CN116029409A

CN116029409A - Nuclear reactor multiparameter optimizing method, device, computer equipment and storage medium

Info

Publication number: CN116029409A
Application number: CN202211149834.XA
Authority: CN
Inventors: 李亮国; 孟祥飞; 南宗宝; 余健明; 刘继墉; 卢冬华; 邢军; 苏前华; 吴小航; 柳红超
Original assignee: China General Nuclear Power Corp; China Nuclear Power Technology Research Institute Co Ltd; China Nuclear Power Engineering Co Ltd; CGN Power Co Ltd
Current assignee: China General Nuclear Power Corp; China Nuclear Power Technology Research Institute Co Ltd; China Nuclear Power Engineering Co Ltd; CGN Power Co Ltd
Priority date: 2022-09-21
Filing date: 2022-09-21
Publication date: 2023-04-28
Also published as: WO2024060480A1

Abstract

The present application relates to a nuclear reactor multiparameter optimization method, apparatus, computer device, storage medium and computer program product. The method comprises the following steps: acquiring each parameter to be optimized of a target optimization object in a nuclear reactor and an initial design scheme of the nuclear reactor; generating an optimization function of a target optimization object according to the initial design scheme and each parameter to be optimized; calling an optimization parameter model to solve an optimization function to obtain an optimization value of each parameter to be optimized, and constructing the optimization parameter model in advance according to a self-adaptive moment estimation algorithm; judging whether the optimization function is converged based on the optimization value of each parameter to be optimized, and if so, determining the optimization value of each parameter to be optimized as a target optimization value of each parameter to be optimized; and updating the initial design scheme according to the target optimization value of each parameter to be optimized. By adopting the method, the efficiency and the accuracy of optimizing the parameters and the design scheme of the nuclear reactor can be effectively improved.

Description

Nuclear reactor multiparameter optimizing method, device, computer equipment and storage medium

Technical Field

The present application relates to the field of big data technology, and in particular, to a method, an apparatus, a computer device, a storage medium, and a computer program product for optimizing multiple parameters of a nuclear reactor.

Background

With the development of nuclear power industry, in order to meet the demands of the fields of space nuclear power, mobile nuclear power stations, sea water desalination and the like, a new generation of reactor with simple structure, small volume, light weight, long service life and good inherent safety is developed in an international competitive manner.

In the conventional technology, the design of the reactor usually depends on experience accumulation and expert judgment to determine relevant design parameters, so that the optimization of relevant design schemes of the reactor is realized, the determination period of the design parameters is longer, the design efficiency of the reactor is low, and the design parameters obtained by manual judgment are easy to have errors, so that the safety and other performances of the designed reactor cannot be ensured.

Disclosure of Invention

In view of the foregoing, there is a need for a nuclear reactor multi-parameter optimization method, apparatus, computer device, computer readable storage medium, and computer program product that can improve the efficiency and accuracy of nuclear reactor parameter optimization.

In a first aspect, the present application provides a method of multi-parameter optimization of a nuclear reactor, the method comprising:

acquiring each parameter to be optimized of a target optimization object in a nuclear reactor and an initial design scheme of the nuclear reactor;

Generating an optimization function of the target optimization object according to the initial design scheme and each parameter to be optimized;

calling an optimization parameter model to solve the optimization function to obtain an optimization value of each parameter to be optimized, wherein the optimization parameter model is pre-constructed according to a self-adaptive moment estimation algorithm;

judging whether the optimization function is converged based on the optimization value of each parameter to be optimized, and if so, determining that the optimization value of each parameter to be optimized is the target optimization value of each parameter to be optimized;

and updating the initial design scheme according to the target optimization value of each parameter to be optimized.

In one embodiment, the generating an optimization function according to the initial design solution and each parameter to be optimized includes:

determining an optimization target of the target optimization object according to the initial design scheme;

determining the optimization weight of each parameter to be optimized based on the optimization target;

and generating an optimization function according to each parameter to be optimized and the optimization weight of each parameter to be optimized.

In one embodiment, the calling the optimization parameter model to solve the optimization function to obtain an optimized value of each parameter to be optimized includes:

Determining the standard value of each parameter to be optimized according to the initial design scheme;

performing iterative updating calculation by taking each standard value as an input value of the adaptive moment estimation algorithm to obtain an iterative updating value of each standard value;

and if the iteration update values meet the update end conditions, determining the optimized values of the parameters to be optimized according to the iteration update values.

In one embodiment, the determining the optimized value of each parameter to be optimized according to each iteration updated value includes:

determining constraint conditions of the parameters to be optimized according to the initial design scheme;

comparing each iteration update value with the constraint condition of the corresponding parameter to be optimized, and if each iteration update value meets the constraint condition of the corresponding parameter to be optimized, determining the optimized value of each parameter to be optimized according to each iteration update value and the standard value of each parameter to be optimized.

In one embodiment, the determining, according to the initial design, the standard value of each parameter to be optimized includes:

determining initial values of the parameters to be optimized according to the initial design scheme;

and carrying out normalization processing on each initial value to obtain a standard value of each parameter to be optimized corresponding to each initial value.

In one embodiment, the determining whether the optimization function converges based on the optimization value of each parameter to be optimized further includes:

obtaining basic parameters of the target optimization object, wherein the basic parameters comprise parameters to be optimized;

modeling the target optimization object according to the basic parameters to obtain an analysis model of the target optimization object;

inputting the optimized value of each parameter to be optimized into the analysis model to obtain the optimized value of each basic parameter, and recording the optimized value of each basic parameter.

