Preheating system and preheating method for nuclear fuel assembly

Abstract

The invention provides a preheating system and method for a nuclear fuel assembly, and the method comprises the steps: inputting a preheating data set of the nuclear fuel assembly under various historical working conditions into a preset preheating model for training, and obtaining a predicted temperature value of a target temperature measurement point on the nuclear fuel assembly in a preset thermal uniformity threshold range; training the predicted temperature value according to a preset loss function, and determining loss values of the preheating model under different historical working conditions; the preheating model is updated through the loss value, and a trained preheating model is obtained; and inputting the preheating data set of the nuclear fuel assembly under the current working condition into the trained preheating model to generate a preheating strategy, and controlling a preheating system to heat argon in response to the preheating strategy. Training is carried out by combining preheating data of the nuclear fuel assembly under different historical working conditions, a preheating system is controlled to execute a preheating strategy, a target temperature measuring point reaches the actually needed temperature, and therefore it is guaranteed that all parts of the nuclear fuel assembly are evenly heated.

Description

Technical Field

[0001] The invention relates to the field of nuclear reactors, and in particular to a preheating system and a preheating method for a nuclear fuel assembly. Background Art

[0002] Before loading new components of a sodium-cooled fast reactor into the core, the nuclear fuel components need to be heated to a certain temperature. When a large number of nuclear fuel components are continuously loaded into the reactor, the new components of a sodium-cooled fast reactor are preheated in a temporary heating container and then loaded into the new component conversion barrel. The new component conversion barrel does not need to be cooled before loading the components, which can effectively save the time of repeatedly cooling and heating the new component conversion barrel again.

[0003] The related technology preheats the nuclear fuel assembly by heating argon gas, and then the fan accelerates the flow of argon gas, and the argon gas heated to the target temperature is quickly circulated and heat exchanged in the container to achieve the function of heating. Although heating by circulating heat exchange with argon gas can ensure relatively uniform heating, it is not easy to ensure that each component can be evenly heated to the target temperature when a large number of nuclear fuel assemblies are continuously loaded into the stack. Since the argon flow rate and heat exchange efficiency at different positions in the container may vary, some components may not be heated enough and some components may be overheated, affecting the performance and safety of the components. Summary of the invention

[0004] In view of this, an object of the present invention is to provide a preheating system and a preheating method for a nuclear fuel assembly, which solves the problem of uneven heating of the nuclear fuel assembly during the preheating process in the prior art.

[0005] In a first aspect, the present application provides a method for preheating a nuclear fuel assembly, the method comprising: Inputting the preheating data set of the nuclear fuel assembly under various historical working conditions into a pre-set preheating model for training, and obtaining the predicted temperature value of the target temperature measurement point on the nuclear fuel assembly within a preset thermal uniformity threshold range; The predicted temperature value is trained according to a preset loss function to determine the loss value of the preheating model under different historical working conditions; and based on a preset update rule, the preheating model is updated by the loss value to obtain a trained preheating model; The preheating data set of the nuclear fuel assembly under the current working condition is input into the trained preheating model to generate a preheating strategy, and the preheating system is controlled to heat the argon gas in response to the preheating strategy.

[0006] In one embodiment, the preset data set at least includes surface temperature data of multiple target temperature measurement points under various historical working conditions, the Biot number of the target temperature measurement point, the opening of the argon regulating valve, the pressure of argon near the target temperature measurement point, the power of the heater, the number of nuclear fuel assemblies, and the speed of the fan; the preheating data set of the nuclear fuel assembly under various historical working conditions is input into a preset preheating model for training to obtain the predicted temperature value of the target temperature measurement point on the nuclear fuel assembly within a preset thermal uniformity threshold range, specifically including: A Markov joint probability distribution is constructed based on the relationship between the surface temperature data of the target temperature measurement point, the Biot number of the target temperature measurement point, the opening of the argon regulating valve, the pressure of the argon near the target temperature measurement point, the power of the heater, the number of nuclear fuel assemblies and the speed of the fan; The predicted temperature value of each target temperature measurement point satisfying the thermal uniformity threshold range is determined based on the Markov joint probability distribution.

[0007] In one embodiment, the Markov joint probability distribution is constructed based on the relationship between the surface temperature data of the target temperature measurement point, the Biot number of the target temperature measurement point, the opening of the argon regulating valve, the pressure of the argon near the target temperature measurement point, the power of the heater, the number of nuclear fuel assemblies and the speed of the fan, specifically including: constructing energy groups according to the surface temperature data and the Biot number of two adjacent target temperature measurement points respectively, and constructing energy groups according to the surface temperature data of the target temperature measurement point, the opening of the argon regulating valve, the pressure of argon near the target temperature measurement point, the power of the heater, the number of nuclear fuel assemblies and the rotation speed of the fan, and establishing an energy function based on the energy group; Generating a corresponding Boltzmann potential function according to the energy function, and determining a change parameter of the Boltzmann potential function based on maximum likelihood estimation; The Markov joint probability distribution is generated based on the variation parameter and the Boltzmann potential function.

