Plate heat exchanger temperature prediction method and system based on proxy model

Abstract

The invention provides a plate heat exchanger temperature prediction method and system based on an agent model, and relates to the technical field of plate heat exchanger performance prediction. According to the method, a finite element simulation model of the plate heat exchanger is established; generating a plurality of sample points in the design space of the key parameter variables, inputting the sample points into the finite element simulation model, and performing simulation calculation to obtain fluid domain temperature field data of the plate heat exchanger under different working conditions; taking the sample points obtained by sampling as input and the corresponding fluid domain temperature field data as output, constructing an experimental data set, and constructing an agent model of the temperature field of the plate heat exchanger based on the experimental data set; and inputting the working condition data under the current known working condition into the agent model to obtain a heat exchanger fluid domain temperature field predicted by the agent model under the to-be-measured working condition. The method can accurately predict the temperature distribution of the fluid domain of the heat exchanger under the condition of inputting the temperature and flow of the cold and hot water inlet of the plate heat exchanger.

Description

Technical Field

[0001] The present invention relates to the technical field of plate heat exchanger performance prediction, and in particular to a plate heat exchanger temperature prediction method and system based on an agent model. Background Art

[0002] As an important equipment in modern industrial production, plate heat exchangers are widely used in various industries that require heat exchange, such as machinery, electricity, chemical industry, and medicine. Plate heat exchangers are made of a series of metal sheets with a certain corrugated shape. Their design allows the fluid to flow alternately between the plates, thereby maximizing the heat transfer efficiency. Compared with traditional tube heat exchangers, plate heat exchangers have the advantages of high heat exchange efficiency, space saving, high pressure resistance, and long service life.

[0003] For many years, how to improve the heat transfer performance of plate heat exchangers has been one of the research directions of many scholars. Improving the heat transfer capacity of heat exchangers is to improve resource utilization.

[0004] However, plate heat exchangers, especially brazed plate heat exchangers, cannot be disassembled, and the internal fluid flow and temperature distribution are black box problems, so external detection methods have great limitations.

[0005] With the rapid development of hardware conditions, numerical simulation has become an important means to solve heat exchanger structural problems and verify design solutions in the analysis of heat transfer characteristics of heat exchangers. However, for complex engineering problems, it takes a lot of time to directly solve the original model. Summary of the invention

[0006] Purpose of the invention: To propose a temperature prediction method for a plate heat exchanger based on an agent model, and to propose a temperature prediction system for a plate heat exchanger for implementing the prediction method, so as to solve the above-mentioned problems existing in the prior art.

[0007] In a first aspect of the present invention, a temperature prediction method for a plate heat exchanger based on an agent model is proposed, and the steps are as follows:

[0008] S1. Establish a three-dimensional model of the plate heat exchanger, analyze the design requirements and operating conditions of the plate heat exchanger, and establish a finite element simulation model;

[0009] S2, determine the boundary conditions and key parameter variables of the plate heat exchanger during the simulation process;

[0010] S3. generating a plurality of sample points in the design space of the key parameter variables, inputting the sample points into the finite element simulation model, and performing simulation calculation to obtain the fluid domain temperature field data of the plate heat exchanger under different working conditions;

[0011] S4, taking the sample points obtained by sampling as input and the corresponding fluid domain temperature field data as output, constructing an experimental data set, and constructing a proxy model of the temperature field of the plate heat exchanger based on the experimental data set;

[0012] S5. Inputting the operating data under the current known operating conditions into the proxy model to obtain the fluid domain temperature field of the heat exchanger predicted by the proxy model under the working condition to be tested.

[0013] In a further embodiment, the plate heat exchanger must meet the following boundary conditions during the simulation process:

[0014] The fluid inlets on both the cold and hot sides are velocity inlets; the outlets are pressure outlets; the wall boundary condition is a no-slip wall; the plate material is 304 stainless steel; and the fluid is liquid water.

[0015] In a further embodiment, during the simulation of the plate heat exchanger, key parameter variables include: temperature and flow rate of the fluid inlets on the cold and hot sides.