In a second aspect, the present application also provides a nuclear reactor multiparameter optimization device, the device comprising:

the parameter acquisition module is used for acquiring each parameter to be optimized of a target optimization object in the nuclear reactor and an initial design scheme of the nuclear reactor;

the function generating module is used for generating an optimization function of the target optimization object according to the initial design scheme and each parameter to be optimized;

the function solving module is used for calling an optimization parameter model to solve an optimal solution for the optimization function to obtain an optimized value of each parameter to be optimized, and the optimization parameter model is pre-constructed according to a self-adaptive moment estimation algorithm;

The parameter optimization module is used for judging whether the optimization function is converged based on the optimization value of each parameter to be optimized, and if so, determining that the optimization value of each parameter to be optimized is the target optimization value of each parameter to be optimized;

and the scheme updating module is used for updating the initial design scheme according to the target optimization value of the parameter to be optimized.

In a third aspect, the present application also provides a computer device comprising a memory storing a computer program and a processor implementing the steps of the above method when the processor executes the computer program.

In a fourth aspect, the present application also provides a computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of the above-described method.

In a fifth aspect, the present application also provides a computer program product comprising a computer program which, when executed by a processor, implements the steps of the method described above.

According to the multi-parameter optimizing method, the device, the computer equipment, the storage medium and the computer program product of the nuclear reactor, all the parameters to be optimized of the target optimizing object in the nuclear reactor and the initial design scheme of the nuclear reactor are obtained, the optimizing function of the target optimizing object is generated according to the initial design scheme and all the parameters to be optimized, the optimizing function model which is constructed in advance according to the self-adaptive moment estimation algorithm is called to solve the optimizing function, the speed of obtaining the optimizing value is greatly improved, the value obtained by solving the optimizing function can be regarded as the parameter value after optimization, whether the optimizing function is converged or not is judged based on the optimizing value of all the parameters to be optimized, when the obtained optimizing value can enable the optimizing function to be converged, the optimizing value of the optimizing function to be the optimal solution of the optimizing function is obtained, the optimizing value of the optimizing function to be determined to be the target optimizing value of the parameters to be optimized, the initial design scheme is updated according to the target optimizing value of the parameters to be optimized, and the efficiency and the accuracy of optimizing the parameters and the design scheme are effectively improved.

Drawings

FIG. 1 is a diagram of an application environment for a nuclear reactor multiparameter optimization method in one embodiment;

FIG. 2 is a flow diagram of a method of multi-parameter optimization of a nuclear reactor in one embodiment;

FIG. 3 is a flowchart illustrating steps for generating an optimization function according to an initial design and parameters to be optimized according to an embodiment;

FIG. 4 is a schematic flow chart of a step of calling an optimization parameter model to solve an optimization function to obtain an optimized value of each parameter to be optimized in one embodiment;

FIG. 5 is a flow chart of a method of optimizing multiple parameters of a nuclear reactor in another embodiment;

FIG. 6 is a flow chart of a method of optimizing multiple parameters of a nuclear reactor in another embodiment;

FIG. 7 is a flow chart of a method of optimizing multiple parameters of a nuclear reactor in another embodiment;

FIG. 8 is a block diagram of a nuclear reactor multi-parameter optimization apparatus in one embodiment;

fig. 9 is an internal structural diagram of a computer device in one embodiment.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application will be further described in detail with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the present application.

The nuclear reactor multi-parameter optimization method provided by the embodiment of the application can be applied to an application environment shown in fig. 1. Wherein the parameter optimization platform 102 communicates with the user terminal 104. The data storage system may store data that the parameter optimization platform 102 needs to process. The data storage system may be integrated on the parameter optimization platform 102 or may be located on the cloud or other network server. The parameter optimization platform obtains all to-be-optimized parameters of a target optimization object in the nuclear reactor and an initial design scheme of the nuclear reactor, generates an optimization function of the target optimization object according to the initial design scheme and all to-be-optimized parameters, calls an optimization parameter model to solve the optimization function to obtain an optimization value of each to-be-optimized parameter, the optimization parameter model is pre-constructed according to a self-adaptive moment estimation algorithm, judges whether the optimization function is converged based on the optimization value of each to-be-optimized parameter, and if so, determines that the optimization value of each to-be-optimized parameter is the target optimization value of each to-be-optimized parameter, and updates the initial design scheme according to the target optimization value of each to-be-optimized parameter. The user terminal 104 may be, but not limited to, various personal computers, notebook computers, smart phones, tablet computers, internet of things devices and portable wearable devices, where the internet of things devices may be smart speakers, smart televisions, smart air conditioners, smart vehicle devices, and the like. The portable wearable device may be a smart watch, smart bracelet, headset, or the like. The parameter optimization platform 102 may be implemented as a stand-alone server or as a cluster of servers.

In one embodiment, the parameter optimization platform can also be integrated on the user terminal, and the user starts the parameter optimization platform to communicate with the parameter optimization platform by triggering the user terminal parameter optimization platform to start operation.

In one embodiment, as shown in fig. 2, a method for optimizing multiple parameters of a nuclear reactor is provided, and the method is applied to the parameter optimization platform in fig. 1 for illustration, and includes the following steps:

step 202, obtaining each parameter to be optimized of a target optimization object in a nuclear reactor and an initial design scheme of the nuclear reactor.