[0008] In one embodiment, the Boltzmann function is expressed as: in, The energy function of the energy group constructed for the surface temperature data of two adjacent target temperature measurement points, The energy function of the energy group constructed by the Biot number of two adjacent target temperature measurement points, An energy function of an energy group constructed for the surface temperature data of a target temperature measuring point, the opening of an argon regulating valve, the pressure of argon near the target temperature measuring point, the power of a heater, the number of nuclear fuel assemblies, and the rotation speed of a fan; in, and are the surface temperature data of two adjacent target temperature measurement points, and are the Biot numbers of two adjacent target temperature measurement points, T is the surface temperature data, V is the opening, P is the pressure, W is the power, R is the speed, N is the number of nuclear fuel assemblies, is the target surface temperature data, Target opening, is the target pressure, is the target power, is the target speed, N t is the target number of nuclear fuel assemblies, is the average surface temperature, , , , , , and are all weight parameters.

[0009] In one embodiment, determining the variation parameter of the Boltzmann potential function based on maximum likelihood estimation specifically includes: The conditional probabilities of the surface temperature data of the target temperature measuring point, the Biot number of the target temperature measuring point, the opening of the argon regulating valve, the pressure of the argon near the target temperature measuring point, the power of the heater, the number of nuclear fuel assemblies and the speed of the fan under the condition that other variables are fixed are constructed by pseudo-likelihood estimation, and the pseudo-approximate likelihood function is determined according to the conditional probabilities; The gradients of the pseudo-approximate likelihood function for the corresponding variables are calculated respectively, and the parameter values ​​of the corresponding variables are updated by the gradient ascent method until the pseudo-approximate likelihood function satisfies the preset iteration conditions, and the parameter values ​​of the corresponding variables are used as the variation parameters of the Boltzmann potential function.

[0010] In one embodiment, the loss value includes a mean square error loss value and a mean absolute error loss value, and the expression of the loss function is: in, is the mean square error loss value, is the mean absolute error loss value, n is the number of target temperature measurement points, The surface temperature data of the target temperature measurement point in the preheating data set; To predict the temperature value.

[0011] In one embodiment, after the preheating data set of the nuclear fuel assembly under the current working condition is input into the trained preheating model to generate the preheating strategy, the method further includes: Matching reference thermal uniformity data corresponding to the current operating condition for each target temperature measurement point based on historical operating conditions; Comparing the uncertainty value between the thermal uniformity data of the preheating strategy of the target temperature measurement point and the reference thermal uniformity data; Marking the target temperature measurement point where the uncertainty value exceeds a preset uncertainty threshold as an abnormal target temperature measurement point; Extract the preheating parameters of the abnormal target temperature measurement point in the preset data set under the historical working conditions corresponding to the reference thermal uniformity data, update the preheating data set according to the preheating parameters, and regenerate a new preheating strategy.

[0012] In one embodiment, updating the preheating data set according to the preheating parameters specifically includes: Randomly delete the data in the preheating data set under the current working condition by using a random forest algorithm, and establish multiple groups of deleted data sets to be restored after deletion for the same data; Based on a pre-established recovery model, recover the deleted data set to obtain a recovery data set; The similarity between the deleted data set and the restored data set is calculated, and the preheating data set is updated with the preheating parameters corresponding to the abnormal target temperature measurement points in the restored data set whose similarity meets the similarity threshold.

[0013] In a second aspect, the present application provides a preheating system for a nuclear fuel assembly, comprising an argon regulating valve for controlling the flow of argon gas, a heater for heating argon gas, a fan for conducting the flow of argon gas, a first temperature sensor for monitoring the surface temperature data of a target temperature measurement point, a second temperature sensor for monitoring the ambient temperature of the nuclear fuel assembly, a camera for identifying the number of nuclear fuel assemblies, and a control center for controlling the argon regulating valve, heater, fan, camera, first temperature sensor and second temperature sensor to execute a preheating strategy, wherein the control center is used to execute a preheating method for a nuclear fuel assembly as described in any one of the first aspects.

[0014] In a third aspect, the present application provides a computer-readable storage medium storing instructions for a processor to load and execute a method for preheating a nuclear fuel assembly as described in any one of the first aspects.

[0015] In the preheating system and preheating method of the nuclear fuel assembly of the present embodiment, by combining the preheating data of the nuclear fuel assembly under different historical operating conditions for training, under the guidance of the thermal uniformity data, the preheating model is prompted to predict the actual required temperature of each target temperature measuring point under actual operating conditions, and control the argon regulating valve, heater and fan of the preheating system to execute the preheating strategy, so that the target temperature measuring point reaches the actual required temperature, thereby ensuring uniform heating of all parts of the nuclear fuel assembly. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0017] Figure 1 Schematic diagram of the process of preheating a nuclear fuel assembly in an embodiment of the present application. DETAILED DESCRIPTION

[0018] The specific embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the description of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0019] In the description of the present invention, unless otherwise clearly specified and limited, the terms "set", "install", "connection" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium. For ordinary technicians in this field, the specific meanings of the above terms can be understood according to specific circumstances.

[0020] The directions or positional relationships indicated by terms such as “upper”, “lower”, “left”, “right”, “front”, “back”, “top”, “bottom”, “inside” and “outside” are based on the directions or positional relationships shown in the accompanying drawings, or are the directions or positional relationships in which the inventive product is usually placed when used. They are only for the convenience of description and simplified description, and do not indicate or imply that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore should not be understood as a limitation on the present invention.

[0021] The terms "first", "second", "third", etc. are merely used to distinguish elements of similar nature, and do not indicate or imply relative importance or a particular order.

[0022] The terms "comprises," "comprising," or any other variations thereof, are intended to cover a non-exclusive inclusion of the elements listed and may also include additional elements not expressly listed.

