CN-122020058-A - Method and related device for predicting critical heat flux density of offshore floating nuclear power plant

CN122020058ACN 122020058 ACN122020058 ACN 122020058ACN-122020058-A

Abstract

The embodiment of the invention provides a critical heat flux density prediction method and a related device for a marine floating nuclear power station, and relates to the technical field of reactor safety analysis. The method comprises the steps of obtaining multidimensional operation parameters of the offshore floating nuclear power station, wherein the multidimensional operation parameters comprise a thermal parameter, a marine condition parameter, a working medium type parameter, a working condition parameter and a heat exchanger morphological parameter, inputting the multidimensional operation parameters into a pre-trained critical heat flow density prediction model, and outputting a corresponding critical heat flow density prediction value through the critical heat flow density prediction model. The method is specially used for accurately predicting the critical heat flux density of the offshore floating nuclear power station by pre-training a critical heat flux density prediction model and comprehensively considering multi-factor coupling influences such as thermal parameters, ocean condition parameters, working medium type parameters, working condition parameters, heat exchanger morphological parameters and the like, so that data support is provided for safe operation of the offshore floating nuclear power station.

Inventors

WANG LANG
LIU WEI
JIA CHANGMING
Lan Siying
LI XUELIN
LIU SONGYANG
GUO JINSONG
GU CHEN
LUO YONG

Assignees

华能核能技术研究院有限公司
华能(福建)能源开发有限公司福州分公司

Dates

Publication Date: 20260512
Application Date: 20260204

Claims (10)

1. A method for predicting critical heat flux density of a floating nuclear power plant at sea, the method comprising: acquiring multidimensional operation parameters of the offshore floating nuclear power station, wherein the multidimensional operation parameters comprise thermal parameters, ocean condition parameters, working medium type parameters, working condition parameters and heat exchanger morphological parameters; inputting the multidimensional operation parameters into a pre-trained critical heat flow density prediction model; and outputting a corresponding critical heat flux density predicted value through the critical heat flux density predicted model.
2. The method of claim 1, wherein the critical heat flux density prediction model comprises an input layer, a plurality of hidden layers and an output layer, and wherein outputting the corresponding critical heat flux density prediction value by the critical heat flux density prediction model comprises: receiving the multidimensional operation parameters through the input layer, and generating initial feature vectors based on the multidimensional operation parameters; performing nonlinear transformation processing on the initial feature vector through the plurality of hidden layers to obtain a corresponding high-order feature vector; And converting the high-order characteristic vector into a corresponding critical heat flow density predicted value through the output layer.
3. The method of critical heat flux density prediction for a floating offshore nuclear power plant of claim 1, wherein prior to inputting the multi-dimensional operating parameters into a pre-trained critical heat flux density prediction model, the method further comprises: normalizing the multidimensional operation parameters; the inputting the multidimensional operation parameters into a pre-trained critical heat flow density prediction model comprises the following steps: and inputting the multidimensional operation parameters after normalization treatment into a pre-trained critical heat flow density prediction model.
4. A critical heat flux density prediction method of a floating offshore nuclear power plant according to any of claims 1-3, characterized in that the critical heat flux density prediction model is trained by: Acquiring sample data under a conventional working condition and sample data under a target working condition, wherein the conventional working condition is a working condition with sufficient sample data, and the target working condition is a working condition with lack of sample data; Pre-training a pre-constructed deep learning model based on the sample data under the conventional working condition to obtain a pre-training model; And performing fine tuning training on the pre-training model based on the sample data under the target working condition to obtain the critical heat flow density prediction model.
5. The method for predicting critical heat flux density of a floating offshore nuclear power plant according to claim 4, wherein the performing fine-tuning training on the pre-training model based on the sample data under the target working condition to obtain the critical heat flux density prediction model comprises: and freezing part of hidden layers of the pre-training model, and adjusting parameters of the hidden layers close to the output layer and the output layer in the pre-training model to obtain the critical heat flow density prediction model.
6. A critical heat flux density prediction apparatus for a floating nuclear power plant at sea, the apparatus comprising: The system comprises an acquisition module, a heat exchanger and a control module, wherein the acquisition module is used for acquiring multidimensional operation parameters of the offshore floating nuclear power station, and the multidimensional operation parameters comprise thermal parameters, ocean condition parameters, working medium type parameters, working condition parameters and heat exchanger morphological parameters; The input module is used for inputting the multidimensional operation parameters into a pre-trained critical heat flow density prediction model; And the processing module is used for outputting a corresponding critical heat flux density predicted value through the critical heat flux density predicted model.
7. The critical heat flux density prediction device of the offshore floating nuclear power plant according to claim 6, wherein the critical heat flux density prediction model comprises an input layer, a plurality of hidden layers and an output layer, the processing module is used for receiving the multidimensional operation parameters through the input layer and generating initial feature vectors based on the multidimensional operation parameters, neurons in the input layer are in one-to-one correspondence with the multidimensional operation parameters, nonlinear transformation processing is conducted on the initial feature vectors through the hidden layers to obtain corresponding high-order feature vectors, and the high-order feature vectors are converted into corresponding critical heat flux density predicted values through the output layer.
8. The critical heat flux density predicting device of a floating offshore nuclear power plant as set forth in claim 6, further comprising: the preprocessing module is used for carrying out normalization processing on the multidimensional operation parameters; The input module is used for inputting the multidimensional operation parameters after normalization processing into a pre-trained critical heat flow density prediction model.
9. An electronic device comprising a processor, a memory and a computer program stored on the memory and executable on the processor, which when executed by the processor, performs the steps of the critical heat flux density prediction method of a floating marine nuclear power plant as claimed in any of claims 1-5.
10. A computer readable storage medium, characterized in that it has stored thereon a computer program which, when executed by a processor, implements the steps of the critical heat flow density prediction method of a floating marine nuclear power plant as claimed in any of claims 1-5.

