe belonging to the primary circuit of a pressurized water reactor. Indeed, the ability to measure the temporal evolution of boron concentration within the primary circuit can be of interest, particularly during an accident causing a loss of water from the primary circuit. In this publication, the evolution of characteristics at different locations in the primary circuit is modeled following the occurrence of a breach of varying dimensions. The characteristics studied include, for example, the void ratio and the boron concentration. The modeling is performed using the CATHARE code (Code for Accidental Thermohydraulic Analysis of Water Reactors), described in the publication P.Emonot, A. Souyri, JL Gandrille, and F. Barré. Cathare-3: A new System code for thermal-hydraulics in the context of the Neptune project. Nuclear Engineering and Design, 241 (11):4476-4481, 2011. The CATHARE code allows for the simulation of the thermohydraulic behavior of nuclear reactors. It is a code based on the two-fluid (liquid, gas) model. six equations, translating the conservation of mass, momentum and total energy.

[0009] The inventors have developed a method for characterizing a breach affecting the primary circuit: this involves identifying the damaged loop of the primary circuit, as well as estimating the size and / or position of the breach on the damaged loop. The method also makes it possible to monitor, over the long term, the evolution of the boron concentration in the primary circuit following an accident. This allows for the prevention of criticality risks. Description of the invention

[0010] A first object of the invention is a method for characterizing a loss of coolant in a primary circuit of a nuclear reactor cooled by the coolant, the primary circuit comprising at least one cooling loop extending around a reactor vessel, each cooling loop comprising a hot branch exiting the vessel, a steam generator, an intermediate branch, and a cold branch, the steam generator extending between the hot branch and the intermediate branch, the cold branch extending between the intermediate branch and the vessel, each cooling loop comprising a pipe, belonging to the hot branch or the intermediate branch or the cold branch, around which is arranged a device for non-destructively measuring the concentration of a neutron absorber in the pipe, the method comprising: a. detection of the occurrence of a loss of heat transfer fluid following a breach formed in a cooling loop, known as a damaged cooling loop; b. following step a), estimation of a neutron absorber concentration, in each cooling loop, by each device respectively arranged in each cooling loop, step b) being carried out in several consecutive instants so as to obtain a temporal evolution of the neutron absorber concentration; c. comparison of each time evolution resulting from step b), for each cooling loop, to different modeled functions, each modeled function representing a time evolution of the neutron absorber concentration, in the pipe, following the formation of the breach, each modeled function being parameterized by a size and / or a position of the breach in the damaged cooling loop; d. Based on the comparisons made in step c), for each cooling loop, identification of the damaged cooling loop and / or estimation of the size and / or position of the breach.

[0011] Steps c) and d) are preferably implemented by a control unit.

[0012] Step c) can implement, in each cooling loop: - the first modeled functions taking into account a breach in another cooling loop; - second modeled functions, taking into account a breach in said cooling loop.

[0013] Step d) may include an estimation of the size and location of the breach in the damaged cooling loop.

[0014] According to one possibility: - each intermediate branch includes a U-shaped branch, arranged between the steam generator and the pump, the U-shaped branch extending on either side of a bottom, forming a low point of the U-shaped branch; - each measuring device is positioned facing the bottom of the U-shaped branch.

[0015] Step c) can be implemented by a maximum likelihood type algorithm.

[0016] Each measuring device may include: - at least one neutron source, configured to emit neutrons into the pipe facing which it is positioned; - at least one neutron detector, configured to detect neutrons emitted by the neutron source and having passed through and / or been backscattered by the conduit.

[0017] A second object of the invention is a system for characterizing a loss of coolant in a primary circuit of a nuclear reactor cooled by the coolant, the primary circuit comprising several cooling loops extending around a reactor vessel, each cooling loop comprising a pipe, the system comprising: - at least one measuring device, the measuring device or devices being configured to determine a concentration of neutron absorber in the pipe at different times; - a control unit, connected to the measuring device(s), and configured to implement steps c) and d) of a process according to the first object of the invention.