[0023] Preheating of nuclear fuel assemblies is an important operation link in the operation of high-temperature nuclear reactors or related tests. If the nuclear fuel assemblies are not preheated when entering the reactor, the various parts of the assembly may undergo uneven thermal expansion due to thermal shock, resulting in stress concentration or even damage. Preheating can make the temperature of various parts of the assembly rise slowly and more evenly, allowing the material to have enough time for thermal expansion, so that the internal stress can be released in advance and evenly distributed, avoiding damage to the assembly due to excessive stress in subsequent operation, affecting the normal operation of the nuclear reactor.

[0024] When preheating the nuclear fuel assembly, the nuclear fuel assembly is placed vertically in the hanging basket and arranged in a hexagon. Its overall structure is honeycomb-shaped. The middle hexagon is the connection mechanism of the hanging basket gripper, which does not hold the nuclear fuel assembly. Then, six nuclear fuel assemblies are evenly arranged in the first circle near the center; twelve nuclear fuel assemblies are evenly arranged in the second circle, and are staggered by half a body with the nuclear fuel assemblies in the first circle; eighteen nuclear fuel assemblies are evenly arranged in the third circle, and are staggered by half a body with the nuclear fuel assemblies in the second circle. It can be understood that the number and arrangement of the above nuclear fuel assemblies can be set according to actual needs.

[0025] The present embodiment provides a preheating system for a nuclear fuel assembly, comprising an argon regulating valve for controlling the flow rate of argon gas, a heater for heating argon gas, a fan for conducting the flow of argon gas, a first temperature sensor for monitoring the surface temperature data of a target temperature measuring point, a second temperature sensor for monitoring the ambient temperature of the nuclear fuel assembly, a camera for identifying the number of nuclear fuel assemblies, a pressure sensor for monitoring the argon pressure, and a control center for controlling the argon regulating valve, heater, fan, camera, first temperature sensor and second temperature sensor to execute a preheating strategy.

[0026] The gas is heated by a heater, and then the gas flow is accelerated by a blower, and the heated gas is quickly circulated in the container for heat exchange to achieve the function of heating. When cooling, the heater is turned off, and air at room temperature is introduced. The blower continuously blows air into the container to exhaust the hot air to achieve the function of cooling. In addition, the argon gas regulating valve includes a pressure relief valve, an air intake valve and an exhaust valve. During the heating process, when the internal pressure of the heating container used to load the nuclear fuel assembly rises above the required value, the pressure relief valve automatically opens to relieve the pressure; when the pressure is lower than the required value, the air intake valve opens to continuously rush in argon gas to increase the pressure. The preheating system of this embodiment can maintain the temperature in the heating container within the range of 200℃~250℃ and the pressure within the range of 5kpa~6kpa.

[0027] like Figure 1 As shown, based on the above-mentioned preheating system of nuclear fuel assembly, this embodiment further provides a preheating method of nuclear fuel assembly, the method comprising: Step S10: inputting the preheating data set of the nuclear fuel assembly under various historical working conditions into a preset preheating model for training, and obtaining the predicted temperature value of the target temperature measurement point on the nuclear fuel assembly within a preset thermal uniformity threshold range; Step S20: training the predicted temperature value according to a preset loss function to determine the loss value of the preheating model under different historical working conditions; and based on a preset update rule, updating the preheating model by the loss value to obtain a trained preheating model; Step S30: inputting the preheating data set of the nuclear fuel assembly under the current working condition into the trained preheating model to generate a preheating strategy, and controlling the preheating system to heat the argon gas in response to the preheating strategy.

[0028] In the preheating method of the nuclear fuel assembly of the present embodiment, by combining the preheating data of the nuclear fuel assembly under different historical operating conditions for training, under the guidance of the thermal uniformity data, the preheating model is prompted to predict the actual required temperature of each target temperature measuring point under the actual operating conditions, and the argon regulating valve, heater and fan of the preheating system are controlled to execute the preheating strategy, so that the target temperature measuring point reaches the actual required temperature, thereby ensuring uniform heating of all parts of the nuclear fuel assembly.

[0029] Step S10: inputting the preheating data set of the nuclear fuel assembly under various historical working conditions into a preset preheating model for training, and obtaining the predicted temperature value of the target temperature measurement point on the nuclear fuel assembly within a preset thermal uniformity threshold range.

[0030] In step S10, the new assembly of the sodium-cooled fast reactor generally includes a nuclear fuel assembly, a control assembly, a conversion zone assembly, a reflector assembly, and a shielding layer assembly, etc. The nuclear fuel assembly includes a fuel rod for loading nuclear fuel inside, a positioning grid for fixing the fuel rod, and an upper and lower tube seat assembly for supporting the fuel assembly; the control assembly includes a control rod and a transmission mechanism for driving the control rod to move, and the control rod is used to control the number of neutrons and adjust the rate of nuclear reaction by inserting or withdrawing the control rod from the core; the conversion zone assembly includes a depleted uranium rod for realizing the proliferation of the nuclear fuel assembly and a sleeve for accommodating the depleted uranium rod; the reflector assembly includes a reflective material for maintaining a chain reaction in the reactor and a structural frame for fixing the reflective material; the shielding layer assembly includes a neutron absorbing material for absorbing leaked neutrons and a protective shell for wrapping the neutron absorbing material.

