CN-122024906-A - Supercritical carbon dioxide heat exchange prediction method, device, electronic equipment, computer readable storage medium and program product

CN122024906ACN 122024906 ACN122024906 ACN 122024906ACN-122024906-A

Abstract

The embodiment of the invention provides a supercritical carbon dioxide heat exchange prediction method, a device, electronic equipment, a computer readable storage medium and a program product, and relates to the field of thermal engineering. According to the method, the working condition processing information is obtained by preprocessing the working condition information to be predicted, which comprises the thermal physical parameters representing the physical properties of the supercritical carbon dioxide, the flow parameters representing the motion state of the supercritical carbon dioxide, the heat exchanger morphological parameters representing the heat exchange environment and the flow direction parameters representing the relative relation between the flow direction and the gravity direction, and the heat exchange coefficient prediction is carried out based on the working condition processing information by utilizing a pre-trained heat exchange prediction model, so that the coupling influence among the physical property change, the flow state, the structural characteristics and the spatial arrangement can be comprehensively reflected. The heat exchange prediction model effectively overcomes the limitation that the traditional experience relation is dependent on manual judgment and single application condition by learning nonlinear mapping relations in a large amount of working condition data, and improves the prediction precision of heat exchange behaviors under complex working conditions.

Inventors

WANG LANG
LIU WEI
JIA CHANGMING
TENG XIANG
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 supercritical carbon dioxide heat exchange prediction method, characterized in that the method comprises: the method comprises the steps of preprocessing received working condition information to be predicted to obtain working condition processing information, wherein the working condition information to be predicted comprises a thermophysical parameter, a flow parameter, a heat exchanger morphological parameter and a flow direction parameter, the thermophysical parameter represents the physical property of supercritical carbon dioxide, the flow parameter represents the motion state of the supercritical carbon dioxide, the heat exchanger morphological parameter represents a heat exchange environment, and the flow direction parameter represents the relative relation between the flow direction of the supercritical carbon dioxide and the gravity direction; and predicting based on the working condition processing information by utilizing a pre-trained heat exchange prediction model to obtain a heat exchange coefficient corresponding to the working condition information to be predicted.
2. The supercritical carbon dioxide heat exchange prediction method according to claim 1, wherein the heat exchanger morphological parameters include a heat exchanger type, the predicting based on the working condition processing information by using a pre-trained heat exchange prediction model to obtain a heat exchange coefficient corresponding to the working condition information to be predicted comprises: Determining a matched heat exchange prediction model from a plurality of pre-trained heat exchange prediction models according to the heat exchanger type in the working condition information to be predicted; And inputting the working condition processing information into a matched heat exchange prediction model for prediction to obtain a heat exchange coefficient corresponding to the working condition information to be predicted.
3. The supercritical carbon dioxide heat exchange prediction method according to claim 2, wherein the heat exchange prediction model comprises an input layer, a plurality of hidden layers and an output layer, the hidden layers adopt a full-connection structure, the working condition processing information is input into a matched heat exchange prediction model for prediction, and a heat exchange coefficient corresponding to the working condition information to be predicted is obtained, and the method comprises the following steps: Inputting the working condition processing information to an input layer of the heat exchange prediction model; Mapping the working condition processing information into working condition characteristics by utilizing a plurality of neurons in the input layer; capturing heat transfer rules layer by layer for the working condition characteristics by utilizing the full-connection structures and the nonlinear activation functions of the plurality of hidden layers, and outputting heat transfer characteristic vectors by the last hidden layer; and converting the heat exchange characteristic vector into a heat exchange coefficient corresponding to the working condition information to be predicted by using the output layer.
4. A supercritical carbon dioxide heat exchange prediction method according to any one of claims 1-3, wherein the heat exchange prediction model is trained by: cleaning and normalizing the original working condition data to obtain a standard training data set; Building a plurality of initial deep learning models containing different numbers of hidden layers; training each initial deep learning model according to the standard training data set to obtain an optimized deep learning model; And determining the heat exchange prediction model from a plurality of optimized deep learning models according to a preset complexity threshold and a preset error threshold.
5. The supercritical carbon dioxide heat exchange prediction method according to claim 2, further comprising: if the working condition data of the target heat exchanger type is less than the training data quantity threshold value, acquiring a target heat exchange prediction model from heat exchange prediction models corresponding to heat exchanger types other than the target heat exchanger type; cleaning and normalizing the working condition data of the target heat exchanger type to obtain target working condition training data; And performing transfer learning fine adjustment on the target heat exchange prediction model according to the target working condition training data to obtain a heat exchange prediction model corresponding to the type of the target heat exchanger.
6. The supercritical carbon dioxide heat exchange prediction method according to claim 1, wherein the preprocessing the received working condition information to be predicted to obtain working condition processing information comprises: and carrying out normalization processing on the thermophysical parameter, the flow parameter, the heat exchanger morphological parameter and the flow direction parameter to obtain the working condition processing information.
7. A supercritical carbon dioxide heat exchange prediction apparatus, the apparatus comprising: The system comprises a processing module, a heat exchanger and a heat exchange module, wherein the processing module is used for preprocessing the received working condition information to be predicted to obtain working condition processing information, the working condition information to be predicted comprises a thermophysical parameter, a flow parameter, a heat exchanger form parameter and a flow direction parameter, the thermophysical parameter represents the physical property of the supercritical carbon dioxide, the flow parameter represents the motion state of the supercritical carbon dioxide, the heat exchanger form parameter represents a heat exchange environment, and the flow direction parameter represents the relative relation between the flow direction of the supercritical carbon dioxide and the gravity direction; And the prediction module is used for predicting based on the working condition processing information by utilizing a pre-trained heat exchange prediction model to obtain a heat exchange coefficient corresponding to the working condition information to be predicted.
8. An electronic device comprising a processor and a memory, the memory storing a computer program executable by the processor, the processor executable by the computer program to implement the supercritical carbon dioxide heat exchange prediction method of any one of claims 1-6.
9. A computer readable storage medium having stored thereon a computer program, which when executed by a processor implements the supercritical carbon dioxide heat exchange prediction method according to any one of claims 1-6.
10. A program product, characterized in that it, when executed by a processor, implements the supercritical carbon dioxide heat exchange prediction method according to any one of claims 1-6.