[0018] Each measuring device may include: - at least one neutron source, configured to emit neutrons into the pipe facing which the measuring device is positioned; - at least one neutron detector, configured to detect neutrons emitted by the neutron source and having passed through and / or been backscattered by the conduit.

[0019] Each measuring device may include: - a front end, configured to be positioned facing the pipe; - a neutron reflector, the neutron source being disposed between the front end and the neutron reflector, the neutron reflector being configured to reflect neutrons, emitted by the neutron source, back towards the front end.

[0020] The reflector is preferably made of a material with an albedo greater than 0.7, albedo being the ratio of the neutron flux incident on the material to the neutron flux backscattered by the material. The reflector may be made of a material containing carbon and / or hydrogen and / or beryllium. The reflector may have a thickness of less than 30 cm or less than 20 cm. The neutron source may be disposed in a cavity extending between the reflector and the front end.

[0021] The measuring device may include a front screen, disposed between the detector or each detector and the front end, made of a metallic material, configured to attenuate at least 75% of gamma radiation with an energy of 1 MeV.

[0022] The invention will be better understood upon reading the description of the exemplary embodiments presented later in this description, in connection with the figures listed below. FIGURES

[0023] Figure 1 schematically illustrates the main elements of a primary circuit of a pressurized water reactor.

[0024] Fig. 2 shows a first example of a measuring device allowing implementation of the invention.

[0025] Fig. 3 shows a second example of a measuring device enabling an implementation of the invention.

[0026] Fig. 4 shows a third example of a measuring device enabling an implementation of the invention.

[0027] Fig. 5 illustrates an example of a modeled breach.

[0028] Fig. 6 shows a model of the evolution of the pressure in the primary circuit (ordinate axis - unit MPa) as a function of time (abscissa axis, unit seconds) following the occurrence of a breach, and this for different breach sizes, between 2 and 10 inches.

[0029] Fig. 7A shows the directions of water flow in a cooling loop following the occurrence of a breach.

[0030] Fig. 7B shows the vacuum ratio in the cooling loop schematically represented in Fig. 7A.

[0031] Fig. 8A shows a time evolution of the boron concentration in the U-shaped branch of an intact cooling loop.

[0032] Fig. 8B shows a time evolution of the boron concentration in the U-shaped branch of a damaged cooling loop.

[0033] Fig. 8C shows a time evolution of the vacuum rate in the U-shaped branch of an intact cooling loop.

[0034] Fig. 8D shows a time evolution of the vacuum rate in the U-shaped branch of a damaged cooling loop.

[0035] Fig. 9 schematically illustrates an arrangement of measuring devices in the four cooling loops of a nuclear reactor comprising four cooling loops.

[0036] Fig. 10 represents the main steps of a method for identifying a cooling loop.

[0037] Figures 11A to 111 illustrate the evolution of boron concentration at the U-shaped branch of cooling loops as schematically shown in Figures 7A or 7B, respectively, for different breach sizes. Each of these figures shows the evolution of boron concentration: - in a damaged cooling loop, including the breach (dashes); - in an intact cooling loop, not including the breach (solid line). PRESENTATION OF SPECIFIC IMPLEMENTATION METHODS

[0038] An important aspect of the invention is to place a measuring device 1 on several cooling loops, ideally on each cooling loop, of a nuclear reactor. Each measuring device 1 is configured to face a pipe of the primary circuit. The portion to be analyzed can be located in the U-shaped branch as described below. In the example described, the nuclear reactor has four cooling loops, only two of which are shown in [Fig. 1]. Four measuring devices are therefore placed on each cooling loop, on the U-shaped branch of each loop.

[0039] Figure 2 schematically illustrates a first example of a measuring device 1 enabling an implementation of the invention. The measuring device 1 is arranged to be positioned facing a pipe 2 of the primary circuit, and preferably in contact with said pipe.

