Method for determining heat exchange data, method and device for training deep learning model

Abstract

The present disclosure provides a heat exchange data determination method, a deep learning model training method, an apparatus, an electronic device, a storage medium and a program product, relates to the technical field of artificial intelligence, and in particular to the technical fields of deep learning, energy, power, industrial big data, mechanical manufacturing and the like. The specific implementation scheme is as follows: inputting flow channel attribute data of a heat exchange device and fluid attribute data of a fluid into a plurality of deep learning models to obtain a plurality of parameter data, the fluid being a medium flowing in the heat exchange device; and determining heat exchange data between the fluid and a heat exchange wall of the heat exchange device based on first physical property data, second physical property data and the plurality of parameter data, the first physical property data being heat exchange related physical property data of the heat exchange wall, and the second physical property data being heat exchange related physical property data of the fluid.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of artificial intelligence, in particular to the technical fields of deep learning, energy, power, industrial big data, and mechanical manufacturing, and more particularly to a heat exchange data determination method, a deep learning model training method, device, electronic equipment, storage medium, and program product. BACKGROUND

[0002] Heat exchange devices are widely used in chemical industry, oil and gas, aerospace, semiconductor, and other fields. Accurate determination of the convective heat transfer performance of a heat exchange device plays an important role in process parameter optimization, energy saving, and consumption reduction. SUMMARY

[0003] The present disclosure provides a heat exchange data determination method, a deep learning model training method, device, electronic equipment, storage medium, and program product.

[0004] According to an aspect of the present disclosure, a heat exchange data determination method is provided, including: inputting flow channel attribute data of a heat exchange device and fluid attribute data of a fluid into a plurality of deep learning models to obtain a plurality of parameter data, wherein the fluid is a medium flowing in the heat exchange device; and determining heat exchange data between the fluid and a heat exchange wall of the heat exchange device based on first physical property data, second physical property data, and the plurality of parameter data, wherein the first physical property data is heat exchange related physical property data of the heat exchange wall, and the second physical property data is heat exchange related physical property data of the fluid.

[0005] According to another aspect of the present disclosure, a deep learning model training method is provided, including: inputting sample flow channel attribute data of a heat exchange device and sample fluid attribute data of a fluid into a plurality of deep learning models to obtain a plurality of sample parameter data, wherein the fluid is a medium flowing in the heat exchange device; determining sample heat exchange data between the fluid and a heat exchange wall of the heat exchange device based on sample first physical property data, sample second physical property data, and the plurality of sample parameter data, wherein the sample first physical property data is heat exchange related physical property data of the heat exchange wall, and the sample second physical property data is heat exchange related physical property data of the fluid; determining a first loss value based on the sample heat exchange data and a sample label, wherein the sample label is true data matched with the sample flow channel attribute data, the sample fluid attribute data, the sample first physical property data, and the sample second physical property data; and adjusting parameters of the plurality of deep learning models based on the first loss value.

[0006] According to another aspect of the present disclosure, a heat exchange data determination apparatus is provided, comprising: a data input module configured to input flow channel attribute data of a heat exchange device and fluid attribute data of a fluid into a plurality of deep learning models to obtain a plurality of parameter data, wherein the fluid is a medium flowing in the heat exchange device; and a data determination module configured to determine heat exchange data between the fluid and a heat exchange wall of the heat exchange device based on first physical property data, second physical property data, and the plurality of parameter data, wherein the first physical property data is heat exchange related physical property data of the heat exchange wall, and the second physical property data is heat exchange related physical property data of the fluid.

[0007] According to another aspect of the present disclosure, a deep learning model training apparatus is provided, comprising: a sample data input module configured to input sample flow channel attribute data of a heat exchange device and sample fluid attribute data of a fluid into a plurality of deep learning models to obtain a plurality of sample parameter data, wherein the fluid is a medium flowing in the heat exchange device; a sample determination module configured to determine sample heat exchange data between the fluid and a heat exchange wall of the heat exchange device based on sample first physical property data, sample second physical property data, and the plurality of sample parameter data, wherein the sample first physical property data is heat exchange related physical property data of the heat exchange wall, and the sample second physical property data is heat exchange related physical property data of the fluid; a loss value determination module configured to determine a first loss value based on the sample heat exchange data and a sample label, wherein the sample label is true data matched with the sample flow channel attribute data, the sample fluid attribute data, the sample first physical property data, and the sample second physical property data; and a parameter adjustment module configured to adjust parameters of the plurality of deep learning models based on the first loss value.

[0008] According to another aspect of the present disclosure, an electronic device is provided, comprising: at least one processor; and a memory communicatively connected with the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform a method according to the present disclosure.

[0009] According to another aspect of the present disclosure, a non-transitory computer readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable a computer to perform a method according to the present disclosure.

[0010] According to another aspect of the present disclosure, a computer program product is provided, comprising a computer program, wherein the computer program, when executed by a processor, implements a method according to the present disclosure.

[0011] It should be appreciated that the content described in this section is not intended to identify key or important features of the embodiments of the disclosure, nor is it intended to limit the scope of the disclosure. Other features of the disclosure will become apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS

[0012] The accompanying drawings are used to better understand the present scheme and do not limit the disclosure. Among them:

[0013] Figure 1 An exemplary system architecture to which the determination method and device of heat exchange data according to embodiments of the disclosure can be applied is schematically shown;

[0014] Figure 2 A flowchart of the determination method of heat exchange data according to embodiments of the disclosure is schematically shown;

[0015] Figure 3 A schematic diagram of determining heat exchange intensity according to another embodiment of the disclosure is schematically shown;

[0016] Figure 4 A flowchart of the determination method of heat exchange data according to another embodiment of the disclosure is schematically shown;

[0017] Figure 5 A flowchart of the training method of a deep learning model according to embodiments of the disclosure is schematically shown;

[0018] Figure 6 A block diagram of the determination device of heat exchange data according to embodiments of the disclosure is schematically shown;

[0019] Figure 7 A block diagram of the training device of a deep learning model according to embodiments of the disclosure is schematically shown; and

[0020] Figure 8 A block diagram of an electronic device suitable for implementing the determination method of heat exchange data according to embodiments of the disclosure is schematically shown. DETAILED DESCRIPTION

[0021] Exemplary embodiments of the disclosure are described below with reference to the accompanying drawings, which include various details of the embodiments of the disclosure to help understanding, and should be considered as merely exemplary. Therefore, those skilled in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the disclosure. Also, for the sake of clarity and conciseness, the description below omits the description of well-known functions and structures.

[0022] The disclosure provides a determination method of heat exchange data, a training method of a deep learning model, devices, electronic devices, storage media and program products.

[0023] According to an embodiment of the present disclosure, the method for determining heat exchange data can comprise: inputting flow channel attribute data of a heat exchange device and fluid attribute data of a fluid into a plurality of deep learning models to obtain a plurality of parameter data, the fluid being a medium flowing in the heat exchange device; and determining heat exchange data between the fluid and a heat exchange wall of the heat exchange device based on first physical property data, second physical property data, and the plurality of parameter data, the first physical property data being heat exchange related physical property data of the heat exchange wall, and the second physical property data being heat exchange related physical property data of the fluid.

[0024] In the technical solutions of the present disclosure, the collection, storage, use, processing, transmission, provision, disclosure, and application of user personal information comply with relevant laws and regulations, necessary security measures are taken, and the public order and good customs are not violated.

[0025] In the technical solutions of the present disclosure, the authorization or consent of the user is obtained before the user personal information is acquired or collected.

[0026] Figure 1 An example system architecture to which the method and device for determining heat exchange data according to an embodiment of the present disclosure can be applied is schematically shown.

