A transient cavity temperature proxy model of a heat flow coupling system and a construction method thereof

Abstract

The present application relates to a kind of flow heat coupling system transient cavity temperature proxy model and its construction method, belong to temperature detection technical field, solve the problem of low precision, slow, poor prediction ability and universality in prior art complex flow heat coupling system transient cavity temperature acquisition difficulty, degree, the proxy model of the present application will complex flow heat coupling system highly model into simple lumped flat heat transfer physical model, can quickly and accurately obtain the internal cavity temperature of complex flow heat coupling system, greatly simplify the difficulty of complex flow heat coupling system transient cavity temperature acquisition, save cost, efficiency and accuracy are high, while simple operation.

Description

Technical Field

[0001] This invention belongs to the field of temperature detection technology, and relates to a transient cavity temperature proxy model for a fluid-thermal coupling system and its construction method. Background Technology

[0002] For mechanical equipment with internal cavity structures, the internal cavity structure is usually quite complex. For the internal cavity structure, one side is usually high-temperature gas and the other side is low-temperature cold gas, which involves complex convective heat transfer problems. Therefore, obtaining the cavity temperature of the internal cavity structure is beneficial to determining the working state of the entire mechanical equipment, and it is also an important basis for the structural strength and life design analysis of mechanical equipment. For the acquisition of transient cavity temperature in complex fluid-thermal coupling systems, there are four main existing technologies: (1) CFD transient calculation method; (2) experimental method; (3) neural network method; (4) fluid network model method.

[0003] The existing technology has at least the following problems:

[0004] (1) Complex operation. CFD transient calculation method requires a lot of work such as fine modeling, mesh generation, and preprocessing; the design and manufacture of test bench is difficult; the fluid network-based thermal analysis method also requires to build a reasonable fluid network for the complex internal structure of fluid machinery, which usually has a large number of fluid elements.

[0005] (2) Long cycle and heavy resource consumption. Complex fluid-thermal coupling systems typically have numerous small internal structures and complex local flows. Fluid-thermal coupling calculations using CFD transient methods require detailed modeling, resulting in huge mesh sizes and heavy computational resource consumption. The calculation cycle of CFD transient methods is too long. If an experimental approach is adopted, the design, manufacturing, testing, and experimentation of the test bench require a long period of time, resulting in heavy financial consumption. Furthermore, it is difficult to test the internal cavity temperature of some fluid machinery rotating chambers.

[0006] (3) Large data requirements and difficulty in obtaining complete boundary conditions. Whether using CFD transient calculation methods for transient fluid-thermal coupling calculations or fluid network methods, sufficient and complete boundary conditions are required, and the accuracy of the boundary conditions will greatly affect the accuracy of cavity temperature calculations. Neural networks also require sufficient data to complete network training. The difficulty in obtaining data in actual use makes the above methods less practical.

[0007] (4) Poor generalizability: For cavity temperatures in complex fluid-thermal coupled systems, the fluid network method largely relies on empirical formulas for wind resistance and heat transfer after coefficient correction based on test data. This unidirectional coupling via empirical formulas lacks the versatility of a true fluid-thermal coupled model. It is often impossible to apply the same set of correction coefficients to different operating conditions, thus failing to predict cavity temperatures. The neural network method predicts cavity temperatures in fluid-thermal network systems using limited known experimental parameters and their variation patterns, without considering the influence of internal fluid-solid heat transfer. It cannot reflect the true physical model and has certain limitations in cavity temperature prediction, making it difficult to promote and apply. Summary of the Invention

[0008] Based on the above analysis, this invention provides a transient cavity temperature proxy model for a fluid-thermal coupling system and its construction method to solve the problems of high difficulty, low accuracy, slow speed, poor predictive ability and poor versatility in obtaining transient cavity temperatures in existing fluid-thermal coupling systems.