[0031] When the preheating system preheats the nuclear fuel assembly, the temperature in the heating container is usually kept within the range of 200℃~250℃ and the pressure is usually kept within the range of 5kpa~6kpa. However, due to the changes in the structure, size and composition of the nuclear fuel assembly, it is easy to cause uneven heat transfer in the surface and deep layers of the nuclear fuel assembly corresponding to the target temperature measurement point. In addition, due to the large pressure deviation near multiple target temperature measurement points caused by the flow of argon gas, it is also easy to affect the heat transfer efficiency of argon gas to the nuclear fuel assembly near the target temperature measurement point, resulting in uneven heating of the nuclear fuel assembly.

[0032] The thermal uniformity data is used to measure the uniformity of the temperature distribution of the entire nuclear fuel assembly during the heat transfer from argon to the nuclear fuel assembly and the heat conduction from the surface to the deep layer of the nuclear fuel assembly. Generally, the closer the value of the thermal uniformity data is to 1, the more uniform the various parts of the nuclear fuel assembly are. In this embodiment, the value of the thermal uniformity threshold range is 0.95-1. It can be understood that the thermal uniformity threshold range can be set according to the actual reaction results of the nuclear fuel assembly under various historical conditions.

[0033] The preset data set includes at least surface temperature data of multiple target temperature measurement points under various historical working conditions, the Biot number of the target temperature measurement point, the opening of the argon regulating valve, the pressure of argon near the target temperature measurement point, the power of the heater, the number of nuclear fuel assemblies and the speed of the fan.

[0034] The surface temperature data of the target temperature measurement point usually refers to the surface temperature of the fuel rod, and the surface temperature data can be detected by an infrared temperature sensor. According to theories such as Planck's law, the infrared radiation intensity of an object is related to temperature. The infrared temperature sensor can measure the surface temperature data of the target temperature measurement point based on the infrared radiation energy emitted by the surface of the nuclear fuel assembly.

[0035] When manufacturing nuclear fuel assemblies, the nuclear fuel assemblies entering the same nuclear reactor have the same specifications and components. The Biot number is the ratio of the thermal resistance per unit thermal conductivity area inside the nuclear fuel assembly to the thermal resistance per unit area (i.e., external thermal resistance). Its expression is: , where B is the Biot number, is the surface heat transfer coefficient, h is the characteristic length, is the solid thermal conductivity. The solid thermal conductivity of each component can be determined by determining the outer surface composition and internal composition of the nuclear fuel assembly. and surface heat transfer coefficient , and then the characteristic length h of the nuclear fuel assembly is determined based on the specifications of the nuclear fuel assembly, so that the Biot number of the target temperature measurement point can be determined.

[0036] The argon regulating valve includes a pressure relief valve, an air inlet valve and an exhaust valve, etc. A flow meter can be set at the valve port of the argon regulating valve to determine the flow of argon gas and the opening of the argon regulating valve according to the flow formula. It can be understood that the opening of the valve body can be controlled by a PID controller.

[0037] A pressure sensor is also provided near the target temperature measurement point, and the pressure of the argon gas near the target temperature measurement point can be detected by the pressure sensor.

[0038] The preheating data set of the nuclear fuel assembly under various historical working conditions is input into a preset preheating model for training to obtain a predicted temperature value of a target temperature measurement point on the nuclear fuel assembly that meets a preset thermal uniformity threshold range, specifically including: Step S11: constructing a Markov joint probability distribution based on the relationship between the surface temperature data of the target temperature measurement point, the Biot number of the target temperature measurement point, the opening of the argon regulating valve, the pressure of the argon near the target temperature measurement point, the power of the heater, the number of nuclear fuel assemblies and the speed of the fan; Step S12: determining the predicted temperature value of each target temperature measurement point that satisfies the thermal uniformity threshold range based on the Markov joint probability distribution.

[0039] In this embodiment, by establishing a joint probability distribution among variables with poor correlation, such as the surface temperature data of the target temperature measurement point, the Biot number of the target temperature measurement point, the opening of the argon regulating valve, the pressure of argon near the target temperature measurement point, the power of the heater, the number of nuclear fuel assemblies, and the speed of the fan, these variables can be associated through a simplified Markov chain regardless of whether they are discrete or continuous, and the predicted temperature value of the target temperature measurement point within the corresponding thermal uniformity threshold range can be predicted.

[0040] In step S11, the Markov joint probability distribution is constructed based on the relationship between the surface temperature data of the target temperature measurement point, the Biot number of the target temperature measurement point, the opening of the argon regulating valve, the pressure of the argon near the target temperature measurement point, the power of the heater, the number of nuclear fuel assemblies and the speed of the fan, which specifically includes: Step S111: constructing an energy group according to the surface temperature data and the Biot number of two adjacent target temperature measurement points respectively, and constructing an energy group according to the surface temperature data of the target temperature measurement point, the opening of the argon regulating valve, the pressure of argon near the target temperature measurement point, the power of the heater, the number of nuclear fuel assemblies and the speed of the fan, and establishing an energy function based on the energy group; Step S112: generating a corresponding Boltzmann potential function according to the energy function, and determining a change parameter of the Boltzmann potential function based on maximum likelihood estimation; Step S113: generating the Markov joint probability distribution based on the variation parameter and the Boltzmann potential function.

[0041] In step S111, in the Markov random field, each variable such as surface temperature data, Biot number, opening of the argon control valve, pressure of argon near the target temperature measurement point, power of the heater, number of nuclear fuel assemblies and speed of the fan is regarded as a node. If there are n target temperature measurement points, there are n temperature nodes; assuming there are n argon control valves, there are n opening nodes; together with the Biot number node, heater power node and fan speed node, a total of the node set of the graph structure of the Markov random field is constituted.