[0027] It should be noted that, Figure 1 The system architecture shown is only an example of a system architecture to which the embodiments of the present disclosure can be applied, to help those skilled in the art understand the technical content of the present disclosure, but does not mean that the embodiments of the present disclosure cannot be applied to other devices, systems, environments, or scenarios. For example, in another embodiment, the example system architecture to which the method and device for determining heat exchange data can be applied can include a terminal device, but the terminal device can not need to interact with the server to implement the method and device for determining heat exchange data provided by the embodiments of the present disclosure.

[0028] As Figure 1 shown, the system architecture 100 according to this embodiment can include terminal devices 101, 102, 103, a network 104, and a server 105. The network 104 is used to provide a fluid for communication links between the terminal devices 101, 102, 103 and the server 105. The network 104 can include various connection types, such as wired and / or wireless communication links, etc.

[0029] The user can use the terminal devices 101, 102, 103 to interact with the server 105 through the network 104 to receive or send messages, etc. Various communication client applications can be installed on the terminal devices 101, 102, 103, such as knowledge reading applications, web browser applications, search applications, instant messaging tools, email clients, and / or social platform software, etc. (only as examples).

[0030] The terminal devices 101, 102, and 103 can be various electronic devices with display screens and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, desktop computers, and the like.

[0031] The server 105 can be a server providing various services, such as a background management server providing support for content browsed by users using the terminal devices 101, 102, and 103 (only as an example). The background management server can perform analysis and the like on received user requests and the like, and feed back the processing results (such as web pages, information, or data obtained or generated according to user requests, and the like) to the terminal devices.

[0032] It should be noted that the method for determining heat exchange data provided by the embodiments of the present disclosure can generally be executed by the terminal devices 101, 102, or 103. Accordingly, the apparatus for determining heat exchange data provided by the embodiments of the present disclosure can also be arranged in the terminal devices 101, 102, or 103.

[0033] Alternatively, the method for determining heat exchange data provided by the embodiments of the present disclosure can also be generally executed by the server 105. Accordingly, the apparatus for determining heat exchange data provided by the embodiments of the present disclosure can be generally arranged in the server 105. The method for determining heat exchange data provided by the embodiments of the present disclosure can also be executed by a server or a server cluster different from the server 105 and capable of communicating with the terminal devices 101, 102, and 103 and / or the server 105. Accordingly, the apparatus for determining heat exchange data provided by the embodiments of the present disclosure can also be arranged in a server or a server cluster different from the server 105 and capable of communicating with the terminal devices 101, 102, and 103 and / or the server 105.

[0034] It should be understood that the number of terminal devices, networks, and servers in the system 100 is only illustrative. Any number of terminal devices, networks, and servers can be provided according to implementation needs. Figure 1

[0035] It should be noted that the serial numbers of the various operations in the following method are only used to represent the operations for description, and should not be regarded as representing the execution sequence of the various operations. The method does not need to be executed in the order shown unless explicitly indicated.

[0036] Figure 2 A flowchart of a method for determining heat exchange data according to an embodiment of the present disclosure is schematically shown.

[0037] As shown in Figure 2 , the method includes operations S210-S220.

[0038] ​In operation S210, the flow channel attribute data of the heat exchange device and the fluid attribute data of the fluid are input into a plurality of deep learning models to obtain a plurality of parameter data.

[0039] In operation S220, based on the first physical property data, the second physical property data, and the plurality of parameter data, heat exchange data between the fluid and the heat exchange wall of the heat exchange device is determined.

[0040] According to an embodiment of the present disclosure, the first physical property data is heat exchange related physical property data of the heat exchange wall. The second physical property data is heat exchange related physical property data of the fluid.

[0041] According to an embodiment of the present disclosure, the heat exchange device can be a device that exchanges energy in a way of convection heat exchange. The heat exchange device can be applied in an industrial manufacturing scene, such as in the fields of chemical industry, power generation, heating, oil and gas, aerospace, automobile, semiconductor, etc.

[0042] According to an embodiment of the present disclosure, the fluid is a medium flowing in the heat exchange device. For example, the fluid can be a medium flowing in the heat exchange device for heat exchange. The type of the fluid is not limited, for example, it can be a gas, a liquid, etc. As long as the medium can be in contact with the heat exchange wall to realize convection heat exchange, it is acceptable.

[0043] According to an embodiment of the present disclosure, the fluid can be a medium with a temperature higher than that of the heat exchange wall of the heat exchange device, for transmitting heat to the heat exchange wall. However, it is not limited thereto. The fluid can also be a medium with a temperature lower than that of the heat exchange wall, for receiving heat transmitted by the heat exchange wall.

[0044] According to an embodiment of the present disclosure, the heat exchange data between the fluid and the heat exchange wall of the heat exchange device can include heat exchange characteristic parameters such as Nusselt number Nu, friction coefficient, heat of the heat exchange unit, enthalpy of the heat exchange unit, heat exchange coefficient of the heat exchange unit, etc. As long as the data can represent the heat exchange capacity between the heat exchange device and the fluid, it is acceptable.

[0045] According to an embodiment of the present disclosure, the fluid attribute data and the flow channel attribute data can be input into a plurality of deep learning models to obtain a plurality of parameter data. The parameter data can be used to represent the heat exchange influencing factors between the solid and the fluid. The plurality of parameter data can represent the heat exchange influencing factors from different angles and different dimensions. Thus, the heat exchange data determined based on the first physical property data, the second physical property data, and the plurality of parameter data has high accuracy and wide application range.

[0046] According to a related example, the heat exchange data can be determined by simulation, for example, by using mass conservation, energy conservation, momentum conservation, etc. to simulate the velocity, pressure, temperature of the fluid flowing in the heat exchange device and the pressure, temperature of the heat exchange wall, etc. and then determine the heat exchange data.

[0047] Compared with the simulation mode, the heat exchange data determination method provided by the embodiment of the present disclosure can determine the heat exchange data based on the actually measured flow channel attribute data, fluid attribute data, first physical property data and second physical property data, avoid determining based on ideal data or simulation data, and avoid the problem that the determined heat exchange data is greatly different from the actual situation.

[0048] According to another related example, the heat exchange data can be determined by using a correction equation, for example, using the thermodynamic properties of the fluid, such as the fluid attribute data and the second physical property data, and combining the flow channel attribute data, etc., to calculate the heat exchange data based on an empirical equation, such as a correction equation. The correction equation can be obtained by analyzing and fitting by staff under laboratory conditions.

[0049] Compared with the determination mode using the correction equation, the heat exchange data determination method provided by the embodiment of the present disclosure can use multiple deep learning models to predict multiple parameter data based on the flow channel attribute data and the fluid attribute data, and determine the heat exchange data based on the multiple parameter data, the first physical property data and the second physical property data, thereby improving the accuracy of the heat exchange data while expanding the application range, and avoiding the problem that the output result of the correction equation is greatly different from the actual situation due to the difference between the actual situation and the experimental conditions.

[0050] According to another related example, the heat exchange data can be determined by using a single deep learning model, linear regression or support vector machine, etc. For example, a single deep learning model is trained using training samples, so that the input data includes the fluid attribute data and the flow channel attribute data, and the output data includes the heat exchange data.

[0051] Compared with the determination mode using a single deep learning model, the heat exchange data determination method provided by the embodiment of the present disclosure can use multiple deep learning models to obtain multiple parameter data based on the flow channel attribute data and the fluid attribute data, and thereby make the network structure of each deep learning model in the multiple deep learning models small, while the training speed is high and the application range is wide.

[0052] According to the embodiment of the present disclosure, for the operation S210 as shown in Figure 1 The flow channel attribute data of the heat exchange device and the fluid attribute data of the fluid are input into the multiple deep learning models to obtain multiple parameter data. It can be understood that the multiple deep learning models are used as sub-models to form a target deep learning model. The flow channel attribute data of the heat exchange device and the fluid attribute data of the fluid can be used as input data and input into the target deep learning model to obtain output data, such as multiple parameter data.

[0053] According to an embodiment of the present disclosure, the initial network structures of the plurality of deep learning models can be the same or different, as long as they are trained models capable of taking the flow passage attribute data and the fluid attribute data as input data and taking the plurality of parameter data as output data.