[0009] On one hand, the method for constructing a transient cavity temperature proxy model for a fluid-thermal coupled system according to the present invention includes the following specific steps:

[0010] The internal cavity structure of the fluid-thermal coupling system is compressed perpendicular to the flow direction of the first and second flow paths to obtain a flat plate internal cavity structure.

[0011] The differential heat conduction dx of a infinitesimal element of length x along the length of the plate in the direction of gas flow is:

[0012]

[0013] Where x is the length of the flat plate along the gas flow direction.

[0014] The length of the plate; ρ(x) is the density of the plate at length x; c is the specific heat capacity of the plate; T(x) is the temperature of the plate at length x; τ is the detection time; λ is the thermal conductivity of the plate; S is the source term;

[0015] For the source term S, the expression is:

[0016] S=(T f1 (x)-T(x)) h1(x)tdx+(T f2 (x)-T(x)) h2(x)tdx; (2)

[0017] Among them, T f h1(x) is the fluid temperature in the first flow path at a length x of the plate, h1(x) is the convective heat transfer coefficient in the first flow path at a length x of the plate, and T f h2(x) is the fluid temperature in the second flow path at a plate length of x, and h2(x) is the convective heat transfer coefficient of the second flow path at a plate length of x.

[0018] The lumped expression of equation (1) is:

[0019]

[0020] Integrating equation (3) along the length of the plate in the direction of gas flow:

[0021]

[0022] The energy equations for the first and second flow paths on both sides of the plate are as follows:

[0023]

[0024]

[0025] in, For the inflow of the first flow path, T f 1,in T is the inlet temperature of the first flow path. f 1,out The outlet temperature of the first flow path; T represents the inlet flow rate of the second flow path. f 2,in T is the inlet temperature of the second flow path. f 2,out The outlet temperature of the second flow path; C P The specific heat capacity at constant pressure of the fluid;

[0026] Combining equations (4), (5), and (6), we obtain the flat plate model:

[0027]

[0028] Where, ρ * For modeled flat plate density; T f 1 * To model the fluid temperature of the first flow path; T f 2 * To model the fluid temperature of the second flow path; T * h1 is the temperature of the molded plate. * h2 is the convective heat transfer coefficient of the first flow path. * V is the convective heat transfer coefficient of the modeled second flow path; V is the volume of the flat plate;

[0029] Model the flat plate to obtain a single-sided heat transfer model:

[0030]

[0031] in, The lumped convective heat transfer coefficient is... The total convective heat transfer area; The temperature of the lumped fluid;

[0032] From equations (7) and (8), we get:

[0033]

[0034]

[0035] Arrange equations (9) and (10) to obtain the proxy model of the internal cavity structure of the fluid-thermal coupling system after modeling:

[0036]

[0037] Where α is the equivalent time constant; β1 is the first Yellow River index; and β2 is the second Yellow River index.

[0038] Optionally, obtain the minimum required parameter set and the transient cavity temperature standard value;

[0039] Based on the transient cavity temperature standard value, a surrogate model of the internal cavity structure of the fluid-thermal coupling system is trained using the minimum requirement parameter set, and the surrogate model of the internal cavity structure of the fluid-thermal coupling system is continuously converged.

[0040] On the other hand, the present invention also provides a transient cavity temperature proxy model for a fluid-thermal coupling system, wherein the proxy model is as follows:

[0041]

[0042] Where α is the equivalent time constant; T * τ is the temperature of the molding plate; τ is the detection time. β1 represents the total fluid temperature; β2 represents the first Yellow River index; T represents the second Yellow River index. f 1,in T is the inlet temperature of the first flow path. f 1,out T represents the outlet temperature of the first flow path. f 2,in T is the inlet temperature of the second flow path. f 2,out T is the outlet temperature of the second flow path. f 1 * To model the fluid temperature in the first flow path, T f 2 * To model the fluid temperature of the second flow path.