[0042] The joint probability distribution between multiple variables can be decomposed into the product of multiple factors based on clusters, and each factor is only related to one energy cluster. There is a close interdependence between the variables in an energy cluster, and the correlation between these variables is characterized by the energy function defined on the energy cluster. The value of the energy function reflects a certain "preference" or "energy" state of the variables in the energy cluster. A high value means that a certain combination of variables in the energy cluster is more likely to occur, and a low value means the opposite.

[0043] The energy group established by the surface temperature data of two adjacent target temperature measurement points can characterize the relationship of heat transfer caused by thermal radiation from one target temperature measurement point to another target temperature measurement point; the energy group established by the Biot number of two adjacent target temperature measurement points can characterize the relationship of heat conduction from one target temperature measurement point to another target temperature measurement point in the nuclear fuel assembly; the energy group constructed based on the surface temperature data of the target temperature measurement point, the opening of the argon regulating valve, the pressure of argon near the target temperature measurement point, the power of the heater, the number of nuclear fuel assemblies and the speed of the fan can characterize the relationship between the heat radiation of the target temperature measurement point affected by the argon flow rate and the argon pressure.

[0044] In step S112, the Boltzmann function is expressed as: in, The energy function of the energy group constructed for the surface temperature data of two adjacent target temperature measurement points, The energy function of the energy group constructed by the Biot number of two adjacent target temperature measurement points, An energy function of an energy group constructed for the surface temperature data of a target temperature measuring point, the opening of an argon regulating valve, the pressure of argon near the target temperature measuring point, the power of a heater, the number of nuclear fuel assemblies, and the rotation speed of a fan; in, and are the surface temperature data of two adjacent target temperature measurement points, and are the Biot numbers of two adjacent target temperature measurement points, T is the surface temperature data, V is the opening, P is the pressure, W is the power, R is the speed, N is the number of nuclear fuel assemblies, is the target surface temperature data, Target opening, is the target pressure, is the target power, is the target speed, N t is the target number of nuclear fuel assemblies, is the average surface temperature, , , , , , and They are all weight parameters, which are used to measure the influence of each variable on the temperature difference between two adjacent target temperature measurement points.

[0045] Determining the change parameter of the Boltzmann potential function based on maximum likelihood estimation specifically includes: Step S1121: constructing the conditional probabilities of the surface temperature data of the target temperature measurement point, the Biot number of the target temperature measurement point, the opening of the argon regulating valve, the pressure of argon near the target temperature measurement point, the power of the heater, the number of nuclear fuel assemblies and the speed of the fan under the condition that other variables are fixed through pseudo-likelihood estimation, and determining the pseudo-approximate likelihood function according to the conditional probabilities; Step S1122: Calculate the gradient of the pseudo-approximate likelihood function for the corresponding variables respectively, and update the parameter value of the corresponding variable by the gradient ascent method until the pseudo-approximate likelihood function meets the preset iteration condition and the parameter value of the corresponding variable is used as the variation parameter of the Boltzmann potential function.

[0046] In this embodiment, the pseudo-likelihood estimation method is used to construct a pseudo-likelihood function and maximize it to estimate the parameters of the Markov random field model, thereby completing the modeling of the relationship between relevant variables in the argon preheating process.

[0047] In step S1121, first set the variable set X for the target temperature measurement point = {T n , B n , V n , P n , W n , R n}, where T n is a set of surface temperature data of all target temperature measurement points, Bn is the set of Biot numbers of all target temperature measurement points, V n is a set including the opening of the argon regulating valve, P n is the set of argon pressures near all target temperature measurement points, W n is the set of powers of all heaters, R n is a set of all fan speeds.

[0048] The surface temperature data variable X T For example, in other variables variable X \T ={ B n , V n , P n , W n , R n}Under fixed conditions, construct its conditional probability P(X T |X \T ;θ T ) obeys Gaussian distribution N(u(X \T ;θ T ), σ 2 (X \T ;θ T )), where u(X \T ;θ T ) and σ 2 (X \T ;θ T ) are other variables X \T and parameter θ T The nonlinear function of θ T is the set of parameter values ​​to be estimated.

[0049] Then the surface temperature data variable X T The pseudo-likelihood function In step S1122, the pseudo-likelihood function Taking the logarithm gives the logarithmic pseudo-likelihood function , the gradient can be calculated according to the gradient ascent method: in, is the parameter value of the surface temperature data to be updated.

[0050] As the gradient is updated, the gradient update can be terminated until the logarithmic pseudo-likelihood function converges or all the data in the preheating data set are iterated.

[0051] For the Biot number variable X B , opening variable X V , pressure variable X P , power variable X W and speed variable XR According to the surface temperature data variable X T Similarly, we can obtain the corresponding parameter values B , V , P , W ,and R .

[0052] After all iterations are completed, all updated parameter values , B , V , P , W , R The thermal uniformity of the surface temperature data of all target temperature measurement points is calculated, and the parameter value that meets the thermal uniformity threshold is selected to calculate the temperature of the target temperature measurement point as the predicted temperature value.

[0053] Step S20: training the predicted temperature value according to a preset loss function to determine the loss value of the preheating model under different historical working conditions; and based on a preset update rule, updating the preheating model through the loss value to obtain a trained preheating model.