[0054] It should be noted that the input data is preferably a combination of the flow passage attribute data and the fluid attribute data, but is not limited thereto, and the input data can also only include the flow passage attribute data or the fluid attribute data.

[0055] According to an embodiment of the present disclosure, the flow passage attribute data can include at least one of a flow passage equivalent diameter, a flow passage geometric shape, and a flow passage length. The fluid attribute data can include at least one of a fluid flow rate, a fluid wind direction, and a fluid flow volume.

[0056] According to an embodiment of the present disclosure, the deep learning model can include at least one of a multilayer perceptron (MLP), a convolutional neural network (CNN), a recurrent neural network (RNN), and a decision tree.

[0057] According to an embodiment of the present disclosure, the plurality of parameter data can be one-to-one corresponding to the plurality of deep learning models. For example, each deep learning model outputs one parameter data. It can also be that the plurality of parameter data is not one-to-one corresponding to the plurality of deep learning models. For example, the plurality of deep learning models combine to output one parameter data. Or for example, one deep learning model outputs a plurality of parameter data.

[0058] According to an embodiment of the present disclosure, the plurality of parameter data is generated based on the flow passage attribute data and the fluid attribute data of the fluid by using the plurality of deep learning models, so that each deep learning model becomes an expert model, and based on the different data types of the fluid attribute data and the flow passage attribute data, a deep learning model matched with the data type is used for processing, so that the plurality of parameter data is accurate and the accurate heat exchange data is determined.

[0059] According to an embodiment of the present disclosure, for the operation S220 as shown in Figure 2 Based on the first physical property data, the second physical property data, and the plurality of parameter data, the heat exchange data between the fluid and the heat exchange wall of the heat exchange device can be determined, which can include: based on the first physical property data, the second physical property data, and the plurality of parameter data, determining heat exchange intensity data between the fluid and the heat exchange wall. Based on the heat exchange intensity data, the heat exchange data between the fluid and the heat exchange wall is determined.

[0060] According to an embodiment of the present disclosure, the first physical property data can include at least one of a density of the heat exchange wall, a thermal conductivity of the heat exchange wall, a temperature of the heat exchange wall, a dynamic viscosity coefficient of the heat exchange wall, and a material of the heat exchange device. The second physical property data can include at least one of a density of the fluid, a thermal conductivity of the fluid, a dynamic viscosity coefficient of the fluid, a temperature of the fluid, a physical property state of the fluid such as a gas state or a liquid state, and a kind of the fluid such as water or oil.

[0061] According to an embodiment of the present disclosure, the heat exchange intensity data can be dimensionless data such as a Nusselt number, a friction coefficient, or the like. The heat exchange data can be heat of the heat exchange unit, enthalpy of the heat exchange unit, or a heat exchange coefficient of the heat exchange unit, or the like.

[0062] According to an embodiment of the present disclosure, determining the heat exchange intensity data between the fluid and the heat exchange wall based on the first physical property data, the second physical property data, and the plurality of parameter data can include adding, subtracting, multiplying, or dividing the first physical property data, the second physical property data, and the plurality of parameter data, but is not limited thereto, and can include performing a weighted summation operation to obtain the heat exchange intensity data between the fluid and the heat exchange wall.

[0063] According to an embodiment of the present disclosure, determining the heat exchange data between the fluid and the heat exchange wall based on the heat exchange intensity data can include inputting the heat exchange intensity data into a heat exchange data equation to obtain the heat exchange data between the fluid and the heat exchange wall.

[0064] According to an embodiment of the present disclosure, the heat exchange data equation can be an empirical equation fitted by a worker based on experimental data. However, it is not limited thereto. The heat exchange data equation can also be a derivation equation derived based on a mass conservation equation, an energy conservation equation, a momentum conservation equation, or the like. The target heat exchange data equation can be determined from a plurality of heat exchange data equations based on a data type of the heat exchange data. The target heat exchange data equation matches the data type of the heat exchange data.

[0065] Taking the heat exchange intensity data as a Nusselt number and the heat exchange data as a convective heat exchange coefficient as an example, the heat exchange data equation can include h = Nu * K / L. In the equation, h represents the convective heat exchange coefficient, K represents a thermal conductivity, L represents an equivalent length, and Nu represents the Nusselt number, which is a dimensionless number representing the degree of intensity of convective heat exchange.

[0066] Taking the heat exchange intensity data as a friction coefficient and the heat exchange data as a convective heat exchange coefficient as an example, the heat exchange data equation can include h = A * f. In the equation, h represents the convective heat exchange coefficient, A represents a constant, and f represents the friction coefficient of the heat exchange wall.

[0067] According to the embodiment of the present disclosure, the deep learning model is combined with the empirical equation or the derived equation, the mechanism and the deep learning model are used jointly, the accuracy of the heat exchange data is ensured, the data types of the heat exchange data are rich, and then the heat exchange data can be applied to various heat exchange devices of different types, so that the application range is wide.

[0068] According to the embodiment of the present disclosure, the heat exchange intensity data between the fluid and the heat exchange wall can be determined based on the first physical property data, the second physical property data and the plurality of parameter data, which can include: determining target physical property data based on the first physical property data and the second physical property data. The first target parameter data matching the data type of the first physical property data is determined from the plurality of parameter data. The heat exchange intensity data between the fluid and the heat exchange wall is determined based on the target physical property data and the first target parameter data.

[0069] According to the embodiment of the present disclosure, the target physical property data can be determined based on the first physical property data and the second physical property data, which can include: performing addition, subtraction, multiplication or division operations on the first physical property data and the second physical property data, but is not limited thereto, and can also perform weighted summation operations. As long as the operation mode can obtain the target physical property data based on the first physical property data and the second physical property data.

[0070] According to the embodiment of the present disclosure, the first target parameter data matching the data type of the first physical property data can be determined from the plurality of parameter data, but is not limited thereto, and the first target parameter data matching the data type of the second physical property data can also be determined from the plurality of parameter data.

[0071] According to the embodiment of the present disclosure, the heat exchange intensity data between the fluid and the heat exchange wall can be determined based on the target physical property data and the first target parameter data, which can include: the first target parameter data and the target physical property data can be added, subtracted, multiplied or divided, but is not limited thereto, and can also be weighted and summed. As long as the operation mode can obtain the heat exchange intensity data based on the target physical property data and the first target parameter data.

[0072] For example, the first physical property data and the second physical property data can be added, subtracted, multiplied or divided to obtain the target physical property data, the first target parameter data can be used as the power coefficient of the target physical property data, the target physical property data can be raised to the power of the first target parameter data to obtain the heat exchange intensity data. The first physical property data and the second physical property data can also be added or subtracted to obtain the target physical property data, the first target parameter data can be used as the weight coefficient of the target physical property data, and the first target parameter data and the target physical property data can be multiplied to obtain the heat exchange intensity data.

[0073] According to the embodiment of the present disclosure, the target parameter data can be determined from the plurality of parameter data according to the data type of the first physical property data, so that the heat exchange intensity data determination manner is targeted.

[0074] According to the embodiment of the present disclosure, based on the target physical property data and the first target parameter data, the heat exchange intensity data between the fluid and the heat exchange wall can be determined, which can include: determining environment data related to the heat exchange environment. The second target parameter data matching the data type of the environment data is determined from the plurality of parameter data. Based on the environment data, the second target parameter data, the target physical property data and the first target parameter data, the heat exchange intensity data is determined.

[0075] According to the embodiment of the present disclosure, the environment data can refer to the flow field data related to the convective heat exchange, and can be dimensionless data related to the fluid or heat exchange. The environment data can be determined by at least two of the fluid property data, the flow passage property data, the first physical property data and the second physical property data, and is used to represent the influence of the superposition of multiple factors in heat transfer.