[0040] Device 1 is intended to estimate the concentration of an isotope 4 in the heat transfer fluid 3. Isotope 4 is a neutron-absorbing isotope. In this example, The isotope is 10B. In this example, the heat transfer fluid 3 is borated water (a mixture of water and boric acid). The diameter of the pipe 2 is typically between 80 and 100 cm. Each measuring device 1 is positioned facing a pipe 2, and preferably in contact with it.

[0041] Coupling to the primary circuit is accompanied by technical difficulties, among which is the large size of the primary circuit pipes, generally between 80 cm and 100 cm.

[0042] In addition to this, there is potentially significant γ radiation due to the presence of γ-emitting isotopes in the primary circuit: these isotopes correspond to activation products present in the heat transfer fluid (for example 60Co), to which fission products (for example 137Cs) may potentially be added.

[0043] Another constraint is a potential temperature variation: in a pressurized water reactor, during shutdown phases, the temperature of the water in the primary circuit can be around 50°C. During operation, the water temperature can reach or exceed 300°C.

[0044] In general, each measuring device can be as described in application FR2410884, filed on 08 / 10 / 2024.

[0045] In the embodiment shown in [Fig. 2], each device comprises a cavity 10, which, in this example, is filled with air. The cavity 10 is delimited by a shell 11 and extends to a front end 12. The shell 11 extends opposite the front end 12. The front end 12 is intended to be positioned facing the conduit 2. In this example, the cavity 10 has a front opening, at the front end 12, leading to the conduit 2. In this example, the cavity is substantially hemispherical in shape. The cavity can extend over a diameter (or longer diagonal) of between 10 cm and 30 cm, and preferably between 10 cm and 20 cm.

[0046] The device includes a neutron source 13. This can be an isotopic source, comprising one or more isotopes, enabling the emission of neutrons. In this example, the isotopic source is an Am-Be type source, whose neutron emission is based on an (a, n) type reaction. Other types of sources, based on the same principle, are conceivable, for example 241Am-Li or 244Cm-Be. The neutron source 13 can be a spontaneous fission type source, the most common isotopes being Cf or Cm. It can also be a neutron generator. The use of a neutron generator allows precise control of the neutron emission, at the expense of significant size and safety-related difficulties for use within a reactor building. In the following section Description, the neutron source used is an Am-Be type source, except where otherwise specified.

[0047] The device includes at least one detector 14, extending between the shell 11 and the front end 12. In the example shown, the device includes two detectors 14, extending into the cavity 10. Using two detectors provides redundancy of the measurement channels. This allows a malfunction affecting one of the two detectors to be detected. The device may include more than two detectors, associated with each measurement channel.

[0048] Each detector 14 is configured to provide a counting rate TC corresponding to a number of neutrons detected per second. In this example, each detector 14 is a boron-deposited gaseous proportional counter. This type of counter is common. Under irradiation with a neutron flux, charged particles (α and βLi) are formed by (ν, α) capture. The charged particles are detected, by polarized electrodes, via secondary ionizations in the detector. Each detector can be a Photonis CPNB64 type proportional counter, whose detection efficiency is 25 counts / cm² (or counts per second per neutrons per cm² per second).

[0049] Each measuring device comprises a processing unit 20, connected to the detectors 14 by a wired link, preferably robust against neutron and gamma radiation, for example using mineral cables suitable for use in a reactor building. The processing unit 20 may include a microprocessor. The processing unit 20 is programmed to perform an estimation of a concentration of 10B in the fluid 3 based on count rates TC respectively measured by each detector 14.

[0050] The processing unit 20 can be programmed to calculate a count rate TC from a detector 14. From the count rate, the concentration of 10B is estimated, taking into account the temperature Tf of the water in the primary circuit. For example, a calibration base has been previously stored in the processing unit.

[0051] According to one possibility, the counting rates of the two detectors can be added together, so as to increase their sensitivity.