[0054] The loss value includes the mean square error loss value and the mean absolute error loss value, and the expression of the loss function is: in, is the mean square error loss value, is the mean absolute error loss value, n is the number of target temperature measurement points, The surface temperature data of the target temperature measurement point in the preheating data set; To predict the temperature value.

[0055] Mean square error loss The smaller the value of is, the smaller the overall deviation between the predicted value of the preheating model and the true value is. The model has a higher accuracy and can accurately predict the surface temperature of the target temperature measurement point. It directly measures the mean absolute deviation between the predicted value and the true value. A smaller value of indicates that the model has a smaller prediction deviation in the average sense, which can more intuitively reflect the average prediction error of the model.

[0056] By calculating the mean square error loss value under different training rounds And the mean absolute error loss , and in the mean square error loss value And the mean absolute error loss When all meet the preset threshold, the round-robin training is stopped and the trained warm-up model is obtained.

[0057] Step S30: inputting the preheating data set of the nuclear fuel assembly under the current working condition into the trained preheating model to generate a preheating strategy, and controlling the preheating system to heat the argon gas in response to the preheating strategy.

[0058] In the preheating strategy, the appropriate opening of the argon regulating valve, the pressure of argon near the target temperature measurement point, the power of the heater, the number of nuclear fuel assemblies and the speed of the fan can be output to maintain the surface temperature data of the target temperature measurement point at a higher thermal uniformity.

[0059] In addition, due to production or assembly errors, the nuclear fuel assembly itself, or the container used to hold the nuclear fuel assembly is different when installing the nuclear fuel assembly, resulting in certain differences in the operating conditions each time the nuclear fuel assembly is preheated, which can easily lead to abnormal heating of some areas of the nuclear fuel assembly.

[0060] After the preheating data set of the nuclear fuel assembly under the current working condition is input into the trained preheating model to generate the preheating strategy, the method further includes: Step S40: matching reference thermal uniformity data corresponding to the current operating condition for each target temperature measurement point based on historical operating conditions; Step S50: comparing the uncertainty value between the thermal uniformity data of the preheating strategy of the target temperature measurement point and the reference thermal uniformity data; Step S60: marking the target temperature measurement point whose uncertainty value exceeds a preset uncertainty threshold as an abnormal target temperature measurement point; Step S70: extracting the preheating parameters of the abnormal target temperature measurement point in the preset data set under the historical working conditions corresponding to the reference thermal uniformity data, updating the preheating data set according to the preheating parameters, and regenerating a new preheating strategy.

[0061] In steps S40-S70, the temperature of similar historical working conditions is selected as the reference thermal uniformity data, and the target temperature measuring point with abnormal heating in some areas of the nuclear fuel assembly can be compared through uncertainty analysis as the abnormal target temperature measuring point, and the preheating parameters of the abnormal target temperature measuring point are corrected according to the historical working conditions. Under the premise of ensuring high thermal uniformity, the abnormal target temperature measuring point is prevented from overheating.

[0062] In step S50, the loss function Determine the uncertainty value between the thermal uniformity data of the preheating strategy and the reference thermal uniformity data, where L is the uncertainty value; X is the thermal uniformity data of the preheating strategy; and Y is the reference thermal uniformity data.

[0063] In step S70, updating the preheating data set according to the preheating parameters specifically includes: Step S701: randomly deleting data in the preheating data set under the current working condition by using a random forest algorithm, and establishing multiple groups of deleted data sets to be restored after deletion for the same data; Step S702: based on a pre-established recovery model, recover the deleted data set to obtain a recovery data set; Step S703: Calculate the similarity between the deleted data set and the restored data set, and update the preheating data set with the preheating parameters corresponding to the abnormal target temperature measurement points in the restored data set whose similarity meets the similarity threshold.

[0064] In steps S701-S703, the data of the preheating data set under the current working condition is randomly deleted by the random forest algorithm, and multiple groups of deleted data sets to be restored after deletion are established for the same data; then, the deleted data set is restored by the recovery model established by the diagnosed damage degree type to obtain a restored data set; and the restored data set is compared with the deleted data set.

[0065] When calculating the similarity between the deleted data set and the restored data set, the abnormal feature difference between the deleted data set and the restored data set is constructed based on the thermal uniformity data combined with the cosine similarity, where the value range of the cosine similarity is between [-1, 1]. The similarity between the deleted data set and the restored data set is calculated. When the cosine similarity between the deleted data set and the restored data set is closer to 1, they are more similar, indicating that the thermal uniformity is higher. However, if the cosine similarity between the deleted data set and the restored data set is close to -1, it means that the deleted data set and the restored data set are extremely dissimilar, and the preheating parameters of the abnormal target temperature measurement point provided by the restored data set are inaccurate. The deleted data set needs to be rebuilt according to the preheating parameters of the historical working conditions until the cosine similarity between the deleted data set and the restored data set is closer to 1.

[0066] Due to the influence of heat transfer, the temperature deviation between two adjacent target temperature measurement points will not be particularly large. When correcting the preheating parameters of historical working conditions, it is not directly replaced, but deleted and restored, which can avoid misjudgment of the model and thus improve the accuracy of model recognition.

[0067] In addition, when a nuclear fuel assembly is heated, the temperature of the outer layer is usually higher than that of the inner layer, and the nuclear fuel assembly has multiple layers from the inside to the outside, so that the outer layer of the nuclear fuel assembly has been heated to the target temperature, while the temperature of the inner layer of the nuclear fuel assembly is still relatively low. Therefore, when generating a preheating strategy, the nuclear fuel assembly is heated in stages, and after each heating to the target temperature of a stage, the nuclear fuel assembly is insulated, so that the inner layer of the nuclear fuel assembly is fully heated while maintaining the temperature of the outer layer of the nuclear fuel assembly.