The target optimization object is equipment to be optimized in equipment constituting a nuclear reactor, such as an inverted U-shaped natural circulation steam generator in the nuclear reactor, a driving steam turbine engine and the like. The parameters to be optimized of the target optimization object are extracted from structural parameters and thermodynamic parameters of the target optimization object, and taking the target optimization object as an inverted U-shaped natural circulation steam generator as an example, the structural parameters of the inverted U-shaped natural circulation steam generator include, but are not limited to, steam generator height, heat transfer tube outer diameter, heat transfer tube number and the like, and the thermodynamic parameters include, but are not limited to, steam generator primary side flow rate, steam generator primary side inlet outlet temperature, steam generator secondary side flow rate, steam generator secondary side inlet outlet temperature and the like. The parameters to be optimized of the steam generator can be the volume (V) of the steam generator, the temperature (T) of the primary side outlet, the maximum flow velocity (V) of the primary side outlet ₁ ) Maximum secondary flow rate (v) ₂ ) Maximum heat flux density (Q), etc.

The initial design scheme is designed by a designer according to experience data and historical use data of the nuclear reactor, and can reflect initial values of basic parameters of devices forming the nuclear reactor, a value range of the basic parameters, coupling relations among the basic parameters, optimization targets of the devices and the like.

Specifically, the parameter optimization platform acquires each parameter to be optimized of a target optimization object in the nuclear reactor sent by a user based on a user terminal and an initial design scheme of the nuclear reactor.

And 204, generating an optimization function of the target optimization object according to the initial design scheme and each parameter to be optimized.

The optimization function is a function for optimizing each parameter to be optimized in the target optimization object, and the optimization function is generated based on an optimization method. It can be understood that the optimization function is an objective function composed of a given number of variables, i.e. parameters to be optimized, and the value of the variable is called an optimal solution by making the objective function value of the variable take a certain set of values to be the largest or smallest.

Specifically, since the initial design scheme includes basic information of each parameter to be optimized of the target optimization object, the parameter optimization platform can generate an optimization function of the target optimization object together with each parameter to be optimized according to the initial design scheme of the nuclear reactor. It can be understood that the optimization function may be an objective function taking the maximum value maxf (x) or an objective function taking the minimum value minf (x), depending on the actual situation.

And 206, calling an optimization parameter model to solve the optimization function to obtain the optimization value of each parameter to be optimized, wherein the optimization parameter model is pre-constructed according to the self-adaptive moment estimation algorithm.

The optimization parameter model is a model for solving an optimization function, and it can be understood that the parameter optimization model is pre-constructed according to an adaptive moment estimation algorithm and is configured on a parameter optimization platform.

The adaptive moment estimation algorithm (adaptive moment estimation, adam) is a random objective function optimization algorithm based on a first-order gradient, the Adam algorithm can be intuitively understood as the sum of an RMSprop algorithm and a Momentum algorithm, combines the advantages of the two algorithms, has high Adam calculation efficiency and low storage capacity requirement, does not change the diagonal rescaling of the gradient, is very suitable for a model with larger problems in data/parameters and complex multidimensional, and has faster convergence speed.

In particular, since the optimization function is generated based on a plurality of parameters to be optimized of the target optimization object, the solution of the optimization function is a critical and complex process. The parameter optimization platform is packaged with a model which is built in advance according to the self-adaptive moment estimation algorithm and can solve various optimization functions, and after the optimization function of the target optimization object is generated, the parameter optimization platform calls a preset optimization parameter model to solve the optimization function, so that the optimization value of each parameter to be optimized is obtained.

And step 208, judging whether the optimization function is converged based on the optimization value of each parameter to be optimized, and if so, determining the optimization value of each parameter to be optimized as the target optimization value of each parameter to be optimized.

Whether the optimized value of each parameter to be optimized obtained by solving the optimized function can enable the optimized function to converge is a key condition for judging whether the optimized value of each parameter to be optimized is an optimal solution of the optimized function.

If the optimized values of the parameters to be optimized can enable the optimized function to be converged, the optimized values of the parameters to be optimized are the optimal solution of the optimized function, and the initial design scheme of the nuclear reactor can be updated according to the optimized values of the parameters to be optimized. If the optimized value of each parameter to be optimized cannot enable the optimized function to converge, it is indicated that the optimized value of each parameter to be optimized is not the optimal solution of the optimized function, and if the initial design scheme is updated directly according to the optimized value of each parameter to be optimized at this time, the parameter values of each parameter in the updated scheme may be inaccurate.

Specifically, the parameter optimization platform judges whether the optimization function is converged based on the optimization value of each parameter to be optimized, and when the optimization value of each parameter to be optimized is determined to enable the optimization function to be converged, the optimization value of each current parameter to be optimized is determined to be the target optimization value of each parameter to be optimized.

Step 210, updating the initial design scheme according to the target optimization value of each parameter to be optimized.