[0076] According to the embodiment of the present disclosure, the environment data includes at least one of the Reynolds number, the Prandtl number, the Schmidt number and the Stanton number.

[0077] According to the embodiment of the present disclosure, the Prandtl number (Pr) can be defined as the ratio of the dynamic viscosity coefficient and the thermal diffusivity, or the ratio of the momentum transfer and the heat transfer effect, and is used to represent the comparison of the momentum diffusion and the heat diffusion capacity in the fluid. The Reynolds number (R) can be defined as the ratio of the viscous force and the inertial force of the fluid, and is used to distinguish the flow state of the fluid, such as laminar flow or turbulent flow. The Schmidt number (Sc) can be defined as the ratio of the kinematic viscosity coefficient and the diffusion coefficient, and is used to represent the relative thickness of the hydrodynamic layer and the mass transfer boundary layer of the fluid with both momentum diffusion and mass diffusion. The Stanton number (St) can be defined as the ratio of the heat transfer in the fluid and the heat capacity of the fluid, and is used to represent the heat transfer in the forced convection process.

[0078] According to the embodiment of the present disclosure, the second target parameter data matching the data type of the environment data can be determined from the plurality of parameter data based on the data type of the environment data. For example, the Prandtl number and the Reynolds number have different definitions and physical meanings, so the parameter data matched by the Prandtl number and the Reynolds number are different.

[0079] According to the embodiment of the present disclosure, based on the environment data, the second target parameter data, the target physical property data and the first target parameter data, the heat exchange intensity data can be determined, which can include: performing addition, subtraction, multiplication or division operations on the environment data, the second target parameter data, the target physical property data and the first target parameter data, but is not limited thereto, and can also perform weighted summation operation.

[0080] According to the embodiment of the present disclosure, the environmental data is taken into account, and various factors can be comprehensively considered, so that the determined heat exchange intensity data is accurate.

[0081] According to the embodiment of the present disclosure, based on the environmental data, the second target parameter data, the target physical property data and the first target parameter data, the heat exchange intensity data is determined, including: determining the first target data based on the target physical property data and the first target parameter data. The second target data is determined based on the environmental data and the second target parameter data. The heat exchange intensity data is determined based on the first target data and the second target data.

[0082] For example, the first target parameter data can be taken as a coefficient of the target physical property data, and the first target data is determined based on the target physical property data and the first target parameter data. For example, the first target parameter data is taken as a power coefficient of the target physical property data, and the target physical property data is raised to the power of the first target parameter data to obtain the first target data. The second target parameter data can be taken as a coefficient of the environmental data, and the second target data is determined. For example, the second target parameter data is taken as a power coefficient of the environmental data, and the environmental data is raised to the power of the second target parameter data to obtain the second target data. The first target data and the second target data are added, subtracted, multiplied or divided, etc. to obtain the heat exchange intensity data.

[0083] For another example, the first target parameter data can be added or subtracted from the target physical property data to obtain the first target data. The second target parameter data can be added or subtracted from the environmental data to obtain the second target data. The first target data and the second target data are multiplied or divided to obtain the heat exchange intensity data.

[0084] According to the embodiment of the present disclosure, the target physical property data is combined with the first target parameter data, and the environmental data is combined with the second target parameter data to determine the heat exchange intensity data, which can make the determination method of the heat exchange intensity data clear and the reasoning process clear.

[0085] According to an embodiment of the present disclosure, a mapping relationship between the data type of the first physical property data and the parameter data can be established in advance. According to the mapping relationship between the data type of the first physical property data and the parameter data, the first target parameter data matching the data type of the first physical property data is determined from the plurality of parameter data. A mapping relationship between the data type of the environmental data and the parameter data can also be established in advance. According to the mapping relationship between the data type of the environmental data and the parameter data, the second target parameter data matching the data type of the environmental data is determined from the plurality of parameter data. However, it is not limited thereto. A deep learning model, such as a Gate Recurrent Unit (GRU), can also be used to determine the matching relationship between the parameter data and the first physical property data or the environmental data. For example, the input data of the Gate Recurrent Unit is the plurality of parameter data, and the output data is a label representing the matching relationship between the parameter data and the first physical property data or the environmental data.

[0086] Figure 3 A schematic diagram for determining heat exchange intensity according to an embodiment of the present disclosure is shown.

[0087] As shown in Figure 3 , the first input data, such as the fluid property data 310 and the flow channel property data 320, can be input into the first Gate Recurrent Unit M310 to obtain the first output data. The plurality of first output data is used to represent the mapping relationship between the fluid property data and the flow channel property data as input data and each of the plurality of deep learning models M320. For example, the fluid property data and the flow channel property data can be remodeled as the first input data to obtain the second input data matching one or more of the plurality of deep learning models M320. The plurality of deep learning models M320 can be understood as a plurality of expert layers.

[0088] According to an embodiment of the present disclosure, the remodeling can refer to splitting the first input data into a plurality of sub-data, and each sub-data is taken as a second input data. However, it is not limited thereto. It can also refer to directly taking the first input data as the second input data.

[0089] As shown in Figure 3 , the second input data can be input into the plurality of deep learning models M320 to obtain the second output data, such as the plurality of parameter data 330. The plurality of parameter data 330 is taken as the third input data and input into the second Gate Recurrent Unit M330 to obtain the third output data 340. The third output data is used to represent the mapping relationship between the parameter data and the data type of the first physical property data and the mapping relationship between the parameter data and the data type of the environmental data.

[0090] According to an embodiment of the present disclosure, the first input data can also be input into the second gating cycle unit to obtain third output data. The input data can be any data that can determine the mapping relationship between the parameter data and the data type of the first physical property data and the mapping relationship between the parameter data and the data type of the environmental data.

[0091] According to another embodiment of the present disclosure, the heat exchange intensity equation can also be used for constraint, for example, the mapping relationship between the parameter data and the data type of the first physical property data and the mapping relationship between the parameter data and the data type of the environmental data are established in advance, and the second output data, for example, the plurality of parameter data, are directly taken as the final output data.

[0092] According to another embodiment of the present disclosure, based on the environmental data, the second target parameter data, the target physical property data and the first target parameter data, the heat exchange intensity data is determined, including: inputting the environmental data, the second target parameter data, the target physical property data and the first target parameter data into the heat exchange intensity equation to obtain the heat exchange intensity data.

[0093] According to an embodiment of the present disclosure, the heat exchange intensity equation can be an empirical equation fitted by a worker based on experimental data. However, it is not limited thereto. The heat exchange intensity equation can also be a derivation equation derived based on mass conservation, energy conservation, momentum conservation and the like. Based on the data type of the heat exchange intensity data, a target heat exchange intensity equation can be determined from a plurality of heat exchange intensity equations. The target heat exchange intensity equation can match the data type of the heat exchange intensity data.

[0094] According to an embodiment of the present disclosure, taking the Nusselt number as an example of the heat exchange intensity data, the heat exchange intensity equation can adopt the following formula (1).

[0095]

[0096] wherein, Nu represents the Nusselt number; A i represents the environmental data, including but not limited to the Reynolds number Re, the Prandtl number Pr, the Schmidt number Sc, the Stanton number St and the like; i represents the i th environmental data; a w，j represents the first physical property data, including but not limited to the density, the dynamic viscosity coefficient, the thermal conductivity and the like; w represents the first physical property category; j represents the first first physical property data or the first second physical property data; a b，j represents the second fluid physical property, including but not limited to the density, the dynamic viscosity coefficient, the thermal conductivity and the like; b represents the second physical property category; M d，i represents the i th second target parameter; M c，j represents the j th first target parameter; M represents a constant; Π i () represents multiplication.