[0052] Some analytical expressions are proposed in the literature that establish a relationship between the boron concentration (C) and the count rate (TC). For example, the publication by Pirat P, "Boronline, a new generation of boron meter," 2nd International Conference on Advancements in Nuclear Instrumentation, Measurement Methods and their Applications, 2011, IEEE, pp. 1-2, describes an expression of the type:

[0053] TC—-----(1) c^cf+bC+c

[0054] Where a, b etc are adjustment parameters, stored in the calibration base: these are real numbers which depend on the temperature of the heat transfer fluid, the detector used and the configuration in which the detector is located, taking into account for example the presence as well as the geometry and nature of a moderator material near the detector.

[0055] A Shinkori expression, well-suited to high boron concentrations, can also be used. This expression comes from the publication by Si Hyeong Sung and Hee Reyoung Kim, “Experimental characterization of the accuracy of multidetector boron meter for operational safety of reactors.” International Journal of Energy Research, 42(8):2701-2709, 2018. According to this expression: [00561 c = <b+Æ+^+^(r)

[0057] The parameters av ai and ai are real numbers, stored in the calibration base, which depend on the temperature of the heat transfer fluid, the detector used and the configuration in which the detector is located.

[0058] The shell 11 delimiting the cavity 10 may advantageously include shielding 16 so as to form a screen for each detector 14 against ambient radiation. The shielding 16 is preferably made of a metal with an atomic number Z greater than 60, or even 80, and a high density, preferably greater than 10 g / cm³. It may, in particular, be lead or a tungsten alloy. The thickness of the shielding may, for example, be 10 cm. It is preferably between 2 cm and 20 cm. A thin layer of cadmium 17, with a thickness between 0.5 mm and 2 mm, may cover the shielding 16 for radiation protection purposes, as cadmium has a high neutron absorption.

[0059] Preferably, each measuring device includes a reflector 15. By reflector is meant a material that allows backscattering of neutrons so as to maximize the quantity of neutrons interrogating the pipe. The backscattered neutrons are neutrons initially emitted in a direction opposite to the pipe, then undergoing one or more successive scatterings in the reflector, so as to propagate towards the pipe.

[0060] It is preferable that the characteristics of the reflector (material, thickness) be such that the albedo is greater than 0.7, or even greater than 0.8. The reflector 15 can extend to a thickness between 2 cm and 20 cm, or between 2 cm and 15 cm. Various materials suitable for forming a reflector have been tested: - Graphite: albedo 0.937 - density 1.6 g.cm 3; - MgO: magnesium oxide: albedo 0.859 - density 3.58 g.cm3; - Zirconium silicide (Zr3Si2): albedo 0.833 - density 5.88 g.cm3.

[0061] Other materials can form a reflector, in particular materials containing light atoms, which are conducive to elastic neutron scattering, for example C or Be or H. The reflector can be formed by a polymer, for example polyethylene.

[0062] Figure 3 represents a second embodiment, in which each device The measuring device 1 includes a front screen 18 at the front end 12. The front screen 18 is made of a material similar to the material forming the shielding 16, that is, a material with a sufficiently high atomic number and density, for example, lead or a tungsten alloy, to absorb a significant portion of the gamma radiation, at least 75% or at least 90%, or even more, of gamma radiation with an energy of 1 MeV. The front screen 18 is intended to protect the detectors 14 from the radiation emanating from the conduit. Thus, the front screen 18 is interposed either at the front end 12 or between the front end and each detector.

[0063] Figure 4 represents a third embodiment, in which the cavity of Each measuring device includes auxiliary detectors 19, formed by proportional counters, without boron deposition. The auxiliary detectors are intended to establish an auxiliary detection signal, solely representative of the irradiation γ to which the detectors 14 are subjected. By calibration, an irradiation level can be extracted from the auxiliary detection signal generated by each detector 19. A contribution of the irradiation γ to the detection signal generated by the detectors 14 can then be estimated. This contribution can be subtracted from the detection signal generated by each detector 14, so as to extract the contribution of the detection signal due to the detected neutrons. This presupposes knowledge of the response of each detector to the irradiation level γ. Other types of auxiliary detectors can be used, for example, Geiger-Müller detectors or detectors based on scintillator crystals.In this configuration, shielding 16 is not required around the measuring cavity.