Description

Supercritical carbon dioxide heat exchange prediction method, device, electronic equipment, computer readable storage medium and program product Technical Field The invention relates to the technical field of supercritical carbon dioxide heat exchange performance prediction in the field of thermal engineering, in particular to a supercritical carbon dioxide heat exchange prediction method, a device, electronic equipment, a computer readable storage medium and a program product. Background Supercritical carbon dioxide (Supercritical Carbon Dioxide, S-CO 2 for short) has wide application prospect in the fields of advanced nuclear energy systems, solar thermal power generation, medium-low temperature waste heat recovery and other high-energy conversion due to lower critical parameters (critical temperature 31.0 ℃ and critical pressure 7.38 MPa) and excellent thermophysical characteristics. The S-CO 2 has high gas diffusivity and high liquid density in the supercritical state, and can realize compact system design and high-heat-efficiency operation. However, accurate prediction of heat exchange performance thereof faces significant technical challenges. At present, the prediction of the heat exchange performance of supercritical carbon dioxide mainly depends on experimental or semi-empirical relation based on experimental data fitting, a plurality of special correlation formulas are needed to be selected according to complex working conditions such as different flow directions, heat exchanger forms and the like, the selection process is highly dependent on manual experience, uniformity and self-adaptation capability are lacked, and nonlinear heat transfer behaviors under the action of severe physical changes and multi-factor coupling are difficult to accurately describe, so that the prediction precision is low and the application range is limited. Disclosure of Invention Accordingly, the present invention is directed to a supercritical carbon dioxide heat exchange prediction method, device, electronic apparatus, computer readable storage medium and program product, which can improve the prediction accuracy of heat exchange behavior under complex working conditions. 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 supercritical carbon dioxide heat exchange prediction method, the method comprising: the method comprises the steps of preprocessing received working condition information to be predicted to obtain working condition processing information, wherein the working condition information to be predicted comprises a thermophysical parameter, a flow parameter, a heat exchanger morphological parameter and a flow direction parameter, the thermophysical parameter represents the physical property of supercritical carbon dioxide, the flow parameter represents the motion state of the supercritical carbon dioxide, the heat exchanger morphological parameter represents a heat exchange environment, and the flow direction parameter represents the relative relation between the flow direction of the supercritical carbon dioxide and the gravity direction; and predicting based on the working condition processing information by utilizing a pre-trained heat exchange prediction model to obtain a heat exchange coefficient corresponding to the working condition information to be predicted. In an alternative embodiment, the heat exchanger morphological parameters include a heat exchanger type, the predicting by using a pre-trained heat exchange prediction model based on the working condition processing information to obtain a heat exchange coefficient corresponding to the working condition information to be predicted includes: Determining a matched heat exchange prediction model from a plurality of pre-trained heat exchange prediction models according to the heat exchanger type in the working condition information to be predicted; And inputting the working condition processing information into a matched heat exchange prediction model for prediction to obtain a heat exchange coefficient corresponding to the working condition information to be predicted. In an optional embodiment, the heat exchange prediction model includes an input layer, a plurality of hidden layers and an output layer, the plurality of hidden layers adopt a fully connected structure, the step of inputting the working condition processing information into the matched heat exchange prediction model for prediction to obtain a heat exchange coefficient corresponding to the working condition information to be predicted includes: Inputting the working condition processing information to an input layer of the heat exchange prediction model; Mapping the working condition processing information into working condition characteristics by utilizing a plurality of neurons in the input layer; capturing heat transfer rules layer by layer for the working conditi