Description

Method and related device for predicting critical heat flux density of offshore floating nuclear power plant Technical Field The invention relates to the technical field of reactor safety analysis, in particular to a critical heat flow density prediction method and a related device of a marine floating nuclear power station. Background The floating nuclear power station at sea is an innovative nuclear power generation facility aimed at providing stable power supply to regions far from land, island countries, and offshore work platforms. With the growing global energy demand and the increasing demand for mobile, deployable energy solutions, the concept of floating nuclear power plants at sea has received international social attention. However, its development and operation face multiple challenges of technology, security, and regulatory. CHF (CRITICAL HEAT Flux, critical heat Flux density) is a key parameter for the safe operation of a thermodynamic device, indicating the point at which the heat transfer rate reaches its maximum limit. Once the actual heat flux density exceeds CHF, film boiling is initiated, resulting in deteriorated heat transfer and a sharp rise in wall temperature, which in turn results in severe consequences such as fuel cladding melting. Therefore, accurate prediction of CHF is critical to ensuring safe operation of a floating nuclear power plant at sea. Traditional CHF predictions rely primarily on land-based empirical relationships or numerical simulations. The land-based empirical relationship may not be suitable for offshore reactor conditions and it is difficult for conventional empirical relationships to accurately reflect CHF characteristics in offshore floating nuclear power plants due to the complexity of the marine environment. The use of numerical simulation methods, such as Computational Fluid Dynamics (CFD), to predict CHF in a floating nuclear power plant at sea, while capable of taking into account the effects of some ocean conditions, is labor intensive and difficult to guarantee with accuracy, especially for complex ocean conditions, such as heave motions, different working fluids and working condition combinations, requires significant computational resources and the results are susceptible to model assumptions and boundary conditions. Furthermore, since the floating nuclear power plant at sea is an emerging field, the relevant experimental data is relatively scarce. In particular, for certain extreme ocean conditions or novel working medium combinations, experimental data may be completely missing, making conventional methods difficult to apply. Disclosure of Invention In view of the above, the present invention is directed to a method and a related device for predicting critical heat flux density of a floating nuclear power plant at sea, so as to solve the problem that it is difficult to accurately predict critical heat flux density of the floating nuclear power plant at sea in the related art. In order to achieve the above object, the technical scheme adopted by the embodiment of the invention is as follows: In a first aspect, the present invention provides a critical heat flux density prediction method for a floating offshore nuclear power plant, the method comprising: acquiring multidimensional operation parameters of the offshore floating nuclear power station, wherein the multidimensional operation parameters comprise thermal parameters, ocean condition parameters, working medium type parameters, working condition parameters and heat exchanger morphological parameters; inputting the multidimensional operation parameters into a pre-trained critical heat flow density prediction model; and outputting a corresponding critical heat flux density predicted value through the critical heat flux density predicted model. In an alternative embodiment, the critical heat flux density prediction model comprises an input layer, a plurality of hidden layers and an output layer, and the outputting of the corresponding critical heat flux density prediction value by the critical heat flux density prediction model comprises the following steps: receiving the multidimensional operation parameters through the input layer, and generating initial feature vectors based on the multidimensional operation parameters; performing nonlinear transformation processing on the initial feature vector through the plurality of hidden layers to obtain a corresponding high-order feature vector; And converting the high-order characteristic vector into a corresponding critical heat flow density predicted value through the output layer. In an alternative embodiment, prior to inputting the multi-dimensional operating parameters into a pre-trained critical heat flux density prediction model, the method further comprises: normalizing the multidimensional operation parameters; the inputting the multidimensional operation parameters into a pre-trained critical heat flow density prediction model comprises the