[0064] Regardless of the embodiment, the cavity 10 can be hollow (for example filled with air) or comprise a reflective material.

[0065] An important aspect of the invention is that during an accident situation, the temporal monitoring of the boron concentration, in at least one, or even in each cooling loop, can first of all make it possible to identify the damaged loop, and possibly to estimate the size, and / or the position of the rupture which caused the loss of heat transfer fluid.

[0066] It is customary to classify APRP-type accidents into three main categories, according to their size. By breach size, we mean a diameter, called the hydraulic diameter Dh, such that:

[0067] D / =

[0068] Where A and P are respectively the surface area and the perimeter of the breach.

[0069] Small breaches correspond to breaches whose size is less than 1 inch, intermediate breaches are those whose size is between 1 and 14 inches, large breaches are those whose size exceeds 14 inches.

[0070] According to the principles described in the prior art publication [TABTI 2024], modeling was conducted simulating the evolution over time of primary circuit characteristics following the occurrence of a buttonhole breach, as shown in [Fig. 5]. A buttonhole breach is defined as a crack or partial rupture in a pipe or conduit. This type of breach resembles an elongated slit, similar to the opening of a buttonhole. The size and position of a buttonhole breach can influence the dynamics of the accident, particularly the extent of refrigerant loss.

[0071] The occurrence of breaches ranging in size from 2 inches (5.08 cm) to 10 inches (25.4 cm) was modeled in 2-inch increments. As mentioned in publication [TABTI 2024], the occurrence of a breach results in four successive phases, during which the pressure in the primary circuit decreases in steps: Figure 6 shows the evolution of pressure (ordinate axis - unit MPa) as a function of time (abscissa axis - unit seconds). The size of the breach is also shown. - Phase 1: Sudden depressurization, leading to the automatic shutdown of the pump reactor and the activation of backup circuits: safety injection circuits (RIS) and the emergency power supply circuit for the steam generators. Natural circulation of the heat transfer fluid, by thermosiphon, is established. The heat transfer fluid is initially single-phase, then, under the effect of depressurization, two-phase. - Phase 2: The water inventory of the primary circuit decreases, in two-phase thermosiphon mode, until continuous circulation within the steam generator becomes impossible, thus ending the two-phase circulation and initiating the so-called "heat pipe" operating mode. The steam generator contains numerous vertical pins, enabling the production of steam in a secondary circuit by heat transfer. During this phase, the steam from the primary circuit, formed following depressurization, condenses in the generator, inducing liquid circulation from the steam generator to the reactor vessel. During this phase, the hot and cold branches are thermodynamically isolated from each other. During this phase, a water plug This forms in the U-shaped branch. Figure 7A shows the direction of liquid water flow in a loop at a specific instant as the water flows from the steam generator to the tank, through the hot branch as well as through the intermediate and cold branches. The loop has been modeled and segmented into a mesh. In each mesh, the direction of flow has been determined and is symbolized by an arrow. The formation of flows in opposite directions is observed within the steam generator and the U-shaped branch.

[0072] Figure 7B shows the accumulation of water in the U-shaped branch at the same instant as that shown in Figure 7A. Figure 7B shows the void ratio in the loop, the void ratio corresponding to the volume fraction occupied by the gaseous phase. A value of zero corresponds to a purely liquid phase. A value of 1 corresponds to a purely gaseous phase. It can be observed that the void ratio is minimal in the U-shaped branch. - Phase 3: Following the decrease in the water inventory of the primary circuit, the reactor core is partially dewatered. This results in a temperature increase, inducing continuous steam formation throughout the entire primary circuit, including the breach. The steam formation causes the expulsion of the water plug formed in each U-shaped branch. - Phase 4: During this phase, the transition of the breach to the vapor phase slows down the decrease in the water inventory and leads to the stabilization of the pressure.

[0073] The boron remains exclusively contained in the liquid phase. However, it is at the U-shaped branch, and more precisely at the bottom of the U-shaped branch, that the quantity of water in the liquid phase is significant during the four phases described in relation to Figures 6, 7A, and 7B. This is due to the specific shape of the U-shaped branch, the bottom of the branch forming a low point, which promotes liquid accumulation. The U-shaped branch is conducive to water accumulation during the pressure drop in the primary circuit.