Specifically, the parameter optimization platform updates the parameter values of the basic parameters corresponding to the target optimization objects in the initial design scheme according to the target optimization values of the parameters to be optimized to obtain an updated optimal design scheme of the nuclear reactor,

according to the multi-parameter optimizing method for the nuclear reactor, all parameters to be optimized of a target optimizing object in the nuclear reactor and an initial design scheme of the nuclear reactor are obtained, an optimizing function of the target optimizing object is generated according to the initial design scheme and all the parameters to be optimized, an optimizing function model which is built in advance according to a self-adaptive moment estimating algorithm is called to solve the optimizing function, the speed of obtaining an optimizing value is greatly improved, the optimizing function is generated according to the initial design scheme and all the parameters to be optimized, therefore, the value obtained by solving the optimizing function can be regarded as a parameter value after optimization, whether the optimizing function is converged or not is judged based on the optimizing value of all the parameters to be optimized, when the obtained optimizing value can enable the optimizing function to be converged, the optimizing value obtained at the moment is regarded as an optimal solution of the optimizing function, the optimizing value of the optimizing function is determined as a target optimizing value of the parameters to be optimized, the initial design scheme is updated according to the target optimizing value of the parameters to be optimized, and the efficiency and the optimizing accuracy of the parameters and the design scheme of the nuclear reactor are effectively improved.

In one embodiment, as shown in fig. 3, generating an optimization function according to the initial design scheme and each parameter to be optimized includes:

step 302, determining an optimization target of the target optimization object according to the initial design scheme.

The optimization targets of the target optimization objects are designed according to the setting scenes of the nuclear reactor, for example, in the setting scenes with higher performance compactness, the optimization targets can be arranged in a high-performance compactness mode. It can be appreciated that when designing an initial design of a nuclear reactor, a designer binds a target optimization object with its corresponding optimization target and records the target optimization object in the initial design.

Specifically, the parameter optimization platform searches an optimization target corresponding to a target optimization object from an initial design scheme according to the target optimization object.

Step 304, determining the optimization weight of each parameter to be optimized based on the optimization objective.

Specifically, the optimization objective of the objective optimization objective may determine the optimization weight of each parameter to be optimized, for example, if the optimization objective is a high-performance compact arrangement, a higher weight needs to be assigned to the parameters such as volume, and the objective optimization objective is taken as an inverted U-shaped natural circulation steam generator of the nuclear reactor as an example to optimize the parameters The number is the steam generator volume (V), the primary side outlet temperature (T), the maximum flow rate (V2) of the primary side outlet, the maximum flow rate (V2) of the secondary side and the maximum heat flux density (Q). For the convenience of calculation, determining the basic model of the optimization function as minf (x), wherein the function expression of the optimization function is as follows: minf (x) =min (W ₁ V+W ₂ T+W ₃ ν ₁ +W ₄ ν ₂ +W ₅ Q), wherein Wn represents the specific gravity of each parameter to be optimized. And further, normalization processing of each parameter to be optimized and f (x) is realized.

If the optimization target of the steam generator is a high-performance compact arrangement steam generator, determining the optimization weights of the parameters to be optimized according to the optimization target as follows: w (W) ₁ ＝0.5，W ₂ ＝0.2，W ₃ ＝0.1，W ₄ ＝0.1，W ₅ ＝0.1。

And 306, generating an optimization function according to each parameter to be optimized and the optimization weight of each parameter to be optimized.

Specifically, after determining the optimization weights corresponding to the parameters to be optimized, the parameter optimization platform generates an optimization function according to the parameters to be optimized and the optimization weights corresponding to the parameters to be optimized.

For example, when the parameter optimization platform determines and further obtains the optimization weights of the parameters to be optimized, the optimization weights are respectively: w (W) ₁ ＝0.5，W ₂ ＝0.2，W ₃ ＝0.1，W ₄ ＝0.1，W ₅ =0.1, generating a final expression of the optimization function according to each parameter to be optimized and the optimization weight corresponding to each parameter to be optimized, wherein the final expression is: minf (x) =min (0.5v+0.2t+0.1v) ₁ +0.1ν ₂ +0.1Q)。

In this embodiment, the parameter optimization platform determines an optimization target of the target optimization object according to the initial design scheme, determines an optimization weight of each parameter to be optimized based on the optimization target, generates an optimization function according to each parameter to be optimized and the optimization weight of each parameter to be optimized, and determines the optimization weight of each parameter to be optimized through the optimization target, so that the generated optimization function can fully consider the important proportion of each parameter to be optimized in the optimization target, and further improves the accuracy of nuclear reactor parameter optimization.

In one embodiment, as shown in fig. 4, the step of calling the optimization parameter model to solve the optimization function to obtain the optimized value of each parameter to be optimized includes:

and step 402, determining the standard value of each parameter to be optimized according to the initial design scheme.

In one embodiment, determining the standard value of each parameter to be optimized according to the initial design scheme includes: determining initial values of the parameters to be optimized according to the initial design scheme; and carrying out normalization processing on each initial value to obtain a standard value of each parameter to be optimized corresponding to each initial value.

Specifically, the initial design scheme of the nuclear reactor includes initial values of parameters to be optimized of a target optimization object, and the initial values are normalized, namely dimensionless values of the initial values are calculated, so that standard values corresponding to the initial values of the parameters to be optimized are obtained. Non-dimensionality (or dimensionality) refers to the removal of some or all units of an equation related to a physical quantity by a suitable variable substitution, so as to simplify the experimental or computational purposes, and is an important processing idea in scientific research. The initial values of the parameters to be optimized are normalized, so that the values of the optimization functions under the array formed by the values of the parameters to be optimized can be intuitively obtained.

For example, the initial value of the flow parameter is 250m ³ And/h, dimensionless the initial value to obtain a standard value of 1, wherein the standard value of 1 corresponds to 250m ³ /h。

And step 404, performing iterative update calculation by taking each standard value as an input value of the adaptive moment estimation algorithm to obtain an iterative update value of each standard value.