[0043] Optionally, the surrogate model is based on compressing the internal cavity structure of the fluid-thermal coupling system perpendicular to the flow direction of the first and second flow paths to obtain a flat plate internal cavity structure; the flat plate model is:

[0044]

[0045] Where, ρ * For modeled flat plate density; h1 * To model the convective heat transfer coefficient of the first flow path, h1 * =h1(η);h2 * V is the convective heat transfer coefficient of the modeled second flow path; V is the volume of the plate.

[0046] Compared with the prior art, the present invention can achieve at least one of the following beneficial effects:

[0047] (1) The proxy model of the present invention highly modeled the complex fluid-thermal coupling system into a simple lumped plate heat transfer physical model, which can quickly and accurately obtain the internal cavity temperature of the complex fluid-thermal coupling system.

[0048] (2) The proxy model of the present invention is a closed proxy model, which can accurately obtain the internal cavity temperature of a complex flow-thermal coupling system by using only a few high-precision boundaries (such as the inlet temperature of each flow path, rotation speed, etc.), greatly simplifying the difficulty of obtaining the transient cavity temperature of a complex flow-thermal coupling system, saving costs, and having high efficiency and accuracy, while also being simple to operate.

[0049] (3) The proxy model of the present invention can quickly obtain the values ​​of each modeling parameter reflecting the local thermal equilibrium by adjusting four dimensionless parameters. Attached Figure Description

[0050] The accompanying drawings are for illustrative purposes only and are not intended to limit the scope of the invention.

[0051] Figure 1 (a)-(b) are schematic diagrams of the method of the present invention for flat plate modeling of the internal cavity structure of the thermal fluid coupling system;

[0052] Figure 2 This is a schematic diagram of the flat plate after the flat plate model of the internal cavity structure of the convection thermal coupling system of the present invention;

[0053] Figure 3 (a)-(b) are schematic diagrams of the dual-flow heat exchange lumped into a single-flow heat exchange in the method of the present invention;

[0054] Figure 4 This is a flowchart illustrating the proxy model construction process of the present invention.

[0055] Figure 5 A schematic diagram of the typical air system structure of the CFM56 aero engine used as an example;

[0056] Figure 6 This is a comparison chart of training values ​​and standard values ​​during the cavity temperature training process of the example proxy model of the present invention.

[0057] Figure 7 This is a comparison chart of the cavity temperature prediction value and the standard value of the example proxy model of the present invention. Detailed Implementation

[0058] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments.

[0059] A specific embodiment of the present invention, such as Figure 1-7 A method for constructing a transient cavity temperature proxy model for a fluid-thermal coupled system is disclosed, which specifically includes the following steps:

[0060] Step 1: Obtain the minimum required parameter set and the standard value of transient cavity temperature;

[0061] Training data for fluid machinery (i.e., complex fluid-thermal coupling systems) is acquired and divided into transient cavity temperature standard values ​​(i.e., true cavity temperatures) and a minimum requirement parameter set. The minimum requirement parameter set consists of the input parameters required by the surrogate model, such as the inlet temperature, flow rate, and convective heat transfer coefficient of each flow path, as well as the volume, specific heat capacity, and density of the internal cavity structure of the fluid-thermal coupling system.

[0062] Step 2: Model the flat plate structure of the internal cavity of the thermal fluid coupling system;

[0063] It is understandable that, such as Figure 1 As shown in (a), the internal cavity structure of a flow-thermal coupling system is typically considered in a two-dimensional plane as a complex solid composed of a plate structure of a certain thickness, reinforcing ribs, and protrusions. One side of this complex solid is a high-temperature gas flow path, and the other side is a low-temperature gas flow path. To obtain the internal cavity temperature of a certain flow path in the flow-thermal coupling system, such as... Figure 1 (a) shows the cavity temperature measurement point on the low-temperature gas side. This cavity temperature is partly affected by its own cold gas flow path, partly by solid heat exchange, and simultaneously by the flow path on the high-temperature gas side. By compressing the internal structure of the fluid-thermal coupling system perpendicular to the flow directions of the high-temperature and low-temperature gases, a non-homogeneous plate state with heat exchange with the gases on both sides is obtained, as shown below. Figure 1 As shown in (b).