[0016] In a further embodiment, establishing a finite element simulation model comprises the following steps:

[0017] The basic size parameters of the plate include the outer dimensions such as length and width, and the design dimensions such as corrugation type, corrugation inclination and corrugation depth.

[0018] After completing the modeling in the 3D software, import it into Fluent and extract the fluid domain through Boolean operations.

[0019] In the Ansys Meshing module, unstructured tetrahedral mesh is selected for meshing.

[0020] In a further embodiment, step S3 specifically includes:

[0021] Evenly distributing a number of sample points within the range of the key parameter variable;

[0022] Input the sample points into the finite element simulation model in groups for simulation, select a certain section of the cold and hot fluid domains after each group of simulations, and export several groups of temperature point data;

[0023] Several groups of temperature point data are aggregated to form the fluid domain temperature field data of the plate heat exchanger under different working conditions.

[0024] In a further embodiment, the mathematical form of the proxy model is:

[0025] Y(x)=β 0 +Z(x)

[0026] In the formula, β 0 is an unknown constant representing the expected value of Y(x); Z(x) is a function of σ2 is the variance and the random floating error with mean 0; for different values ​​within the design variable range, the random error is correlated, expressed as:

[0027] Cov[Z(x),Z(x ′ )]=σ 2 R(x,x ′ )

[0028] In the formula, R(x,x ′ ) are input variables x and x ′ The value of this function is only related to the difference between the sample points composed of different variables, and decreases as the difference increases.

[0029] In a further embodiment, the input variables x and x ′ The correlation function R(x,x ′ )Select the Gaussian function.

[0030] In a further embodiment, the hot and cold water inlet temperatures and flow rate data under the current known working conditions are input into the proxy model, and the temperature distribution under the working conditions to be tested is predicted by weighted calculation of the sample points with the minimum mean square error as the optimization target.

[0031] In a further embodiment, after obtaining the proxy model, the RMSE and MAPE values ​​of the proxy model may be further calculated; if the error between the predicted value and the true value of the proxy model exceeds a predetermined range, the proxy model is optimized until the model accuracy meets the requirements.

[0032] In a further embodiment, if the accuracy of the proxy model does not meet the requirements, the proxy model is updated by adding sample points using the EI point addition criterion;

[0033] The basic form of the EI bonus rule is:

[0034]

[0035] Where Φ and φ are the standard normal cumulative distribution function and the standard normal distribution probability density function, respectively; μ(x) is the predicted mean of the Kriging model at point x; σ(x) is the predicted variance of the Kriging model at point x.

[0036] In a further embodiment, the method of adding sample points and updating the proxy model by using the EI point adding criterion specifically includes:

[0037] In the design space of key parameter variables, the predicted mean μ(x) and variance σ(x) of each sample point are calculated, and then the EI value of each point is calculated according to the formula;

[0038] The point with the largest EI value is selected as the next sampling point; the new sampling point can be one or more; when there are multiple points, the sampling point is determined by calculating multiple EI values ​​in parallel;

[0039] Add new sampling points to the existing training data set and retrain the proxy model; calculate the new model accuracy RMSE and MAPE values ​​to determine whether the model meets the accuracy requirements;

[0040] If the proxy model error still exceeds the predetermined range, repeat the EI calculation and point addition, and continue to add new sample points until the accuracy of the proxy model meets the requirements.

[0041] A second aspect of the present invention provides a temperature prediction system for a plate heat exchanger, the system comprising a model building unit, a simulation execution unit, a prediction pre-processing unit, and a prediction output unit.

[0042] The model building unit is used to establish a three-dimensional model of the plate heat exchanger, analyze the design requirements and operating conditions of the plate heat exchanger, and establish a finite element simulation model.

[0043] The simulation execution unit is used to determine the boundary conditions and key parameter variables of the plate heat exchanger during the simulation process; generate a number of sample points in the design space of the key parameter variables, input the sample points into the finite element simulation model, and simulate and calculate to obtain the fluid domain temperature field data of the plate heat exchanger under different working conditions.

[0044] The prediction pre-processing unit takes the sample points obtained by sampling as input and the corresponding fluid domain temperature field data as output, constructs an experimental data set, and builds a proxy model of the plate heat exchanger temperature field based on the experimental data set.