In the adaptive moment estimation algorithm, each standard of each parameter to be optimized in the optimization function is used as a group of input parameters, and the input parameters are multidimensional vectors.

Specifically, first, setting initialization parameters in the adaptive moment estimation algorithm, wherein the initialization parameters include a step value delta, and initial parameters of each parameter to be optimizedNumber theta _i The initial parameters are the standard values corresponding to the parameters to be optimized obtained in the step 402, and the numerical stability epsilon and the first-order momentum attenuation coefficient beta ₁ Second order momentum decay coefficient beta ₂ For intermediate variables involved in the calculation: the first-order momentum s and the second-order momentum r are initialized to 0. It can be understood that the numerical stabilization amount, the first-order momentum attenuation coefficient and the second-order momentum attenuation coefficient can all use preset values of the adaptive moment estimation algorithm, and can also be set by a designer according to actual conditions.

In one embodiment, the numerical stability is 10 ^-8 The first order momentum decay coefficient takes 0.9 and the second order momentum decay coefficient takes 0.999.

Second, the gradient of the objective function f (x) of the optimization function with respect to x is a vector consisting of partial derivatives:

each bias element in the gradient

Representing f (x) at x _i Is a rate of change of (c).

For any function f (x) if

Is the local minimum of f (x) and f (x) is +.>

There is a slight sensation in the region, there is a need for a +.>

When in the gradient direction

The direction derivative f (x) is minimized. Therefore, the gradient descent algorithm can be continuously reducedThe value of the objective function f (x):

when in the gradient direction

The direction derivative f (x) is minimized. Therefore, the value of the objective function f (x) can be continuously reduced by the following gradient descent algorithm:

determining a gradient of a loss function of the current position for x _i The gradient expression is as follows:

multiplying the step length by the gradient of the loss function to obtain the falling distance of the current position, namely:

the adaptive moment estimation algorithm not only calculates an estimate v of the second moment of the square of the exponentially decaying gradient _t It also calculates an estimate m of the first moment of the exponential decay gradient _t Similar to momentum.

m _t ＝β ₁ m _t-1 +(1-β ₁ )g _t

Wherein m is _t Initializing to 0 vector for estimation of first moment (mean); v _t For the estimation of the second moment (variance), the 0 vector is initialized. When the attenuation ratio is small (beta ₁ And beta ₂ Close to 1), m _t And v _t Eventually tending to 0. Thus, the adaptive moment estimation algorithm eventually corrects the first and second momentsPositive (. Beta.) ₁ And beta ₂ Superscript t of (c) stands for power t):

finally, the updating rule of the adaptive moment estimation algorithm is as follows:

and step 406, if each iteration update value meets the update end condition, determining an optimized value of each parameter to be optimized according to each iteration update value.

And if the gradient descending distance of each iteration updating value is smaller than the numerical value stabilizing amount, determining the iteration updating value as an optimized value of each parameter to be optimized. If the gradient descent distance of each iteration update value is larger than the value stabilizing amount, returning to the iteration step again to continue the iteration update until each iteration update value meets the update ending condition.

Specifically, it is determined whether or not all the calculation results θ _i The gradient descent distances of (2) are all less than epsilon, and if they are less than epsilon, the algorithm is terminated, based on all theta present _i (i=0, 1, n) determining the optimal value of each parameter to be optimized. If theta is _i If the gradient of (a) decreases by a distance not less than ε, then return to calculating the estimate v of the second moment that calculates the square of the exponentially decaying gradient _t Estimation m of first moment from exponential decay gradient _t Is carried out by a method comprising the steps of.

In the above embodiment, the standard value of each parameter to be optimized is obtained according to the initial value of each parameter to be optimized, and the standard value is used as a part of the initialized parameter in the adaptive moment estimation algorithm to be input into the parameter optimization model for iterative computation, and when each iterative update value meets the update end condition, the optimized value of each parameter to be optimized is obtained according to each iterative update value, so that a data basis is provided for subsequently obtaining the target optimized value of each parameter to be optimized.

In one embodiment, determining the optimized value for each parameter to be optimized from each iteratively updated value comprises: determining constraint conditions of each parameter to be optimized according to an initial design scheme; comparing each iteration update value with the constraint condition of the corresponding parameter to be optimized, and if each iteration update value meets the constraint condition of the corresponding parameter to be optimized, determining the optimized value of each parameter to be optimized according to each iteration update value and the standard value of each parameter to be optimized.

The constraint condition of each parameter to be optimized is a constraint condition of the value of each parameter to be optimized, and can be a constraint condition of the value range of each parameter to be optimized, and it can be understood that the constraint condition of each parameter to be optimized is determined by a designer according to the use condition of a target optimization object corresponding to each parameter to be optimized in the actual use process, and is recorded in an initial design scheme after being bound with each parameter to be optimized in advance, and the constraint condition can be represented by an equation or an inequality.

In the general concept of the optimization method, the optimization function is f (x), each parameter to be optimized is all variables in the optimization function, if the variables have constraint conditions, the constraint conditions comprise all equations and inequalities related to variable values, a feasible domain is a region surrounded by the constraint conditions in space, each point in the feasible domain is a feasible point of the original problem, and the optimal solution is a feasible solution capable of enabling the optimization function to be maximum or minimum.