[0064] The flat plate state after the internal cavity structure is modeled as follows Figure 2 As shown, H is the height of the plate along the compression direction, L is the length of the plate along the gas flow direction, and t is the width of the plate perpendicular to both the height and length directions. The differential heat conduction dx of a infinitesimal element with length x along the length direction of the plate along the gas flow direction is:

[0065]

[0066] Where x is the length of the plate along the gas flow direction; ρ(x) is the plate density at length x; c is the specific heat capacity of the plate; T(x) is the plate temperature at length x; τ is the detection time; λ is the thermal conductivity of the plate; and S is the source term.

[0067] For the source term S, the expression is:

[0068] S=(T f1 (x)-T(x)) h1(x)tdx+(T f2 (x)-T(x)) h2(x)tdx; (2)

[0069] Among them, T f h1(x) is the fluid temperature in the first flow path at a length x of the plate, h1(x) is the convective heat transfer coefficient in the first flow path at a length x of the plate, and T f h2(x) is the fluid temperature in the second flow path at a plate length of x, and h2(x) is the convective heat transfer coefficient of the second flow path at a plate length of x.

[0070] For the infinitesimal element at length x of the plate, the heat conduction of the solid (i.e., the plate) is usually smaller than the heat exchange with the fluid, and the second derivative of temperature can be ignored. Furthermore, by lumping the complex internal cavity structure, the lumped expression of equation (1) is:

[0071]

[0072] Integrating equation (3) along the length of the plate in the direction of gas flow:

[0073]

[0074] The energy equations for the first and second flow paths on both sides of the plate are as follows:

[0075]

[0076]

[0077] in, For the inflow of the first flow path, T f 1,in T is the inlet temperature of the first flow path. f 1,out The outlet temperature of the first flow path; T represents the inlet flow rate of the second flow path. f 2,in T is the inlet temperature of the second flow path. f 2,out The outlet temperature of the second flow path; C P The specific heat capacity at constant pressure of the fluid;

[0078] Combining equations (4), (5), and (6), we obtain the flat plate model:

[0079]

[0080] Within the horizontal space from 0 to L, we find a spatial point η, i.e., x = η, and model the parameters of the spatial point η: ρ. * To model the density of the flat plate, ρ * =ρ(η);T f 1 * To model the fluid temperature in the first flow path, T f 1 * =T f1 (η); T f 2 * To model the fluid temperature in the second flow path, T f 2 * =T f 2(η); T * T is the temperature at which the molding plate is heated. * =T(η);h1 * To model the convective heat transfer coefficient of the first flow path, h1 * =h1(η);h2 * h2 is the convective heat transfer coefficient for modeling the second flow path. * =h2(η); V is the volume of the plate;

[0081] The flat plate model is further lumped into a model to obtain Figure 3 (b) Flat plate single-sided heat transfer model:

[0082]

[0083] in, The lumped convective heat transfer coefficient is... The total convective heat transfer area; The temperature of the lumped fluid;

[0084] For a flat plate, the differential equations of heat conduction of the solid before and after lumping are equivalent, therefore, from equations (7) and (8), we get:

[0085]

[0086]

[0087] Arrange equations (9) and (10) to obtain the proxy model of the internal cavity structure of the fluid-thermal coupling system after modeling:

[0088]

[0089] Ultimately obtained Tf 2,out The value is the temperature value at the cavity temperature measuring point.

[0090] The model of the internal cavity structure of the fluid-thermal coupling system after modeling includes the following dimensionless parameters:

[0091] (1) Equivalent time constant:

[0092]

[0093] The equivalent time constant α determines the speed of the solid's temperature response and is a measure of thermal inertia. Increasing the equivalent time constant α by a factor of 10 increases the transient cavity temperature change rate while reducing the cavity effect.