[0045] The prediction output unit is used to input the operating condition data under the current known operating condition into the proxy model to obtain the temperature field of the heat exchanger fluid domain predicted by the proxy model under the working condition to be tested.

[0046] Beneficial effects: The present invention can effectively capture the complex change trend of the temperature field of the plate heat exchanger and achieve high-precision temperature prediction by constructing a simulation model of the cold and hot fluid domains of the plate heat exchanger, clarifying the value range of key parameter variables in the simulation process of the plate heat exchanger, using Latin hypercube sampling to design sample points in the design space, constructing a large number of sample data sets required for training the proxy model, evaluating the model accuracy, and introducing the EI point-adding criterion to iteratively optimize the initially constructed proxy model. Compared with traditional numerical simulation methods, this method greatly reduces the calculation time and resource consumption while ensuring the prediction accuracy, and provides prediction data for the temperature field prediction of the plate heat exchanger, thereby guiding the innovation and optimization of the actual process. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1It is a flow chart of a temperature prediction method of a plate heat exchanger based on an agent model in an embodiment of the present invention.

[0048] Figure 2 It is a three-dimensional model diagram of the heat exchange plate of the plate heat exchanger drawn in an embodiment of the present invention.

[0049] Figure 3 It is a temperature field prediction comparison diagram under a certain working condition in an embodiment of the present invention.

[0050] Figure 4 It is a distribution diagram of temperature field prediction error under a certain working condition in an embodiment of the present invention. DETAILED DESCRIPTION

[0051] In the following description, a large number of specific details are provided to provide a more thorough understanding of the present invention. However, it is apparent to those skilled in the art that the present invention can be implemented without one or more of these details. In other examples, in order to avoid confusion with the present invention, some technical features known in the art are not described.

[0052] Step 1: Analyze the plate design requirements and operating conditions of the plate heat exchanger, establish a three-dimensional model of the plate heat exchanger and establish a finite element simulation model.

[0053] Step 1.1: Determine the basic size parameters of the plate, including length, width and other dimensions, as shown in Table 1:

[0054] Table 1 Corrugated plate parameters

[0055] name parameter Material 304 Plate length 527mm Plate width 244mm Board spacing 1.20mm Corrugation Angle 60° Corrugation height 2.05mm Plate thickness 8.5mm Corner hole diameter 45mm

[0056] Step 1.2: Draw a 3D model in UG software and import it into Space Claim to extract the fluid domain;

[0057] Step 1.3: Perform unstructured tetrahedral meshing in Ansys Meshing.

[0058] Step 2: Determine the boundary conditions and key parameter variables and their corresponding value ranges during the plate heat exchanger simulation process.

[0059] Step 2.1: The boundary conditions for plate heat exchanger fluid simulation are set as shown in the table below:

[0060] The wall conditions were set to constant wall temperature and no-slip wall, including fluid-structure interaction conditions.

[0061] The material of the herringbone corrugated plate of the plate heat exchanger is set to 304 steel, and the material properties are shown in Table 2.

[0062] Table 2 Plate material related parameters

[0063] parameter Numeric unit density 7930 <![CDATA[Kg / m 3 ]]> Specific heat 500 J / (kg·K) Thermal conductivity 16.3 W / (m·K)

[0064] The fluid medium is set to liquid water, and the material properties are shown in Table 3

[0065] Table 3 Physical parameters of fluid medium

[0066] Physical properties Numeric unit density 998.2 <![CDATA[Kg / m 3 ]]> Specific heat 4182 J / (kg·K) Thermal conductivity 0.6 W / (m·K)

[0067] The inlet boundary condition is set as velocity inlet, with initial velocity only in the positive direction along the y-axis, and the heat exchanger inlet temperature is considered to be equal to the initial temperature set for the fluid.

[0068] The outlet boundary condition is set to pressure outlet, and the pressure of the given flow outlet is 0, indicating that there is no backflow at the outlet.