Specifically, the parameter optimization platform determines constraint conditions of each parameter to be optimized from an initial design scheme according to each parameter to be optimized, compares each iteration update value with the constraint conditions of the corresponding parameter to be optimized, if an update value which cannot meet the constraint conditions of the corresponding parameter to be optimized exists in each iteration update value, the iteration update value obtained by the method cannot be used, if the safety of a nuclear reactor is easy to risk due to strong use, and the iteration update value of each parameter to be optimized needs to be readjusted according to gradient and step reverse optimization.

And if each iteration update value meets the constraint condition of the corresponding parameter to be optimized, determining the optimized value of each parameter to be optimized according to each iteration update value and the standard value of each parameter to be optimized. For example, the standard value of the flow optimization parameter is 1 to 250m ³ And (h) obtaining an iteration update value of 1.2 after iteration update calculation, wherein the optimized value of the flow optimization parameter is 1.2 multiplied by 250m ³ /h＝300m ³ /h。

In this embodiment, after obtaining the iteration update value of each parameter to be optimized, the parameter optimization platform compares the iteration update value of each parameter to be optimized with the constraint condition corresponding to each parameter to be optimized, so as to determine whether the iteration update value can be directly used as the optimization value of each parameter to be optimized, thereby further ensuring the safety and accuracy of the optimization value of each parameter to be optimized.

Because the basic parameters of the nuclear reactor prototype have coupling relations with each other, when one parameter changes, other parameters also need to be correspondingly changed due to the coupling relations. In one embodiment, as shown in fig. 5, determining whether the optimization function converges based on the optimization value of each parameter to be optimized further includes:

step 502, obtaining basic parameters of a target optimization object, wherein the basic parameters comprise parameters to be optimized.

Specifically, the parameter optimization platform acquires basic parameters of the target optimization object from the initial design scheme, wherein the basic parameters are structural parameters and thermodynamic parameters of the target optimization object, and the basic parameters comprise parameters to be optimized which need to be optimized.

And step 504, modeling the target optimization object according to the basic parameters to obtain an analysis model of the target optimization object.

Specifically, the parameter optimization platform models the target optimization object according to the obtained basic parameters to obtain an analysis model of the target optimization object.

In one embodiment, the parameter optimization platform substitutes RELAP, CATHARE and other reactor system safety analysis programs model basic parameters of the target optimization object to obtain an analysis model of the target optimization object.

And step 506, inputting the optimized values of the parameters to be optimized into the analysis model to obtain the optimized values of the basic parameters, and recording the optimized values of the basic parameters.

Specifically, the parameter optimization platform inputs the obtained optimized values of the parameters to be optimized into an analysis model, correspondingly optimizes the basic parameters through the optimized values of the parameters to be optimized based on coupling relations among the basic parameters, obtains the optimized values of the basic parameters, and records and stores the optimized values of the basic parameters.

In one embodiment, the parameter optimization platform stores the optimized values of the basic parameters in a matrix form, and if the optimized values of the parameters to be optimized cannot enable the optimized function to converge, the stored matrix is directly used as a multidimensional vector to be used as an input parameter in the adaptive moment estimation algorithm, so that the calculation is more convenient and faster.

In one embodiment, updating the initial design scheme according to the target optimization value of each parameter to be optimized includes: inputting the target optimization values of the parameters to be optimized into an analysis model to obtain target optimization values of the basic parameters of the target optimization object, and updating the initial design scheme according to the target optimization values of the basic parameters of the target optimization object to obtain the optimal design scheme of the nuclear reactor.

In the above embodiment, the parameter optimization platform models the target optimization object according to the basic parameters of the target optimization object to obtain an analysis model of the target optimization object, and the analysis model analyzes the optimized values of the basic parameters according to the optimized values of the parameters to be optimized to obtain the optimized values of the basic parameters, and the parameter optimization platform records the optimized values of the basic parameters. Based on the coupling relation between each parameter to be optimized and other parameters in the basic parameters, the data base is provided for updating the initial design scheme and obtaining a more accurate optimal design scheme through the optimized value of each parameter to be optimized.

In one embodiment, a method for optimizing multiple parameters of a nuclear reaction is provided, and as shown in fig. 6, the method is generally divided into 5 steps, which are respectively:

step 601, a multi-objective optimization function of an objective optimization object is established.

Step 602, modeling a target optimization object.

And 603, calling an optimization parameter model to perform multi-objective parameter optimization analysis based on the adaptive moment estimation algorithm.

Step 604, obtaining a multi-objective parameter optimization result.

Step 605, updating the initial design to obtain an optimized design.

Specifically, as shown in fig. 7, the method is described by taking an inverted U-shaped natural circulation steam generator of a nuclear reactor as a target optimization object, and the method is implemented as follows:

firstly, a parameter optimization platform determines each parameter to be optimized of a steam generator and an initial design scheme of a target optimization object. Wherein each parameter to be optimized comprises the volume (V) of the steam generator, the temperature (T) of the primary side outlet, the maximum flow velocity (V) of the primary side outlet ₂ ) Maximum secondary flow rate (v) ₂ ) The initial design scheme comprises basic parameters of each target optimization object, initial values of each parameter to be optimized, constraint boundary conditions of each parameter to be optimized, optimization targets of each target optimization object and the like.

Determining an optimization function minf (x) =min (0.5v+0.2t+0.1v) according to the initial design scheme and each parameter to be optimized ₁ +0.1ν ₂ +0.1Q), standard values of the parameters to be optimized, constraint boundary conditions of the parameters to be optimized.