[0094] (2) The heat balance coefficient of the first flow path, i.e.: the first Yellow River index:

[0095]

[0096] The first Yellow River index β1 represents the degree to which the temperature of the first flow path is affected by heat transfer through the flat plate wall under steady state. Increasing the first Yellow River index by 10 times reduces the influence of convective heat transfer on the outlet temperature of the first flow path, making the outlet temperature of the first flow path closer to the inlet temperature, while the outlet temperature of the second flow path slightly approaches the inlet temperature of the first flow path.

[0097] (3) The heat balance coefficient of the second flow path, i.e., the second Yellow River index:

[0098]

[0099] The second Yellow River index β2 represents the degree to which the temperature of the second flow path is affected by heat transfer through the flat plate wall under steady-state conditions. Increasing the second Yellow River index by 10 times reduces the influence of convective heat transfer on the outlet temperature of the second flow path, making the outlet temperature of the second flow path closer to the inlet temperature, while the outlet temperature of the first flow path slightly approaches the inlet temperature of the second flow path.

[0100] (4) Relative convective heat transfer intensity, i.e., tug-of-war index:

[0101]

[0102] The tug-of-war index γ determines the solid temperature under steady state. Increasing or decreasing the tug-of-war index by 10 times will correspondingly enhance the heat exchange capacity of the first or second flow path, causing the outlet temperature of the first or second flow path and the solid temperature to generally approach the inlet temperature of the first or second flow path.

[0103] Step 3: Train the surrogate model of the internal cavity structure of the fluid-thermal coupling system obtained in Step 2;

[0104] Based on the transient cavity temperature standard value, a surrogate model of the internal cavity structure of the convection-thermal coupling system is trained using a minimum requirement parameter set. Then, according to different internal cavity structures of the convection-thermal coupling system under test, four dimensionless parameters are adjusted to obtain the transient cavity temperature test value (i.e., T) obtained by the surrogate model of the internal cavity structure of the convection-thermal coupling system. f 2,out The error between the transient cavity temperature and the standard value meets the accuracy requirements and continuously converges to the standard value of the transient cavity temperature.

[0105] It is understandable that adjusting the four dimensionless parameters is to adjust the flow rate of the first flow path. Second flow path flow First flow path convective heat transfer coefficient Second flow path convective heat transfer coefficient Adjusting the flow rate of the first flow path Second flow path flow First flow path convective heat transfer coefficient Second flow path convective heat transfer coefficient for:

[0106]

[0107] in, and These are the initial values ​​for the flow rate of the first flow path, the flow rate of the second flow path, the convective heat transfer coefficient of the first flow path, and the convective heat transfer coefficient of the second flow path, respectively, in the surrogate model; C m1 C m2 C h1 and are respectively the flow rate scaling factor for the first flow path, the flow rate scaling factor for the second flow path, the convective heat transfer coefficient scaling factor for the first flow path, and the convective heat transfer coefficient scaling factor for the second flow path.

[0108] When the accuracy of the surrogate model for the internal cavity structure of the fluid-thermal coupling system meets the requirements, the four scaling factors are fixed and no longer changed, and the surrogate model completes training. At this point, the surrogate model has the capability to measure the transient cavity temperature of this specific complex fluid-thermal coupling system. For subsequent different operating states of the fluid-thermal coupling system, simply inputting the minimum required parameters into the surrogate model will allow for rapid measurement of the cavity temperature at the measuring point. Therefore, the trained surrogate model possesses predictive capabilities, enabling "one look, a lifetime of knowledge."

[0109] Another embodiment of the present invention discloses a transient cavity temperature surrogate model for a fluid-thermal coupling system. The surrogate model is generated using the aforementioned method and is used to detect the transient cavity temperature of the fluid-thermal coupling system. Specifically, a transient cavity temperature detection model for a fluid-thermal coupling system is disclosed, wherein the surrogate model is:

[0110]

[0111] Where α is the equivalent time constant; T* τ is the temperature of the molding plate; τ is the detection time. β1 represents the total fluid temperature; β2 represents the first Yellow River index; T represents the second Yellow River index. f1,in T is the inlet temperature of the first flow path. f1,out T represents the outlet temperature of the first flow path. f2,in T is the inlet temperature of the second flow path. f2,out T is the outlet temperature of the second flow path. f1 * To model the fluid temperature in the first flow path, T f2 * To model the fluid temperature of the second flow path.