[0069] Step 2.2: The variable parameters include cold water inlet temperature, cold water inlet speed, hot water inlet temperature, and hot water inlet speed. The value ranges are shown in Table 4 below:

[0070] Table 4 Variable parameter variables

[0071] Variable Name unit scope Cold water inlet velocity m / s 0～3.5 Cold water inlet temperature ℃ 15～35 Hot water inlet speed m / s 0～3.5 Hot water inlet temperature ℃ 25～45

[0072] Step 3: Use the Latin hypercube sampling method to generate several sample points in the parameter variable design space, input the sample point parameters into the simulation model, and use the numerical simulation method to calculate the fluid domain temperature field data of the plate heat exchanger under different working conditions.

[0073] Step 3.1: Latin Hypercube Sampling Design 100 sample points according to the pre-requirements, and divide the value range of each variable into 100 intervals evenly. For each variable value, there is only one sample point selected in each interval. Combine the sample points of all variables to form a complete sample. Ensure that the distribution of samples is uniform in each dimension, reducing the clustering of samples.

[0074] Step 3.2: Input the extracted 100 sample points into the simulation model to obtain the corresponding temperature field data of the output.

[0075] Step 4: Using the sample point data obtained by sampling as input and the corresponding temperature field data as output, construct an experimental data set, 80% as a training data set and 20% as a test data set, and build a Kriging proxy model of the temperature field of the plate heat exchanger based on the data set.

[0076] Step 4.1: The basic form of the Kriging surrogate model is:

[0077] Y(x)=β 0 +Z(x)

[0078] In the above formula, β 0 is an unknown constant representing the expected value of Y(x); Z(x) is a function of σ 2 is the variance, a random floating error with a mean of 0. For different values ​​within the design variable range, these random errors have a certain correlation, expressed as:

[0079] Cov[Z(x),Z(x ′ )]=σ 2 R(x,x ′ )

[0080] In the formula, R(x,x ′ ) are input variables x and x ′ The value of this function is only related to the difference between the sample points composed of different variables, and decreases as the difference increases.

[0081] Step 4.2: In the Kriging surrogate model, the Gaussian function is selected as its correlation function.

[0082] Step 4.3: Based on the input operating condition data and the temperature field data obtained by simulation, the temperature distribution under unknown operating conditions is predicted by performing weighted calculation on the sample points and taking the minimum mean square error as the optimization goal.

[0083] Step 5: Calculate the RMSE and MAPE values ​​of the Kriging model. If there is a large error between the predicted value of the surrogate model and the true value, optimize the surrogate model until the model accuracy meets the requirements.

[0084] Step 5.1: RMSE represents the root mean square error between the model prediction value and the true value, and MAPE represents the mean absolute percentage error between the model prediction value and the true value. The smaller the values ​​of the two, the better the model accuracy.

[0085] Step 5.2 The basic form of the EI scoring criteria is:

[0086]

[0087] Where: Φ and φ are the standard normal cumulative distribution function and the standard normal distribution probability density function, respectively. μ(x) is the predicted mean of the Kriging model at point x. σ(x) is the predicted variance of the Kriging model at point x.

[0088] Step 5.3 In the parameter design space, calculate the predicted mean μ(x) and variance σ(x) of each sample point, and then calculate the EI value of each point according to the formula.

[0089] Step 5.4 In the design space, select the point with the largest EI value as the next sampling point. The new sampling point can be one or more. When there are multiple sampling points, the sampling point can be determined by calculating multiple EI values ​​in parallel.

[0090] Step 5.5: Add the new sampling points to the existing training data set and retrain the Kriging proxy model. Calculate the new model accuracy evaluation index to determine whether the model meets the accuracy requirements.

[0091] Step 5.6 If the model error is still large, repeat the above EI value calculation and point addition steps, and continue to add new sample points until the accuracy of the proxy model meets the requirements.

[0092] Step 6: For the working condition to be tested, input the cold and hot water inlet temperature, flow rate and other data to obtain the temperature field of the heat exchanger fluid domain predicted by the Kriging proxy model.

[0093] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0094] The plate heat exchanger temperature prediction method disclosed in the above embodiment can be presented in the form of a software system. It is referred to as a "plate heat exchanger temperature prediction system" herein, hereinafter referred to as a prediction system or system. The prediction system can be composed of a model building unit, a simulation execution unit, a prediction pre-processing unit, and a prediction output unit.