[0074] Figures 8A and 8B represent the evolution of the Boron concentration (ordinate axis - unit ppm), as a function of time (abscissa axis - unit seconds) respectively at the bottom of the U-shaped branch, respectively: - in an integrated cooling loop, without the breach; - in the damaged cooling loop, within which the breach is located.

[0075] In each of these figures, the boron concentration has been modeled for different breccia sizes, ranging from 2 to 10 inches.

[0076] It is observed that the temporal evolution of the boron concentration differs depending on the loop considered (intact branch or damaged branch), but also depending on the size of the breach. The temporal evolution of the boron concentration can thus be used as a signature of the accident, making it possible to identify the damaged cooling loop, as well as to estimate the characteristics of the breach, for example, its position and / or size.

[0077] The breach can be positioned: - at the level of the cold branch, between the safety injection and the core; - or at the cold branch, between the primary circuit pump and the safety injection; - or at the level of the intermediate branch; - or at the level of the hot branch.

[0078] The invention is based on the fact that by monitoring the temporal evolution of the boron concentration in the primary circuit, and particularly in the U-shaped branch, it is possible to identify whether the loop in which the measurement is taken is intact or damaged. It is also possible to estimate the size of the breach and, if necessary, its location. This also allows for the long-term management of criticality risk associated with the accidental release of boron (ARBP) by monitoring the dynamics of the boron concentration in the days following the accident. Indeed, a device directly applied to the primary circuit provides real-time information on the evolution of the boron concentration. This makes it possible to prevent any excessive dilution of boron, with the aim of controlling the criticality risk.

[0079] Figures 8C and 8D represent the evolution of the void ratio (ordinate axis) as a function of time (abscissa axis, unit seconds) respectively at the bottom of the U-shaped branch: - in an integrated cooling loop, without the breach; - in the damaged cooling loop, within which the breach is located;

[0080] and this for gap sizes between 2 and 10 inches.

[0081] Beyond a critical value of the void ratio, the measurement is not usable. The critical value can be determined by calibrating a bore meter, during which measurements are taken for different known void ratios.

[0082] It is noted that for small or medium-sized breaches, that is to say for breaches of less than 6 inches, the void ratio at the level of the U-shaped branch is relatively low, which allows usable measurements to be taken.

[0083] Figure 9 schematically shows a top view of tank V to which four cooling loops are connected. Figure 9 shows a measuring system, comprising Four measuring devices, lb 12, I3, 14, as described in relation to Figures 2 to 4, are positioned at the U-shaped branches of each cooling loop. The system includes a control unit 30, connected to each measuring device, and configured to collect the measured concentrations over time, following the detection of a breach in the primary circuit.

[0084] The occurrence of a breach in the primary circuit is detected by the usual control means of the nuclear reactor, for example by a drop in the primary circuit pressure. After the formation of a breach has been detected, measurements of the boron concentration can be initiated in order to identify the damaged cooling loop, and possibly to estimate the characteristics of the breach, in particular its size and / or position on the damaged loop.

[0085] Figure 10 schematically illustrates the main steps of the process.

[0086] Step 100: detection of the formation of the breach, for example by noting a drop in pressure at the level of the primary circuit.

[0087] Step 110: Acquisition of measurements. As previously indicated, following the detection of the breach formation, the boron concentration is measured at different times, for example, every second. At each time t, each measuring device h generates a count rate TCi(t), which corresponds to the number of neutrons detected per second by each detector 14. The subscript i designates each cooling loop. In the example shown in [Fig. 9], i is an integer between 1 and 4. It should be noted that the boron concentration is preferably measured in the primary circuit during the normal operating phases of the reactor.

[0088] Step 120: Concentration calculation. From the counting rate TC, each measuring device h determines a boron concentration Ci(t) at time t. This allows for a temporal evolution of the boron concentration for each cooling loop, whether intact or damaged. The transition between each counting rate and each concentration can be performed using expressions such as those described previously: see (1) or (1'). This assumes consideration of the temperature of the fluid circulating in the primary circuit.