[0112] Optionally, the surrogate model is based on compressing the internal cavity structure of the fluid-thermal coupling system perpendicular to the flow direction of the first and second flow paths to obtain a flat plate internal cavity structure; the flat plate model is:

[0113]

[0114] Where, ρ * For modeled flat plate density; h1 * To model the convective heat transfer coefficient of the first flow path, h1 * =h1(η);h2 * V is the convective heat transfer coefficient of the modeled second flow path; V is the volume of the plate.

[0115] To facilitate understanding of the present invention, the method of the present invention is illustrated in detail below with examples. However, the present invention can also be applied to other embodiments different from this one. Therefore, the scope of protection of the present invention is not limited to the following examples.

[0116] Taking the classic CFM56 aero engine as an example, its internal air system structure is as follows: Figure 5 As shown, this complex fluid-thermal coupling system consists of a central rotor shaft and two sides of cold-fluid heat exchangers.

[0117] Obtain the inlet temperature, pressure, and rotational speed of the two flow paths; use the free disk heat exchange method to obtain... By employing the linear relationship between the converted flow rate and the relative converted speed (derived from the compressor characteristic curve), the following is obtained: At this point, all the model's inputs are complete.

[0118] To obtain Figure 5 The internal cavity temperature to be measured is shown. A transient cavity temperature proxy model is obtained using the method of constructing a transient cavity temperature proxy model for a fluid-thermal coupling system according to the present invention. The final calculated cavity temperature result is compared with the standard value, as shown below. Figure 6 As shown, the time-averaged error is less than 14K. At this point, the scaling factors no longer change, and the surrogate model has completed training.

[0119] For the engine's actual minimum set of input parameters, the transient cavity temperature prediction value can be obtained through the aforementioned surrogate model. Comparison with the standard value shows good agreement, with a time-averaged error of less than 7K. Figure 7 As shown, the trained surrogate model has excellent fast cavity temperature prediction capabilities.