[0097] According to an embodiment of the present disclosure, the heat exchange intensity equation is improved on the basis of a modified equation of heat exchange intensity data. The multiple parameter data output by the deep learning model can replace the constants in the modified equation, thereby obtaining the heat exchange intensity equation. By combining the modified equation with the deep learning model, the multiple parameter data of the multiple deep learning models are physically constrained by the modified equation, thereby generating a heat exchange intensity equation in the form of a mechanism-deep learning hybrid model. The multiple deep learning models can be used to process complex and variable fluid property data and pipe property data, and the modified equation can be used to provide experimental experience, thereby combining artificial intelligence with physical constraints, improving the accuracy of the heat exchange intensity data, and improving the generalization of the processing.

[0098] According to an optional embodiment of the present disclosure, the environment data, the second target parameter data, the target property data, and the first target parameter data are input into the heat exchange intensity equation to obtain the heat exchange intensity data, and the method can further include determining third target parameter data from the multiple parameter data. The environment data, the second target parameter data, the target property data, the first target parameter data, and the third target parameter data are input into the heat exchange intensity equation to obtain the heat exchange intensity data.

[0099] According to an embodiment of the present disclosure, taking the Nusselt number as an example of the heat exchange intensity data, the heat exchange intensity equation can also use the following formula (2). For example, the M constant in the above formula (1) can be modified to generate a variable parameter M a that replaces the constant M. a The variable parameter M i may vary based on changes in the fluid property data and the pipe property data.

[0100]

[0101] wherein Nu represents the Nusselt number; A i represents the environment data, including but not limited to the Reynolds number Re, the Prandtl number Pr, the Schmidt number Sc, the Stanton number St, etc.; i represents the i-th environment data; a w，j represents the first property data, including but not limited to the density, the dynamic viscosity coefficient, the thermal conductivity, etc.; w represents the first property category; j represents the first first property data or the first second property data; a b，j represents the second fluid property, including but not limited to the density, the dynamic viscosity coefficient, the thermal conductivity, etc.; b represents the second property category; M d，i represents the i-th second target parameter; M c，j represents the j-th first target parameter; M a represents the third target parameter; Π i () represents multiplication.

[0102] According to an embodiment of the present disclosure, the third target parameter data is determined from the plurality of parameter data, each of which can be obtained by processing the fluid attribute data and the pipeline attribute data by using the deep learning model. The third target parameter data varies with the attribute data of the fluid and the heat exchange device respectively. Therefore, on the basis of the environmental data, the second target parameter data, the target physical property data, and the first target parameter data, the third target parameter data as an additional coefficient can make the determined heat exchange intensity data accurate and flexible.

[0103] Figure 4 A flowchart of a method for determining heat exchange data according to an embodiment of the present disclosure is schematically shown.

[0104] As shown in Figure 4 , the pipeline attribute data 410 and the fluid attribute data 420 can be input into a plurality of deep learning models M410 to obtain a plurality of parameter data 440. Based on the first physical property data 451 and the second physical property data 452, the target physical property data 453 is determined. The first target parameter data 441 matching the data type of the first physical property data 451 is determined from the plurality of parameter data 440. The second target parameter data 442 matching the data type of the environmental data 430 is determined from the plurality of parameter data 440. The third target parameter data 443 is determined from the plurality of parameter data 440. The environmental data 430, the second target parameter data 442, the target physical property data 453, the first target parameter data 441, and the third target parameter data 443 are input into a heat exchange intensity equation M420 to obtain heat exchange intensity data. The heat exchange intensity data is input into a heat exchange data equation M430 to obtain heat exchange data 460.

[0105] Figure 5 A flowchart of a method for training a deep learning model according to an embodiment of the present disclosure is schematically shown.

[0106] As shown in Figure 5 , the method includes operations S510-S540.

[0107] In operation S510, sample pipeline attribute data of a heat exchange device and sample fluid attribute data of a fluid are input into a plurality of deep learning models to obtain a plurality of sample parameter data, wherein the fluid is a medium flowing in the heat exchange device.

[0108] In operation S520, based on sample first physical property data, sample second physical property data, and the plurality of sample parameter data, sample heat exchange data between the fluid and a heat exchange wall of the heat exchange device is determined, wherein the sample first physical property data is heat exchange related physical property data of the heat exchange wall; and the sample second physical property data is heat exchange related physical property data of the fluid.

[0109] At operation S530, a first loss value is determined based on the sample heat exchange data and the sample label, where the sample label is true data matched with the sample flow channel attribute data, the sample fluid attribute data, the sample first physical property data, and the sample second physical property data.

[0110] At operation S540, parameters of the plurality of deep learning models are adjusted based on the first loss value.

[0111] According to embodiments of the present disclosure, the fluid attribute data and the flow channel attribute data can be input into the trained plurality of deep learning models to obtain a plurality of parameter data. The parameter data can be used to represent heat exchange influencing factors between the solid and the fluid. The plurality of parameter data can represent the heat exchange influencing factors from different angles and different dimensions. In this way, the heat exchange data determined based on the first physical property data, the second physical property data, and the plurality of parameter data has high accuracy and wide application range.

[0112] According to embodiments of the present disclosure, the sample flow channel attribute data includes at least one of the following: flow channel equivalent diameter, flow channel geometric shape, and flow channel length. The sample fluid attribute data includes at least one of the following: fluid flow rate, fluid wind direction, and fluid flow. The sample first physical property data includes at least one of the following: density, thermal conductivity, and dynamic viscosity coefficient. The sample second physical property data includes at least one of the following: density, thermal conductivity, and dynamic viscosity coefficient.

[0113] According to embodiments of the present disclosure, for operation S520 as shown in Figure 5 Determining the sample heat exchange data between the fluid and the heat exchange wall based on the sample first physical property data, the sample second physical property data, and the plurality of sample parameter data can include: determining sample heat exchange intensity data between the fluid and the heat exchange wall based on the sample first physical property data, the sample second physical property data, and the plurality of sample parameter data. The sample heat exchange data between the fluid and the heat exchange wall is determined based on the sample heat exchange intensity data.

[0114] According to embodiments of the present disclosure, determining the sample heat exchange data between the fluid and the heat exchange wall based on the sample heat exchange intensity data can include: inputting the sample heat exchange intensity data into a sample heat exchange data equation to obtain the sample heat exchange data between the fluid and the heat exchange wall.

[0115] It should be noted that the sample heat exchange data equation can be trained as part of the deep learning model, and the parameters in the sample heat exchange data equation can be adjusted as the plurality of deep learning models. However, it is not limited thereto. The sample heat exchange data equation can also be used as a fixed equation and no longer be modified.

[0116] According to an embodiment of the present disclosure, determining the sample heat exchange intensity data between the fluid and the heat exchange wall based on the sample first physical property data, the sample second physical property data, and the plurality of sample parameter data can include determining sample target physical property data based on the sample first physical property data and the sample second physical property data. Determining sample first target parameter data matching a data type of the sample first physical property data from the plurality of sample parameter data. Determining the sample heat exchange intensity data between the fluid and the heat exchange wall based on the sample target physical property data and the sample first target parameter data.

[0117] According to an embodiment of the present disclosure, determining the sample heat exchange intensity data between the fluid and the heat exchange wall based on the sample target physical property data and the sample first target parameter data can include determining sample environment data related to the heat exchange environment. Determining sample second target parameter data matching a data type of the sample environment data from the plurality of sample parameter data. Determining the sample heat exchange intensity data based on the sample environment data, the sample second target parameter data, the sample target physical property data, and the sample first target parameter data.

[0118] According to an embodiment of the present disclosure, the sample environment data includes at least one of a Reynolds number, a Prandtl number, a Schmidt number, and a Stanton number.

[0119] According to an embodiment of the present disclosure, determining the sample heat exchange intensity data based on the sample environment data, the sample second target parameter data, the sample target physical property data, and the sample first target parameter data includes determining sample first target data based on the sample target physical property data and the sample first target parameter data. Determining sample second target data based on the sample environment data and the sample second target parameter data. Determining the sample heat exchange intensity data based on the sample first target data and the sample second target data.