And calling an optimization parameter model, carrying out parameter optimization on each parameter to be optimized according to the optimization function, the standard value of each parameter to be optimized and the constraint boundary condition of each parameter to be optimized, and searching the optimization value of each parameter to be optimized. And determining whether the optimized value of each parameter to be optimized meets the boundary limit value in the constraint boundary condition, and if not, carrying out inverse optimization according to the gradient and the step length.

If the parameters meet the requirements, an analysis model is generated according to each basic parameter of the target optimization object, wherein the basic parameters are the thermal parameters and the structural parameters of the target optimization object, and the basic parameters mainly comprise the structural parameters such as the height of the steam generator, the height of the heat transfer tubes, the outer diameters of the heat transfer tubes, the number of the heat transfer tubes and the like, and the thermal parameters such as the primary side flow rate of the steam generator, the primary side outlet temperature of the steam generator, the secondary side flow rate of the steam generator, the secondary side outlet temperature of the steam generator and the like.

Inputting the optimized values of the parameters to be optimized into an analysis model, analyzing and recording the optimized values of the basic parameters, judging whether the objective function is converged according to the optimized values of the parameters to be optimized, and if not, continuing to perform positive optimization according to the gradient and the step length. If the parameters are converged, recording the optimal solution of each parameter to be optimized as a target optimal value, and analyzing the target optimal value according to the target optimal value to obtain the target optimal value of each basic parameter. And updating the initial design scheme according to the target optimization value of each basic parameter to obtain an optimal design scheme.

It should be understood that, although the steps in the flowcharts related to the embodiments described above are sequentially shown as indicated by arrows, these steps are not necessarily sequentially performed in the order indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in the flowcharts described in the above embodiments may include a plurality of steps or a plurality of stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of the steps or stages is not necessarily performed sequentially, but may be performed alternately or alternately with at least some of the other steps or stages.

Based on the same inventive concept, the embodiment of the application also provides a nuclear reactor multi-parameter optimization device for realizing the above related nuclear reactor multi-parameter optimization method. The implementation of the solution provided by the device is similar to that described in the above method, so the specific limitations in the embodiments of the multi-parameter optimization device for a nuclear reactor provided below can be referred to as the limitations of the multi-parameter optimization method for a nuclear reactor hereinabove, and are not repeated herein.

In one embodiment, as shown in FIG. 8, a nuclear reactor multiparameter optimization device is provided, comprising: a parameter acquisition module 801, a function generation module 802, a function solution module 803, a parameter optimization module 804 and a scheme update module 805, wherein:

the parameter obtaining module 801 is configured to obtain each parameter to be optimized of a target optimization object in the nuclear reactor, and an initial design scheme of the nuclear reactor.

The function generating module 802 is configured to generate an optimization function of the target optimization object according to the initial design solution and each parameter to be optimized.

And the function solving module 803 is used for calling an optimization parameter model to solve an optimal solution for the optimization function to obtain an optimal value of each parameter to be optimized, and the optimization parameter model is pre-constructed according to the self-adaptive moment estimation algorithm.

The parameter optimization module 804 is configured to determine whether the optimization function converges based on the optimized value of each parameter to be optimized, and if so, determine that the optimized value of each parameter to be optimized is the target optimized value of each parameter to be optimized.

The solution updating module 805 is configured to update the initial design solution according to the target optimization value of the parameter to be optimized.

According to the multi-parameter optimizing device for the nuclear reactor, all parameters to be optimized of a target optimizing object in the nuclear reactor and an initial design scheme of the nuclear reactor are obtained, an optimizing function of the target optimizing object is generated according to the initial design scheme and all the parameters to be optimized, an optimizing function model which is built in advance according to a self-adaptive moment estimating algorithm is called to solve the optimizing function, the speed of obtaining an optimizing value is greatly improved, the optimizing function is generated according to the initial design scheme and all the parameters to be optimized, therefore, the value obtained by solving the optimizing function can be regarded as a parameter value after optimization, whether the optimizing function is converged or not is judged based on the optimizing value of all the parameters to be optimized, when the obtained optimizing value can enable the optimizing function to be converged, the optimizing value obtained at the moment is regarded as an optimal solution of the optimizing function, the optimizing value of the optimizing function is determined as a target optimizing value of the parameters to be optimized, the initial design scheme is updated according to the target optimizing value of the parameters to be optimized, and efficiency and accuracy of optimizing the parameters and the design scheme of the nuclear reactor are effectively improved.

In one embodiment, the function generation module is further to: determining an optimization target of a target optimization object according to the initial design scheme; determining the optimization weight of each parameter to be optimized based on the optimization target; and generating an optimization function according to each parameter to be optimized and the optimization weight of each parameter to be optimized.

In one embodiment, the function solving module is further configured to: determining the standard value of each parameter to be optimized according to the initial design scheme; performing iterative updating calculation by taking each standard value as an input value of the adaptive moment estimation algorithm to obtain an iterative updating value of each standard value; and if the iteration update values meet the update end conditions, determining the optimization values of the parameters to be optimized according to the iteration update values.

In one embodiment, the function solving module is further configured to: determining constraint conditions of each parameter to be optimized according to an initial design scheme; comparing each iteration update value with the constraint condition of the corresponding parameter to be optimized, and if each iteration update value meets the constraint condition of the corresponding parameter to be optimized, determining the optimized value of each parameter to be optimized according to each iteration update value and the standard value of each parameter to be optimized.