Claims

1. A method for constructing a transient cavity temperature surrogate model for a fluid-thermal coupled system, characterized in that, The specific steps include: The internal cavity structure of the fluid-thermal coupling system is compressed perpendicular to the flow direction of the first and second flow paths to obtain a flat plate internal cavity structure. The length of the flat plate along the gas flow direction is x The thermal conductivity differential of infinitesimal elements dx for: （1） in, x The length of the plate along the direction of gas flow; ρ ( x ) is of length x The density of the flat plate at that location; c The specific heat capacity of the flat plate; T ( x The length of the flat plate is... x The temperature of the plate at that location; τ For the detection time; λ The thermal conductivity of the flat plate; S For source terms; H The height of the plate along the compression direction. t The width of the plate is perpendicular to its height and length. For source item S The expression is: S =( T f1 ( x )- T ( x )) h 1( x ) tdx +( T f2 ( x )- T ( x )) h 2( x ) tdx ；（2） in, T f 1 ( x The length of the flat plate is... x The fluid temperature in the first flow path at the location, h 1( x The length of the flat plate is... x The convective heat transfer coefficient of the first flow path at the location, T f 2 ( x The length of the flat plate is... x The fluid temperature in the second flow path at that location, h 2( x The length of the flat plate is... x The convective heat transfer coefficient of the second flow path at the location; The lumped expression of equation (1) is: ρ ( x ) cHtdx· ( T ( x ) / τ )= ( T f1 ( x )- T ( x )) h 1( x ) tdx +( T f2 ( x )- T ( x )) h 2( x ) tdx ；（3） Integrating equation (3) along the length of the plate in the direction of gas flow: ；（4） in, L The length of the plate along the gas flow direction; The energy equations for the first and second flow paths on both sides of the plate are as follows: ；（5） ；（6） in, For the inflow of the first-line path, T f 1,in The inlet temperature of the first flow path, T f 1,out The outlet temperature of the first flow path; For the inflow of the second flow path, T f 2,in The inlet temperature of the second flow path. T f 2,out This is the outlet temperature of the second flow path; C P The specific heat capacity at constant pressure of the fluid; By combining equations (4), (5), and (6), we obtain the flat plate model: ； （7） in, ρ * To model the density of the flat plate; T f 1 * To model the fluid temperature in the first flow path; T f 2 * To model the fluid temperature in the second flow path; T * The temperature of the molded plate; h 1 * The convective heat transfer coefficient for the first flow path is used to model the heat transfer coefficient. h 2 * The convective heat transfer coefficient for the modeled second flow path; V Let be the volume of the flat plate; Model the flat plate to obtain a single-sided heat transfer model: ；（8） in, The lumped convective heat transfer coefficient is... The total convective heat transfer area; The temperature of the lumped fluid; From equations (7) and (8), we get: ； （9） ；（10） Arrange equations (9) and (10) to obtain a proxy model of the internal cavity structure of the fluid-thermal coupling system after modeling: （11） in, α It is the equivalent time constant; β 1 represents the first Yellow River index; β 2 represents the second Yellow River index; Equivalent time constant α It is the equivalent time constant, which determines the speed of the solid's temperature response and is a measure of thermal inertia. Its expression is: （12） First Yellow River Index β 1 represents the degree to which the temperature of the first flow path is affected by heat transfer from the flat plate wall under steady state, expressed as: ；（13） Second Yellow River Index β 2 represents the degree to which the temperature of the second flow path is affected by heat transfer from the flat plate wall under steady-state conditions, expressed as: （14）。 2. The construction method according to claim 1, characterized in that: Obtain the minimum required parameter set and the standard value of transient cavity temperature; Based on the transient cavity temperature standard value, a surrogate model of the internal cavity structure of the fluid-thermal coupling system is trained using the minimum requirement parameter set, and the surrogate model of the internal cavity structure of the fluid-thermal coupling system is continuously converged.

3. A transient cavity temperature proxy model for a fluid-thermal coupled system, characterized in that: The proxy model is as follows: ； in, α It is the equivalent time constant; T * The temperature of the molded plate; τ For the detection time; The temperature of the lumped fluid; β 1 represents the first Yellow River index; β 2 represents the second Yellow River index; T f 1,in The inlet temperature of the first flow path, T f 1,out The outlet temperature of the first flow path; T f 2,in The inlet temperature of the second flow path. T f 2,out This is the outlet temperature of the second flow path; T f 1 * To model the fluid temperature in the first flow path, T f 2 * To model the fluid temperature in the second flow path; Equivalent time constant α It is the equivalent time constant, which determines the speed of the solid's temperature response and is a measure of thermal inertia. Its expression is: in, ρ * To model the density of the flat plate; h 1 * The convective heat transfer coefficient for the first flow path is used to model the heat transfer coefficient. h 2 * The convective heat transfer coefficient for the modeled second flow path; V Let be the volume of the flat plate; The lumped convective heat transfer coefficient is... The total convective heat transfer area; First Yellow River Index β 1 represents the degree to which the temperature of the first flow path is affected by heat transfer from the flat plate wall under steady state, expressed as: in, The inflow rate of the first flow path; C P The specific heat capacity at constant pressure of the fluid; Second Yellow River Index β 2 represents the degree to which the temperature of the second flow path is affected by heat transfer from the flat plate wall under steady-state conditions, expressed as: in, This refers to the inflow rate of the second flow path.

4. The proxy model according to claim 3, characterized in that: The proxy model is based on compressing the internal cavity structure of the fluid-thermal coupling system perpendicular to the flow directions of the first and second flow paths to obtain a flat plate internal cavity structure; the flat plate model is as follows: in, h 1 * = h 1( η ).

Images