[0120] According to another embodiment of the present disclosure, determining the sample heat exchange intensity data based on the sample environment data, the sample second target parameter data, the sample target physical property data, and the sample first target parameter data includes inputting the sample environment data, the sample second target parameter data, the sample target physical property data, and the sample first target parameter data into a sample heat exchange intensity equation to obtain the sample heat exchange intensity data.

[0121] According to an optional embodiment of the present disclosure, inputting the sample environment data, the sample second target parameter data, the sample target physical property data, and the sample first target parameter data into the sample heat exchange intensity equation to obtain the sample heat exchange intensity data can further include determining sample third target parameter data from the plurality of sample parameter data. Inputting the sample environment data, the sample second target parameter data, the sample target physical property data, the sample first target parameter data, and the sample third target parameter data into the sample heat exchange intensity equation to obtain the sample heat exchange intensity data.

[0122] According to an embodiment of the present disclosure, the operation S530 shown in the method for determining the first loss value based on the sample heat exchange data and the sample label can include: inputting the sample heat exchange data and the sample label into a first loss function to obtain the first loss value. The first loss function can include a cross-entropy loss function. Figure 5

[0123] According to an optional embodiment of the present disclosure, the first loss value can also be determined based on the sample heat exchange intensity data and the sample label matched with the sample heat exchange intensity data. For example, the sample heat exchange intensity data and the sample label matched with the sample heat exchange intensity data are input into a third loss function to obtain the first loss value. The third loss function includes a mean square error or an absolute error.

[0124] According to an embodiment of the present disclosure, taking the sample heat exchange intensity data as an example of the Nusselt number. The third loss function can include formula (3).

[0125]

[0126] wherein L1 represents the first loss value; Nu represents the sample heat exchange intensity data; and L represents the sample label matched with the sample heat exchange intensity data.

[0127] According to an embodiment of the present disclosure, the operation S540 shown in the method for adjusting the parameters of the plurality of deep learning models based on the first loss value can include the following operations. Figure 5 For example, the physical constraint condition is determined based on the sample first physical property data and the sample second physical property data. The second loss value is determined based on the physical constraint condition and the plurality of sample parameter data. The parameters of the plurality of deep learning models are adjusted based on the first loss value and the second loss value.

[0128] According to an embodiment of the present disclosure, in the case of inputting the sample environment data, the sample second target parameter data, the sample target physical property data and the sample first target parameter data into the sample heat exchange intensity equation to obtain the sample heat exchange intensity data, the second loss value can be determined by using the physical constraint condition matched with the sample heat exchange intensity data.

[0129] According to an embodiment of the present disclosure, the physical constraint condition can refer to using the parameters in the sample heat exchange intensity equation or the sample heat exchange data equation as a reference to constrain the plurality of sample parameter data output by the plurality of deep learning models.

[0130] According to an embodiment of the present disclosure, the mechanism-deep learning hybrid model can be established based on a modified equation of convective heat transfer.

[0131]

[0132] ​​​According to an embodiment of the present disclosure, a modified equation of the Nusselt number is taken as an example. One modified equation of the Nusselt number can be shown as equation (4). The sample heat exchange intensity equation shown as equation (2) can be established based on the modified equation of the Nusselt number.

[0133] Nu = 0.886 * (Re) 0.5 * (Pr) 0.5 ; equation (4)

[0134] According to an embodiment of the present disclosure, the parameters shown as equation (4), for example, 0.886 and 0.5, can be taken as the physical constraint sample labels, the sample parameter data output by the deep learning model can be taken as the prediction data, the physical constraint sample labels and the sample parameter data can be input into the second loss function to obtain the second loss value. The second loss function can include a cross-entropy loss function.

[0135] Taking equation (2) as an example of the sample heat exchange intensity equation, the second loss value can be determined by equation (5).

[0136] L2 = l(M' a , M' a，eq ) +∑ i l(M' b，i , M' b，i，eq ) +∑ j l(M' c，j , M' c，j，eq ); equation (5)

[0137] wherein, i represents the i-th sample environment data; j represents the j-th sample first physical property data or the j-th sample second physical property data; M' b，i represents the i-th sample second target parameter; M' c，j represents the j-th sample first target parameter; M' a represents the sample third target parameter; M' a，eq represents a constant matched with the sample third target parameter; M' b，i，eq represents a constant matched with the i-th sample second target parameter; M' c，j，eq represents a constant matched with the j-th sample first target parameter.

[0138] According to an embodiment of the present disclosure, the first loss value and the second loss value are weighted and summed to obtain a target loss value. Based on the target loss value, the parameters in the sample heat exchange intensity equation (for example, a mechanism-deep learning hybrid model) constructed by the plurality of deep learning models are adjusted, the mechanism-deep learning hybrid model is trained by using automatic differentiation and gradient descent method, and a trained sample heat exchange intensity equation is obtained.

[0139] According to the embodiment of the present disclosure, the data loss value determined by the sample heat exchange data and the sample label is taken as the first loss value, and the sample parameter data of the mechanism-deep learning hybrid model and the constant in the correction equation are taken as the second loss value, so that the data loss value and the mechanism loss value can be combined to improve the training precision.

[0140] Figure 6 A block diagram of a heat exchange data determination apparatus according to an embodiment of the present disclosure is schematically shown.

[0141] As shown in Figure 6 The heat exchange data determination apparatus 600 can include a data input module 610 and a data determination module 620.

[0142] The data input module 610 is configured to input the flow channel attribute data of the heat exchange device and the fluid attribute data of the fluid into a plurality of deep learning models to obtain a plurality of parameter data, wherein the fluid is a medium flowing in the heat exchange device.

[0143] The data determination module 620 is configured to determine heat exchange data between the fluid and the heat exchange wall of the heat exchange device based on the first physical property data, the second physical property data, and the plurality of parameter data, wherein the first physical property data is heat exchange related physical property data of the heat exchange wall, and the second physical property data is heat exchange related physical property data of the fluid.

[0144] According to the embodiment of the present disclosure, the first determination module includes a first determination submodule and a second determination submodule.

[0145] The first determination submodule is configured to determine heat exchange intensity data between the fluid and the heat exchange wall based on the first physical property data, the second physical property data, and the plurality of parameter data.

[0146] The second determination submodule is configured to determine the heat exchange data between the fluid and the heat exchange wall based on the heat exchange intensity data.

[0147] According to the embodiment of the present disclosure, the first determination submodule includes a first determination unit, a first matching unit, and a second determination unit.

[0148] The first determination unit is configured to determine target physical property data based on the first physical property data and the second physical property data.

[0149] The first matching unit is configured to determine first target parameter data matched with the data type of the first physical property data from the plurality of parameter data.

[0150] The second determination unit is configured to determine the heat exchange intensity data between the fluid and the heat exchange wall based on the target physical property data and the first target parameter data.

[0151] According to an embodiment of the present disclosure, the second determining unit comprises a first determining subunit, a second determining subunit and a third determining subunit.

[0152] The first determining subunit is configured to determine environment data related to the heat exchange environment.

[0153] The second determining subunit is configured to determine, from the plurality of parameter data, second target parameter data matching a data type of the environment data.

[0154] The third determining subunit is configured to determine, based on the environment data, the second target parameter data, target physical property data and first target parameter data, heat exchange intensity data.

[0155] According to an embodiment of the present disclosure, the third determining subunit comprises determining, based on the target physical property data and the first target parameter data, first target data; determining, based on the environment data and the second target parameter data, second target data; and determining, based on the first target data and the second target data, the heat exchange intensity data.

[0156] According to an embodiment of the present disclosure, the third determining subunit comprises inputting the environment data, the second target parameter data, the target physical property data and the first target parameter data into a heat exchange intensity equation to obtain the heat exchange intensity data.