[0095] The model building unit is used to establish a three-dimensional model of the plate heat exchanger, analyze the design requirements and operating conditions of the plate heat exchanger, and establish a finite element simulation model.

[0096] The simulation execution unit is used to determine the boundary conditions and key parameter variables of the plate heat exchanger during the simulation process; a number of sample points are generated in the design space of the key parameter variables, and the sample points are input into the finite element simulation model, and the fluid domain temperature field data of the plate heat exchanger under different working conditions is obtained by simulation calculation.

[0097] The prediction pre-processing unit takes the sample points obtained by sampling as input and the corresponding fluid domain temperature field data as output to construct an experimental data set, and then builds a proxy model of the temperature field of the plate heat exchanger based on the experimental data set.

[0098] The operating condition data under the current known operating condition is input into the proxy model by using the prediction output unit to obtain the temperature field of the heat exchanger fluid domain predicted by the proxy model under the working condition to be tested.

[0099] When the system is running, it can automatically execute all the process steps of the plate heat exchanger temperature prediction method disclosed in the above embodiment, which will not be described in detail here.

[0100] In addition, the plate heat exchanger temperature prediction method disclosed in the above embodiment can be embedded in an electronic device for operation. The electronic device includes a processor, a memory, a communication interface and a communication bus. The processor, the memory and the communication interface communicate with each other through the communication bus. The memory is used to store at least one executable instruction, and the executable instruction enables the processor to execute all the process steps of the plate heat exchanger temperature prediction method disclosed in the above embodiment, which will not be repeated here.

[0101] The electronic device may also communicate with one or more external devices (e.g., keyboards, pointing devices, Bluetooth devices, etc.), may communicate with one or more devices that enable a user to interact with the electronic device, and / or may communicate with any device that enables the electronic device to communicate with one or more other computing devices (e.g., routers, modems, etc.). This communication may be performed through an input / output (I / O) interface. In addition, the electronic device may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) through a network adapter. The network adapter communicates with other modules of the electronic device through a bus. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc.

[0102] Furthermore, the method for demarcating the priority protection area for biodiversity maintenance function disclosed in the above embodiment can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiment can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0103] As described above, although the present invention has been shown and described with reference to specific preferred embodiments, it should not be construed as limiting the present invention itself. Various changes may be made to it in form and detail without departing from the spirit and scope of the present invention as defined in the appended claims.

Claims

1. A temperature prediction method for a plate heat exchanger based on an agent model, characterized in that: The steps include: S1. Establish a three-dimensional model of the plate heat exchanger, analyze the design requirements and operating conditions of the plate heat exchanger, and establish a finite element simulation model; S2, determine the boundary conditions and key parameter variables of the plate heat exchanger during the simulation process; S3. generating a plurality of sample points in the design space of the key parameter variables, inputting the sample points into the finite element simulation model, and performing simulation calculation to obtain the fluid domain temperature field data of the plate heat exchanger under different working conditions; S4, taking the sample points obtained by sampling as input and the corresponding fluid domain temperature field data as output, constructing an experimental data set, and constructing a proxy model of the temperature field of the plate heat exchanger based on the experimental data set; S5. Inputting the operating data under the current known operating conditions into the proxy model to obtain the fluid domain temperature field of the heat exchanger predicted by the proxy model under the working condition to be tested.

2. The plate heat exchanger temperature prediction method based on the proxy model according to claim 1, characterized in that: During the simulation of the plate heat exchanger, the following boundary conditions must be met at the same time: The fluid inlets on both the cold and hot sides are velocity inlets; the outlets are pressure outlets; the wall boundary condition is a no-slip wall; the plate material is 304 stainless steel; and the fluid is liquid water.

3. The plate heat exchanger temperature prediction method based on the proxy model according to claim 2 is characterized in that: During the simulation of the plate heat exchanger, the key parameter variables include: The temperature and flow rate of the fluid inlets on the cold and hot sides.