[0068] This embodiment also establishes a thermal insulation model based on the Bayesian network, so that the nuclear fuel assembly can achieve higher thermal uniformity during the thermal insulation process.

[0069] The specific method of establishing the insulation model is as follows: Based on the multi-dimensional transient heat conduction equation, an objective function between the insulation temperature and the heater power is established, and the surface temperature data actually measured by the first temperature sensor is input into the objective function; then, a priori probability density function of the surface temperature data and the internal temperature data of the nuclear fuel assembly is established, and each parameter in the prior probability density function satisfies the normal distribution; the objective function is combined with the likelihood function of Bayesian reasoning to obtain the posterior distribution probability density function; then, the posterior probability density function is trained according to the preheating data set under historical working conditions to obtain the insulation model.

[0070] The multidimensional transient heat conduction equation is: Where T is the holding temperature, t is the holding time, is the heat diffusion efficiency, , k is the thermal conductivity of the nuclear fuel assembly, is the density of the nuclear fuel assembly, is the specific heat capacity, (x, y, z) are the coordinates of the nuclear fuel assembly at different positions, and Q is the heating intensity of the heating assembly.

[0071] By setting the target holding temperature to , the power of the heating component is P, and the objective function J can be obtained. The objective function J can be defined as: , where t1 and t2 are the start and end insulation time nodes within the preset time period, respectively, and is the heating power coefficient, and its value range is usually 0-1.

[0072] Assume that the internal temperature of the nuclear fuel is T i , the surface temperature data is T s . Then the prior probability density function is p(T i |T s ) has the form of normal distribution: Where n is the dimension of the surface temperature data, which is determined by the number and arrangement of nuclear fuel assemblies. is the mean vector, is the covariance vector, and Determined based on the preheating data set under historical operating conditions.

[0073] Therefore, the posterior probability density function is: .

[0074] Based on the same inventive concept as the above embodiment, this embodiment further provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions, wherein the instructions are used for a processor to load and execute the above-mentioned method for preheating a nuclear fuel assembly.

[0075] In the embodiments of the mobile terminal and computer-readable storage medium provided in the present application, all technical features of the above-mentioned control method embodiments are included, and the expansion and explanation content of the specification are basically the same as those of the above-mentioned method embodiments, which will not be repeated here.

[0076] The embodiment of the present application further provides a computer program product, which includes a computer program code. When the computer program code runs on a computer, the computer executes the methods in the above various possible implementation modes.

[0077] An embodiment of the present application also provides a chip, including a memory and a processor, the memory is used to store a computer program, and the processor is used to call and run the computer program from the memory, so that a device equipped with the chip executes the methods in various possible implementation modes as described above.

[0078] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0079] In the present application, the same or similar terminology concepts, technical solutions and / or application scenario descriptions are generally described in detail only the first time they appear. When they appear again later, they are generally not repeated for the sake of brevity. When understanding the technical solutions and other contents of the present application, for the same or similar terminology concepts, technical solutions and / or application scenario descriptions that are not described in detail later, reference can be made to the previous related detailed descriptions.

[0080] In the present application, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0081] The various technical features of the technical solution of the present application can be arbitrarily combined. In order to make the description concise, not all possible combinations of the various technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of the present application.

[0082] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product is stored in a storage medium as above, including a number of instructions for a terminal device to execute the method of each embodiment of the present application. The above is only a preferred embodiment of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made by using the content of the specification and drawings of this application, or directly or indirectly used in other related technical fields, is similarly included in the patent protection scope of the present application.

[0083] It should be noted that the various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0084] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the attached claims.

Claims

1. A method for preheating a nuclear fuel assembly, characterized in that: The method comprises: Inputting the preheating data set of the nuclear fuel assembly under various historical working conditions into a pre-set preheating model for training, and obtaining the predicted temperature value of the target temperature measurement point on the nuclear fuel assembly within a preset thermal uniformity threshold range; The predicted temperature value is trained according to a preset loss function to determine the loss value of the preheating model under different historical working conditions; and based on a preset update rule, the preheating model is updated by the loss value to obtain a trained preheating model; The preheating data set of the nuclear fuel assembly under the current working condition is input into the trained preheating model to generate a preheating strategy, and the preheating system is controlled to heat the argon gas in response to the preheating strategy.

2. The method for preheating a nuclear fuel assembly according to claim 1, characterized in that: The preheating data set at least includes surface temperature data of multiple target temperature measurement points under various historical working conditions, the Biot number of the target temperature measurement point, the opening of the argon regulating valve, the pressure of argon near the target temperature measurement point, the power of the heater, the number of nuclear fuel assemblies and the speed of the fan; the preheating data set of the nuclear fuel assembly under various historical working conditions is input into a pre-set preheating model for training to obtain the predicted temperature value of the target temperature measurement point on the nuclear fuel assembly within a preset thermal uniformity threshold range, specifically including: A Markov joint probability distribution is constructed based on the relationship between the surface temperature data of the target temperature measurement point, the Biot number of the target temperature measurement point, the opening of the argon regulating valve, the pressure of the argon near the target temperature measurement point, the power of the heater, the number of nuclear fuel assemblies and the speed of the fan; The predicted temperature value of each target temperature measurement point satisfying the thermal uniformity threshold range is determined based on the Markov joint probability distribution.