In one embodiment, the function solving module is further configured to: determining initial values of parameters to be optimized according to an initial design scheme; and carrying out normalization processing on each initial value to obtain the standard value of each parameter to be optimized corresponding to each initial value.

In one embodiment, the nuclear reactor multiparameter optimizing device further comprises: the analysis module is used for acquiring basic parameters of the target optimization object, wherein the basic parameters comprise parameters to be optimized; modeling the target optimization object according to the basic parameters to obtain an analysis model of the target optimization object; and inputting the optimized values of the parameters to be optimized into an analysis model to obtain optimized values of the basic parameters, and recording the optimized values of the basic parameters.

The various modules in the above-described nuclear reactor multi-parameter optimization apparatus may be implemented in whole or in part by software, hardware, and combinations thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.

In one embodiment, a computer device is provided, which may be a terminal or a server integrated with a parameter optimization platform, and the internal structure diagram thereof may be as shown in fig. 9. The computer device includes a processor, a memory, and a network interface connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the computer equipment is used for storing data such as an initial design scheme, parameters to be optimized, an optimized parameter model and the like. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program when executed by a processor implements a method of nuclear reactor multiparameter optimization.

It will be appreciated by those skilled in the art that the structure shown in fig. 9 is merely a block diagram of a portion of the structure associated with the present application and is not limiting of the computer device to which the present application applies, and that a particular computer device may include more or fewer components than shown, or may combine some of the components, or have a different arrangement of components.

In one embodiment, a computer device is provided comprising a memory and a processor, the memory having stored therein a computer program, the processor when executing the computer program performing the steps of:

generating an optimization function of a target optimization object according to the initial design scheme and each parameter to be optimized;

calling an optimization parameter model to solve an optimization function to obtain an optimization value of each parameter to be optimized, and constructing the optimization parameter model in advance according to a self-adaptive moment estimation algorithm;

judging whether the optimization function is converged based on the optimization value of each parameter to be optimized, and if so, determining the optimization value of each parameter to be optimized as a target optimization value of each parameter to be optimized;

In one embodiment, the processor when executing the computer program further performs the steps of:

determining an optimization target of a target optimization object according to the initial design scheme;

and if the iteration update values meet the update end conditions, determining the optimization values of the parameters to be optimized according to the iteration update values.

determining constraint conditions of each parameter to be optimized according to an initial design scheme;

determining initial values of parameters to be optimized according to an initial design scheme;

and carrying out normalization processing on each initial value to obtain the standard value of each parameter to be optimized corresponding to each initial value.

obtaining basic parameters of a target optimization object, wherein the basic parameters comprise parameters to be optimized;

and inputting the optimized values of the parameters to be optimized into an analysis model to obtain optimized values of the basic parameters, and recording the optimized values of the basic parameters.

In one embodiment, a computer readable storage medium is provided having a computer program stored thereon, which when executed by a processor, performs the steps of:

In one embodiment, the computer program when executed by the processor further performs the steps of:

In one embodiment, a computer program product is provided comprising a computer program which, when executed by a processor, performs the steps of:

It should be noted that, user information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, presented data, etc.) referred to in the present application are information and data authorized by the user or sufficiently authorized by each party.

Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, database, or other medium used in the various embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, high density embedded nonvolatile Memory, resistive random access Memory (ReRAM), magnetic random access Memory (Magnetoresistive Random Access Memory, MRAM), ferroelectric Memory (Ferroelectric Random Access Memory, FRAM), phase change Memory (Phase Change Memory, PCM), graphene Memory, and the like. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory, and the like. By way of illustration, and not limitation, RAM can be in the form of a variety of forms, such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), and the like. The databases referred to in the various embodiments provided herein may include at least one of relational databases and non-relational databases. The non-relational database may include, but is not limited to, a blockchain-based distributed database, and the like. The processors referred to in the embodiments provided herein may be general purpose processors, central processing units, graphics processors, digital signal processors, programmable logic units, quantum computing-based data processing logic units, etc., without being limited thereto.

The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.

The above examples only represent a few embodiments of the present application, which are described in more detail and are not to be construed as limiting the scope of the present application. It should be noted that it would be apparent to those skilled in the art that various modifications and improvements could be made without departing from the spirit of the present application, which would be within the scope of the present application. Accordingly, the scope of protection of the present application shall be subject to the appended claims.

Claims

1. A method of multi-parameter optimization for a nuclear reactor, the method comprising:

2. The method of claim 1, wherein generating an optimization function from the initial design and each of the parameters to be optimized comprises:

3. The method of claim 1, wherein the calling an optimization parameter model to solve the optimization function to obtain an optimized value of each parameter to be optimized comprises:

4. A method according to claim 3, wherein said determining an optimized value for each of said parameters to be optimized from each of said iteratively updated values comprises:

5. A method according to claim 3, wherein said determining a standard value for each of said parameters to be optimized according to said initial design scheme comprises:

6. The method according to any one of claims 1 to 5, wherein said determining whether the optimization function converges based on the optimization value of each of the parameters to be optimized further comprises:

7. A nuclear reactor multiparameter optimizing device, said device comprising:

8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any of claims 1 to 6 when the computer program is executed.

9. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 6.

10. A computer program product comprising a computer program, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 6.