4. The plate heat exchanger temperature prediction method based on the proxy model according to claim 1, characterized in that: Step S3 specifically includes: Evenly distributing a number of sample points within the range of the key parameter variable; Input the sample points into the finite element simulation model in groups for simulation, select a certain section of the cold and hot fluid domains after each group of simulations, and export several groups of temperature point data; Several groups of temperature point data are aggregated to form the fluid domain temperature field data of the plate heat exchanger under different working conditions.

5. The plate heat exchanger temperature prediction method based on the proxy model according to claim 1, characterized in that: The mathematical form of the proxy model is: Y(x)=β0+Z(x) Where β0 is an unknown constant representing the expected value of Y(x); Z(x) is a function of σ 2 is the variance and the random floating error with mean 0; for different values ​​within the design variable range, the random error is correlated, expressed as: Cov[Z(x),Z(x ′ )]=σ 2 R(x,x ′ ) In the formula, R(x,x ′ ) are input variables x and x ′ The value of this function is only related to the difference between the sample points composed of different variables, and decreases as the difference increases; Input variables x and x ′ The correlation function R(x,x ′ )Select the Gaussian function.

6. The plate heat exchanger temperature prediction method based on the proxy model according to claim 5, characterized in that: The cold and hot water inlet temperatures and flow rate data under the current known working conditions are input into the proxy model, and the temperature distribution under the working conditions to be tested is predicted by weighted calculation of the sample points with the minimum mean square error as the optimization target.

7. The plate heat exchanger temperature prediction method based on the proxy model according to claim 1, characterized in that: Also includes: Calculate the RMSE and MAPE values ​​of the surrogate model; If the error between the predicted value and the true value of the proxy model exceeds the predetermined range, the proxy model is optimized until the model accuracy meets the requirements.

8. The plate heat exchanger temperature prediction method based on the proxy model according to claim 7, characterized in that: If the accuracy of the proxy model does not meet the requirements, the EI point addition criterion is used to increase sample points and update the proxy model; The basic form of the EI bonus rule is: Where Φ and φ are the standard normal cumulative distribution function and the standard normal distribution probability density function, respectively; μ(x) is the predicted mean of the Kriging model at point x; σ(x) is the predicted variance of the Kriging model at point x.

9. The plate heat exchanger temperature prediction method based on the proxy model according to claim 8, characterized in that: The method of using the EI point adding criterion to increase sample points and update the proxy model specifically includes: In the design space of key parameter variables, the predicted mean μ(x) and variance σ(x) of each sample point are calculated, and then the EI value of each point is calculated according to the formula; The point with the largest EI value is selected as the next sampling point; the new sampling point can be one or more; when there are multiple points, the sampling point is determined by calculating multiple EI values ​​in parallel; Add new sampling points to the existing training data set and retrain the proxy model; calculate the new model accuracy RMSE and MAPE values ​​to determine whether the model meets the accuracy requirements; If the proxy model error still exceeds the predetermined range, repeat the EI calculation and point addition, and continue to add new sample points until the accuracy of the proxy model meets the requirements.

10. A plate heat exchanger temperature prediction system, used for automatically executing the plate heat exchanger temperature prediction method based on the agent model as described in any one of claims 1 to 9; characterized in that: The plate heat exchanger temperature prediction system includes: Model building unit; the model building unit is used to establish a three-dimensional model of the plate heat exchanger, analyze the design requirements and operating conditions of the plate heat exchanger, and establish a finite element simulation model; A simulation execution unit; the simulation execution unit is used to determine the boundary conditions and key parameter variables of the plate heat exchanger during the simulation process; generate a number of sample points in the design space of the key parameter variables, input the sample points into the finite element simulation model, and simulate and calculate to obtain the fluid domain temperature field data of the plate heat exchanger under different working conditions; Prediction pre-processing unit; the prediction pre-processing unit takes the sample points obtained by sampling as input and the corresponding fluid domain temperature field data as output, constructs an experimental data set, and constructs a proxy model of the plate heat exchanger temperature field based on the experimental data set; Prediction output unit: The prediction output unit is used to input the operating data under the current known operating conditions into the proxy model to obtain the temperature field of the heat exchanger fluid domain predicted by the proxy model under the working conditions to be tested.

Images