[0157] According to an embodiment of the present disclosure, inputting the environment data, the second target parameter data, the target physical property data and the first target parameter data into the heat exchange intensity equation to obtain the heat exchange intensity data comprises determining, from the plurality of parameter data, third target parameter data; and inputting the environment data, the second target parameter data, the target physical property data, the first target parameter data and the third target parameter data into the heat exchange intensity equation to obtain the heat exchange intensity data.

[0158] According to an embodiment of the present disclosure, the flow channel attribute data comprises at least one of the following: flow channel equivalent diameter, flow channel geometric shape, flow channel length.

[0159] According to an embodiment of the present disclosure, the fluid attribute data comprises at least one of the following: fluid flow rate, fluid wind direction, fluid flow volume.

[0160] According to an embodiment of the present disclosure, the first physical property data comprises at least one of the following: density, thermal conductivity, dynamic viscosity coefficient.

[0161] According to an embodiment of the present disclosure, the second physical property data comprises at least one of the following: density, thermal conductivity, dynamic viscosity coefficient.

[0162] According to an embodiment of the present disclosure, the environment data comprises at least one of the following: Reynolds number, Prandtl number, Schmidt number, Stanton number.

[0163] Figure 7 A block diagram of a training apparatus of a deep learning model is shown.

[0164] As shown in Figure 7 The training apparatus 700 of the deep learning model includes a sample data input module 710, a sample determination module 720, a loss value determination module 730, and a parameter adjustment module 740.

[0165] The sample data input module 710 is configured to input sample flow channel attribute data of a heat exchange device and sample fluid attribute data of a fluid into a plurality of deep learning models to obtain a plurality of sample parameter data, wherein the fluid is a medium flowing in the heat exchange device.

[0166] The sample determination module 720 is configured to determine sample heat exchange data between the fluid and a heat exchange wall of the heat exchange device based on sample first physical property data, sample second physical property data, and the plurality of sample parameter data, wherein the sample first physical property data is heat exchange related physical property data of the heat exchange wall, and the sample second physical property data is heat exchange related physical property data of the fluid.

[0167] The loss value determination module 730 is configured to determine a first loss value based on the sample heat exchange data and a sample label, wherein the sample label is real data matched with the sample flow channel attribute data, the sample fluid attribute data, the sample first physical property data, and the sample second physical property data.

[0168] The parameter adjustment module 740 is configured to adjust parameters of the plurality of deep learning models based on the first loss value.

[0169] According to an embodiment of the present disclosure, the parameter adjustment module includes a constraint condition determination sub-module, a loss value determination sub-module, and a parameter adjustment determination sub-module.

[0170] The constraint condition determination sub-module is configured to determine a physical constraint condition based on the sample first physical property data and the sample second physical property data.

[0171] The loss value determination sub-module is configured to determine a second loss value based on the physical constraint condition and the plurality of sample parameter data.

[0172] The parameter adjustment determination sub-module is configured to adjust the parameters of the plurality of deep learning models based on the first loss value and the second loss value.

[0173] According to an embodiment of the present disclosure, the present disclosure further provides an electronic device, a readable storage medium, and a computer program product.

[0174] According to an embodiment of the disclosure, an electronic device comprises at least one processor; and a memory connected with the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to an embodiment of the disclosure.

[0175] According to an embodiment of the disclosure, a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to enable a computer to perform the method according to an embodiment of the disclosure.

[0176] According to an embodiment of the disclosure, a computer program product comprises a computer program, and the computer program, when executed by a processor, implements the method according to an embodiment of the disclosure.

[0177] Figure 8 A schematic block diagram of an example electronic device 800 that can be used to implement embodiments of the disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptops, desktops, tablets, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular telephones, smartphones, wearable devices, and other similar computing devices. The components shown here, their connections and relationships, and their functions, are meant to be examples only, and are not meant to limit implementations of the present disclosure described and / or claimed in this document.

[0178] As shown in Figure 8 The device 800 includes a computing unit 801 that can perform various appropriate actions and processes in accordance with a computer program stored in a read-only memory (ROM) 802 or a computer program loaded from a storage unit 808 into a random access memory (RAM) 803. Various programs and data required for the operation of the device 800 can also be stored in the RAM 803. The computing unit 801, the ROM 802, and the RAM 803 are connected to each other through a bus 804. An input / output (I / O) interface 805 is also connected to the bus 804.

[0179] Various components in the device 800 are connected to the I / O interface 805, including an input unit 806, such as a keyboard, a mouse, and the like; an output unit 807, such as various types of displays, speakers, and the like; the storage unit 808, such as a magnetic disk, an optical disk, and the like; and a communication unit 809, such as a network card, a modem, a wireless communication transceiver, and the like. The communication unit 809 allows the device 800 to exchange information / data with other devices through a computer network, such as the Internet, and / or various telecommunication networks.

[0180] The computing unit 801 can be various general and / or special purpose processing components with processing and computing capabilities. Some examples of the computing unit 801 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The computing unit 801 performs various methods and processes described above, such as the heat exchange data determination method. For example, in some embodiments, the heat exchange data determination method can be implemented as a computer software program that is tangibly embodied in a machine-readable medium, such as the storage unit 808. In some embodiments, part or all of the computer program can be loaded and / or installed onto the device 800 via the ROM 802 and / or the communication unit 809. When the computer program is loaded onto the RAM 803 and executed by the computing unit 801, one or more steps of the heat exchange data determination method described above can be performed. Alternatively, in other embodiments, the computing unit 801 can be configured to perform the heat exchange data determination method by any other appropriate means, such as by means of firmware.

[0181] Various implementations of the systems and techniques described above can be realized in digital electronic circuitry, integrated circuitry, a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), a system on a chip (SOC), a complex programmable logic device (CPLD), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include implementation in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which can be special or general purpose, coupled to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device.

[0182] Program code for carrying out methods of the present disclosure can be written in any combination of one or more programming languages. The program code can be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the program code, when executed by the processor or controller, produces a means for implementing the functions / acts specified in the flowcharts and / or block diagrams. The program code can be executed entirely on a machine, partially on a machine, partially on a machine as a stand-alone software package, partially on a machine and partially on a remote machine or entirely on a remote machine or server.

[0183] In the context of this disclosure, a machine-readable fluid can be a tangible fluid that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable fluid can be a machine-readable signal fluid or a machine-readable storage fluid. Machine-readable fluids can be, but are not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0184] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0185] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as a data server), or computing systems that include middleware components (e.g., an application server), or computing systems that include frontend components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with embodiments of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via any form or fluid digital data communication (e.g., a communication network). Examples of communication networks include local area networks (LANs), wide area networks (WANs), and the Internet.

[0186] Computer systems can include clients and servers. Clients and servers are generally located far apart and typically interact via communication networks. Client-server relationships are created by computer programs running on the respective computers and having a client-server relationship with each other. Servers can be cloud servers, distributed system servers, or servers incorporating blockchain technology.

[0187] It should be understood that the various forms of flow shown above can be used to reorder, add, or delete steps. For example, the steps described in the present disclosure can be performed in parallel, in series, or in a different order, as long as the desired results of the technology disclosed in the present disclosure can be achieved, which is not limited herein.

[0188] The specific implementation described above does not constitute a limitation on the protection scope of the present disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent replacements, and improvements made within the spirit and principles of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A method for determining heat exchange data, comprising: inputting flow channel attribute data of a heat exchange device and fluid attribute data of a fluid into a plurality of deep learning models to obtain a plurality of parameter data, wherein the fluid is a medium flowing in the heat exchange device, and the plurality of parameter data represent heat exchange influencing factors between a heat exchange wall of the heat exchange device and the fluid from different dimensions; determining, from the plurality of parameter data, first target parameter data matching a data type of first physical data, the first physical data being heat exchange related physical data of the heat exchange wall; determining, from the plurality of parameter data, second target parameter data matching a data type of environmental data, the environmental data being related to a heat exchange environment; inputting the environmental data, the second target parameter data, target physical data and the first target parameter data into a heat exchange intensity equation to obtain heat exchange intensity data, the target physical data being determined based on the first physical data and second physical data, the second physical data being heat exchange related physical data of the fluid; and determining, based on the heat exchange intensity data, heat exchange data between the fluid and the heat exchange wall.