3. The method for preheating a nuclear fuel assembly according to claim 2, characterized in that: The Markov joint probability distribution is constructed based on the relationship between the surface temperature data of the target temperature measurement point, the Biot number of the target temperature measurement point, the opening of the argon regulating valve, the pressure of the argon near the target temperature measurement point, the power of the heater, the number of nuclear fuel assemblies and the speed of the fan, specifically including: constructing energy groups according to the surface temperature data and the Biot number of two adjacent target temperature measurement points respectively, and constructing energy groups according to the surface temperature data of the target temperature measurement point, the opening of the argon regulating valve, the pressure of argon near the target temperature measurement point, the power of the heater, the number of nuclear fuel assemblies and the rotation speed of the fan, and establishing an energy function based on the energy group; Generating a corresponding Boltzmann potential function according to the energy function, and determining a change parameter of the Boltzmann potential function based on maximum likelihood estimation; The Markov joint probability distribution is generated based on the variation parameter and the Boltzmann potential function.

4. The method for preheating a nuclear fuel assembly according to claim 3, characterized in that: The expression of the Boltzmann potential function is: in, The energy function of the energy group constructed for the surface temperature data of two adjacent target temperature measurement points, The energy function of the energy group constructed by the Biot number of two adjacent target temperature measurement points, An energy function of an energy group constructed for the surface temperature data of a target temperature measuring point, the opening of an argon regulating valve, the pressure of argon near the target temperature measuring point, the power of a heater, the number of nuclear fuel assemblies, and the rotation speed of a fan; in, and are the surface temperature data of two adjacent target temperature measurement points, and are the Biot numbers of two adjacent target temperature measurement points, T is the surface temperature data, V is the opening, P is the pressure, W is the power, R is the speed, N is the number of nuclear fuel assemblies, is the target surface temperature data, Target opening, is the target pressure, is the target power, is the target speed, N t is the target number of nuclear fuel assemblies, is the average surface temperature, , , , , , and are all weight parameters.

5. The method for preheating a nuclear fuel assembly according to claim 3, characterized in that: The determining of the change parameter of the Boltzmann potential function based on maximum likelihood estimation specifically includes: The conditional probabilities of the surface temperature data of the target temperature measuring point, the Biot number of the target temperature measuring point, the opening of the argon regulating valve, the pressure of the argon near the target temperature measuring point, the power of the heater, the number of nuclear fuel assemblies and the speed of the fan under the condition that other variables are fixed are constructed by pseudo-likelihood estimation, and the pseudo-approximate likelihood function is determined according to the conditional probabilities; The gradients of the pseudo-approximate likelihood function for the corresponding variables are calculated respectively, and the parameter values ​​of the corresponding variables are updated by the gradient ascent method until the pseudo-approximate likelihood function satisfies the preset iteration conditions, and the parameter values ​​of the corresponding variables are used as the variation parameters of the Boltzmann potential function.

6. The method for preheating a nuclear fuel assembly according to claim 1, characterized in that: The loss value includes a mean square error loss value and a mean absolute error loss value, and the expression of the loss function is: in, is the mean square error loss value, is the mean absolute error loss value, n is the number of target temperature measurement points, The surface temperature data of the target temperature measurement point in the preheating data set; To predict the temperature value.

7. The method for preheating a nuclear fuel assembly according to claim 1, characterized in that: After the preheating data set of the nuclear fuel assembly under the current working condition is input into the trained preheating model to generate the preheating strategy, the method further includes: Matching reference thermal uniformity data corresponding to the current operating condition for each target temperature measurement point based on historical operating conditions; Comparing the uncertainty value between the thermal uniformity data of the preheating strategy of the target temperature measurement point and the reference thermal uniformity data; Marking the target temperature measurement point where the uncertainty value exceeds a preset uncertainty threshold as an abnormal target temperature measurement point; Extract the preheating parameters of the abnormal target temperature measurement point in the preheating data set under the historical working conditions corresponding to the reference thermal uniformity data, update the preheating data set according to the preheating parameters, and regenerate a new preheating strategy.

8. The method for preheating a nuclear fuel assembly according to claim 7, characterized in that: The updating of the preheating data set according to the preheating parameters specifically includes: Randomly delete the data in the preheating data set under the current working condition by using a random forest algorithm, and establish multiple groups of deleted data sets to be restored after deletion for the same data; Based on a pre-established recovery model, recover the deleted data set to obtain a recovery data set; The similarity between the deleted data set and the restored data set is calculated, and the preheating data set is updated with the preheating parameters corresponding to the abnormal target temperature measurement points in the restored data set whose similarity meets the similarity threshold.

9. A preheating system for a nuclear fuel assembly, characterized in that: It includes an argon regulating valve for controlling the flow rate of argon, a heater for heating argon, a fan for conducting the flow of argon, a first temperature sensor for monitoring the surface temperature data of a target temperature measuring point, a second temperature sensor for monitoring the ambient temperature of a nuclear fuel assembly, a camera for identifying the number of nuclear fuel assemblies, and a control center for controlling the argon regulating valve, the heater, the fan, the camera, the first temperature sensor and the second temperature sensor to execute a preheating strategy, wherein the control center is used to execute the preheating method of a nuclear fuel assembly as described in any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, and the instructions are used by a processor to load and execute the preheating method of a nuclear fuel assembly as described in any one of claims 1-8.

Images