[0089] Step 130: During this step, each time evolution, determined for each cooling loop, is compared with different models: - of the first evolution functions determined by modeling, and representative of the evolution of boron concentration in an integral state of the cooling loop i, taking into account different parameters 8 of breach in a damaged cooling loop, different from the cooling loop'; - of the second evolution functions determined by modeling, and representative of the evolution of boron concentration in a damaged state of the cooling loop.

[0090] refers to the parameters of the breach, which may include the size of the breach or its position.

[0091] Figures 1 IA to 111 represent evolution functions f and taking into account gap sizes respectively equal to 1 inch, 2 inches, 3 inches, 4 inches, 5 inches, 6 inches, 7 inches, 8 inches, 10 inches.

[0092] During step 130, the evolution of the boron concentration, determined for each loop, is compared to each evolution function in order to determine the evolution function closest to the measured evolution. This makes it possible to identify the damaged cooling loop, which corresponds to the cooling loop for which the time evolution of the boron concentration follows an evolution function So representative of the occurrence of a breach with parameters #. In the other cooling loops, the time evolution of the boron concentration follows a function f (r

[0093] Step 130 can be implemented by defining a likelihood function, or log-likelihood function, so as to determine, for each cooling loop i, the function f 0 or Sg which best corresponds to the time evolution of the concentration Ci(t) resulting from the measurements, and this on each loop.

[0094] During this step, a reconstruction algorithm such as the Expectation-Maximization algorithm can be implemented to calculate the maximum of the log-likelihood

[0095] Suppose that the boron content CB(t) at a given time t is modeled by a function f;(CB(t)l0) and gi(CB(t)IO), where: - 0 represents the characteristics of the breach - f; and g; represent the functions describing the time evolution of boron content in intact and broken loops respectively.

[0096] The likelihood function can be: [00971 L (0) =

[0098] The log-likelihood is then: [00991 1 (6) = E^lnf,-(^(()19) +lng,.(C w ( Goose)

[0100] (js corresponds to the concentration of boron in loop i at time t and I corresponds to the number of loops considered.

[0101] ln corresponds to the natural logarithm.

[0102] The maximum log-likelihood is then calculated to estimate the parameter 0 corresponding to the characteristics of the breach (size and position):

[0103] £ j(a\ LJ 0 = argranax l 0 ü X /

[0104] Figures 1 IA to 1 IC are representative of small breaches, respectively 1 inch, 2 inches, 3 inches. In this type of situation, temporal and real-time monitoring of the boron concentration in the primary circuit allows an estimation of the breach size, but does not allow, or makes it more difficult, to identify the damaged loop.

[0105] Figures 11D to 11F are representative of average-sized breaches, respectively 4 inches, 5 inches, and 6 inches. In this type of situation, temporal and real-time monitoring of the boron concentration in the primary circuit allows for an estimation of the breach size, as well as the identification of the damaged loop.

[0106] Figures 1IG and 1II are representative of large-sized breaches, for which temporal and real-time monitoring of boron concentration can be used to characterize the end of the transient, which can extend from a few hours to a few days after the occurrence of the breach. Step 140

[0107] During step 130, parameters 6 are determined that maximize the likelihood function L(0) or Z(0) for each cooling loop. This allows the size of the breach and possibly its position in the damaged cooling loop to be estimated.

[0108] Estimating the position of the breach assumes that the evolution functions ifc» or „ ï (c / tW) are defined for different sizes and different positions of the breach.