2. The method of claim 1, wherein, The inputting the environmental data, the second target parameter data, the target physical data and the first target parameter data into the heat exchange intensity equation to obtain the heat exchange intensity data comprises: determining, from the plurality of parameter data, third target parameter data; inputting the environmental data, the second target parameter data, the target physical data, the first target parameter data and the third target parameter data into the heat exchange intensity equation to obtain the heat exchange intensity data.

3. The method of claim 1 or 2, wherein, The flow channel attribute data comprises at least one of the following: flow channel equivalent diameter, flow channel geometry, flow channel length; The fluid attribute data comprises at least one of the following: fluid flow rate, fluid wind direction, fluid flow volume; The first physical data comprises at least one of the following: density, thermal conductivity, dynamic viscosity coefficient; The second physical data comprises at least one of the following: density, thermal conductivity, dynamic viscosity coefficient; The environmental data comprises at least one of the following: Reynolds number, Prandtl number, Schmidt number, Stanton number. 4.A method for training a deep learning model, comprising: inputting sample flow channel attribute data of a heat exchange device and sample fluid attribute data of a fluid into a plurality of deep learning models to obtain a plurality of sample parameter data, wherein the fluid is a medium flowing in the heat exchange device, and the plurality of sample parameter data represent heat exchange influencing factors between a heat exchange wall of the heat exchange device and the fluid from different dimensions; determining, from the plurality of sample parameter data, sample first target parameter data matching a data type of sample first physical data, the sample first physical data being heat exchange related physical data of the heat exchange wall; determining, from the plurality of sample parameter data, sample second target parameter data matching a data type of sample environmental data, the sample environmental data being related to a heat exchange environment; inputting the sample environment data, the sample second target parameter data, sample target physical property data and the sample first target parameter data into a heat exchange intensity equation to obtain sample heat exchange intensity data, the sample target physical property data being determined based on the sample first physical property data and sample second physical property data, the sample second physical property data being physical property data of the fluid related to heat exchange; and determining sample heat exchange data between the fluid and the heat exchange wall based on the sample heat exchange intensity data; determining a first loss value based on the sample heat exchange data and a sample label, wherein the sample label is true data matched with the sample flow channel attribute data, sample fluid attribute data, the sample first physical property data and the sample second physical property data; and adjusting parameters of the plurality of deep learning models based on the first loss value.

5. The method of claim 4, wherein, The adjusting parameters of the plurality of deep learning models based on the first loss value comprises: determining a physical constraint condition based on the sample first physical property data and the sample second physical property data; determining a second loss value based on the physical constraint condition and the plurality of sample parameter data; and adjusting the parameters of the plurality of deep learning models based on the first loss value and the second loss value.

6. A heat exchange data determination apparatus, comprising: a data input module configured to input flow channel attribute data of a heat exchange device and fluid attribute data of a fluid into a plurality of deep learning models to obtain a plurality of parameter data, wherein the fluid is a medium flowing in the heat exchange device; and a data determination module configured to determine heat exchange data between the fluid and a heat exchange wall of the heat exchange device based on first physical property data, second physical property data and the plurality of parameter data, wherein the first physical property data is physical property data of the heat exchange wall related to heat exchange; and the second physical property data is physical property data of the fluid related to heat exchange; the first determination module comprises: a first determination sub-module configured to determine heat exchange intensity data between the fluid and the heat exchange wall based on the first physical property data, the second physical property data and the plurality of parameter data; and a second determination sub-module configured to determine the heat exchange data between the fluid and the heat exchange wall based on the heat exchange intensity data; the first determination sub-module comprises: a first determination unit configured to determine target physical property data based on the first physical property data and the second physical property data; a first matching unit configured to determine first target parameter data matched with a data type of the first physical property data from the plurality of parameter data; and a second determination unit configured to determine the heat exchange intensity data between the fluid and the heat exchange wall based on the target physical property data and the first target parameter data; the second determination unit comprises: a first determination sub-unit configured to determine environment data related to a heat exchange environment; a second determination sub-unit configured to determine second target parameter data matched with a data type of the environment data from the plurality of parameter data; and The third determining sub-unit is configured to input the environment data, the second target parameter data, the target physical property data, and the first target parameter data into a heat exchange intensity equation to obtain the heat exchange intensity data.

7. The apparatus of claim 6, wherein, The inputting the environment data, the second target parameter data, the target physical property data, and the first target parameter data into the heat exchange intensity equation to obtain the heat exchange intensity data comprises: determining third target parameter data from the plurality of parameter data; inputting the environment data, the second target parameter data, the target physical property data, the first target parameter data, and the third target parameter data into the heat exchange intensity equation to obtain the heat exchange intensity data.

8. The apparatus of claim 6 or 7, wherein, The flow channel attribute data comprises at least one of the following: flow channel equivalent diameter, flow channel geometric shape, and flow channel length. The fluid attribute data comprises at least one of the following: fluid flow rate, fluid wind direction, and fluid flow volume. The first physical property data comprises at least one of the following: density, thermal conductivity, and dynamic viscosity coefficient. The second physical property data comprises at least one of the following: density, thermal conductivity, and dynamic viscosity coefficient. The environment data comprises at least one of the following: Reynolds number, Prandtl number, Schmidt number, and Stanton number.

9. A device for training a deep learning model, comprising: a sample data input module configured to input sample flow channel attribute data of a heat exchange device and sample fluid attribute data of a fluid into a plurality of deep learning models to obtain a plurality of sample parameter data, wherein the fluid is a medium flowing in the heat exchange device, and the plurality of sample parameter data represent heat exchange influencing factors between a heat exchange wall of the heat exchange device and the fluid from different dimensions; a sample determining module configured to determine sample heat exchange data between the fluid and the heat exchange wall of the heat exchange device based on sample first physical property data, sample second physical property data, and the plurality of sample parameter data, wherein the sample first physical property data is heat exchange related physical property data of the heat exchange wall, and the sample second physical property data is heat exchange related physical property data of the fluid; a loss value determining module configured to determine a first loss value based on the sample heat exchange data and a sample label, wherein the sample label is true data matched with the sample flow channel attribute data, sample fluid attribute data, sample first physical property data, and sample second physical property data; and a parameter adjusting module configured to adjust parameters of the plurality of deep learning models based on the first loss value. The sample determining module is specifically configured to: determine sample first target parameter data matched with a data type of the sample first physical property data from the plurality of sample parameter data; determine sample second target parameter data matched with a data type of sample environment data from the plurality of sample parameter data, wherein the sample environment data is related to a heat exchange environment. inputting the sample environment data, the sample second target parameter data, sample target physical property data and the sample first target parameter data into a heat exchange intensity equation to obtain sample heat exchange intensity data, the sample target physical property data being determined based on the sample first physical property data and sample second physical property data; and determining sample heat exchange data between the fluid and the heat exchange wall based on the sample heat exchange intensity data.

10. The apparatus of claim 9, wherein, The parameter adjusting module comprises: a constraint condition determining sub-module configured to determine a physical constraint condition based on the sample first physical property data and the sample second physical property data; a loss value determining sub-module configured to determine a second loss value based on the physical constraint condition and the plurality of sample parameter data; and a parameter adjusting determining sub-module configured to adjust parameters of the plurality of deep learning models based on the first loss value and the second loss value. 11.An electronic device comprising: at least one processor; and a memory connected to the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1 to 5.

12. A non-transitory computer readable storage medium having stored thereon computer instructions, wherein, The computer instructions are used to enable the computer to perform the method according to any one of claims 1 to 5. 13.A computer program product comprising a computer program which, when executed by a processor, implements the method according to any one of claims 1 to 5.

Images