Claims

Demands

1. A method for characterizing a loss of coolant (3) in a primary circuit of a coolant-cooled nuclear reactor, the primary circuit comprising at least one cooling loop extending around a reactor vessel, each cooling loop comprising a hot branch (H) emerging from the vessel, a steam generator (SG), an intermediate branch (I), and a cold branch (C), the steam generator extending between the hot branch and the intermediate branch, the cold branch extending between the intermediate branch and the vessel, each cooling loop comprising a pipe (2), belonging to the hot branch or the intermediate branch or the cold branch, around which is arranged a non-destructive measuring device (1, lb 12, G, 14) for measuring the concentration of a neutron absorber in the pipe, the method comprising: a. detection of the occurrence of a loss of heat transfer fluid following a breach (B) formed in a cooling loop, referred to as a damaged cooling loop; b. following step a), estimation of a neutron absorber concentration, in each cooling loop, by each device respectively arranged in each cooling loop, step b) being carried out in several consecutive instants so as to obtain a temporal evolution of the neutron absorber concentration; c. comparison of each time evolution resulting from step b), for each cooling loop, to different modeled functions each modeled function representing a time evolution of the neutron absorber concentration in the pipe following the formation of the breach, each modeled function being parameterized by a size and / or position of the breach in the damaged cooling loop; d. Based on the comparisons made in step c), for each cooling loop, identification of the damaged cooling loop and / or estimation of breach size and / or position; steps c) and d) being implemented by a processing unit.

2. A method according to claim 1, wherein step c) implements, in each cooling loop: - first modeled functions . ( t ) 10 ) ), taking into account a breach in another cooling loop; - second modeled functions ( g ( Çg • ( t ) 10 ) )' taking into account a breach in said cooling loop.

3. A method according to any one of the preceding claims, wherein step d) comprises an estimation of the size and location of the breach in the damaged cooling loop.

4. A method according to any one of the preceding claims, wherein: - each intermediate branch comprises a U-shaped branch, disposed between the steam generator and the pump, the U-shaped branch extending on either side of a bottom, forming a low point of the U-shaped branch; - each measuring device is disposed facing the bottom of the U-shaped branch.

5. A method according to any one of the preceding claims, wherein step c) is implemented by a maximum likelihood type algorithm.

6. A method according to any one of the preceding claims, wherein each measuring device comprises: - at least one neutron source, configured to emit neutrons into the pipe facing which it is positioned; - at least one neutron detector, configured to detect neutrons emitted by the neutron source and having passed through and / or having been backscattered through the pipe.

7. System for characterizing a loss of coolant in a primary circuit of a nuclear reactor cooled by the coolant, the primary circuit comprising several cooling loops extending around a nuclear reactor vessel, each cooling loop comprising a pipe, the system comprising: - at least one measuring device, the measuring device or each measuring device (1, lb 12, 12, I4) being configured to determine a concentration of neutron absorber in the analysis pipe at different times; - a control unit (30), connected to the measuring device or each measuring device, and configured to carry out steps c) and d) of a method according to any one of the preceding claims.

8. System according to claim 7, wherein each measuring device comprises: - at least one neutron source (13), configured to emit neutrons into the pipe opposite which the measuring device is positioned; - at least one neutron detector (14), configured to detect neutrons emitted by the neutron source and having passed through and / or having been backscattered through the pipe.

9. System according to claim 8, wherein each measuring device comprises - a front end (12), configured to be disposed facing the pipe; - a neutron reflector (15), the neutron source being disposed between the front end and the neutron reflector, the neutron reflector being configured to reflect neutrons, emitted by the neutron source, back to the front end.

10. System according to claim 9, wherein the reflector is formed of a material having an albedo greater than 0.7, the albedo being a ratio between a neutron flux incident to the material and a neutron flux backscattered by the material.

11. System according to any one of claims 9 or 10, wherein the reflector is formed of a material comprising carbon and / or hydrogen and / or beryllium.

12. System according to any one of claims 10 or 11, wherein the reflector extends to a thickness of less than 30 cm or less than 20 cm.

13. System according to any one of claims 9 to 12, wherein the neutron source is disposed in a cavity, extending between the reflector and the front end.

14. System according to any one of claims 9 to 13, wherein the measuring device comprises a front screen (18), disposed between the detector or each detector and the front end, formed of a metallic material, configured to attenuate at least 75% of gamma radiation with an energy of 1 MeV.