A thermal model correction system and method for inverse design based on thermal test data
Through the inverse design thermal model correction system based on thermal test data, the conjugate gradient method inversely calculates the parameters to be corrected, and the problem of low multi-parameter correction accuracy in satellite thermal design is solved, and high-precision thermal model correction and reduction of invalid tests are achieved Make up time.
Patent Information
- Application Number
- CN202111642728.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-12-29
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2041-12-29
AI Technical Summary
In satellite thermal design, when multiple parameters are corrected simultaneously, the accuracy of the thermal model correction results is not high, and existing methods are difficult to effectively improve the correction accuracy.
The inverse design thermal model correction system based on thermal test data is adopted, the conjugate gradient method is used to inversely calculate the parameters to be corrected, and parameterized settings are established through the software interface module and satellite thermal model, and the temperature field gradient and search step size are iteratively solved, and the satellite is automatically corrected. Thermal model.
The thermal model simulation accuracy is improved, the invalid trial time is reduced, and high-precision model correction is achieved under the thermal equilibrium test conditions.
Smart Images

Figure CN114417658B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of thermal test data processing, and in particular to an inverse design thermal model correction system and method based on thermal test data. Background Art
[0002] When designing satellite thermal design, the satellite is usually first analyzed using Thermal Desktop and Sinda / Fluint software, and then the model is corrected through thermal balance tests. The correction parameters include the model's surface coating thermo-optical properties and contact heat transfer coefficient, etc. However, when correcting multiple parameters at the same time, despite the use of multiple methods, the accuracy of the correction results is not high. Summary of the invention
[0003] The present invention provides an inverse design thermal model correction system and method based on thermal test data, aiming to correct the satellite thermal model through an inverse design algorithm and improve the model correction accuracy.
[0004] The present invention provides an inverse design thermal model correction system based on thermal test data, comprising:
[0005] Software interface module: provides an interface for automatic operation of thermal control analysis software;
[0006] Satellite thermal model: According to the known input conditions such as the current overall layout, heat consumption, working mode of the satellite and the assumed values of the contact heat transfer coefficient to be corrected parameters, a satellite thermal analysis model is established on the thermal control analysis software; in the thermal model, the parameters to be corrected are parameterized;
[0007] Conjugate gradient calculation module: Based on the temperature field of the thermal test and the temperature field calculated by the satellite thermal model during correction, the conjugate gradient method is used to inversely calculate the parameters to be corrected in the thermal model;
[0008] The software interface module receives the parameter value to be corrected calculated by the conjugate gradient calculation module, passes the parameter value to be corrected to the thermal control analysis software, initializes the thermal control analysis software and starts the analysis software; obtains the result calculated by the thermal control analysis software, and feeds the calculation result back to the conjugate gradient calculation module.
[0009] As a further improvement of the present invention, the operation process of the conjugate gradient calculation module includes:
[0010] Assume the parameter vector to be corrected is P, define the function S
[0011] S(P)=[YT(P)] T [YT(P)] (1)
[0012] Where Y is the target value of the satellite temperature field, that is, the thermal test temperature field, and T(P) is the temperature field of the satellite when the parameter to be corrected is P;
[0013] The goal is to find the minimum value of function S
[0014]
[0015] As a further improvement of the present invention, in each iteration process of the conjugate gradient calculation module, a new correction parameter value is obtained according to conditions including the temperature field of the thermal balance test, the temperature field calculated by the actual received satellite thermal model, and the temperature change gradient of the satellite thermal model:
[0016] P k+1 =P k -β k d k (3)
[0017] Among them, β k is the search step length, d k is the descending direction;
[0018]
[0019] where γ k is the conjugate coefficient;
[0020]
[0021]
[0022]
[0023]
[0024] The present invention provides an inverse design thermal model correction method based on thermal test data, comprising the following steps:
[0025] S1. Assume the initial parameter to be corrected is P 0 , define k = 0;
[0026] S2. The software interface module sets the initial parameter value to be corrected P 0 The value of the parameter to be corrected is P. k The satellite temperature field at time T(P k );
[0027] S3. The software interface module converts T(P k ) is fed back to the conjugate gradient calculation module, which calculates the conjugate gradient according to the convergence criterion S(P k+1)<ε to determine whether it converges, if it converges, stop the correction process; if it does not converge, continue to step S4;
[0028] S4. Calculation parameter conditions: Solve the parameter to be corrected and take the value as P k The temperature field gradient S-function gradient Conjugation coefficient γ k , descending direction d k , search step length β k ;
[0029] S5. Obtain a new value P of the parameter to be corrected according to the parameter condition in step S4. k+1 , change the satellite thermal model parameter to be corrected to P k+1 ;
[0030] S6. Define k=k+1, and return to execute step S2.
[0031] As a further improvement of the present invention, the value of the parameter to be corrected in step S4 is P k The temperature field gradient S-function gradient The numerical solution is obtained by finite difference method.
[0032] As a further improvement of the present invention, the conjugate coefficient γ in step S4 is k The calculation equation is:
[0033]
[0034] As a further improvement of the present invention, the descending direction d in step S4 k The calculation equation is:
[0035]
[0036] As a further improvement of the present invention, the search step length β in step S4 is k The calculation equation is:
[0037]
[0038] in
[0039]
[0040] As a further improvement of the present invention, the new correction parameter P in step S5 k+1 The calculation equation is:
[0041] P k+1 =P k -βk d k .
[0042] The beneficial effect of the present invention is that under the condition of known thermal balance test temperature field, the parameters to be corrected in the thermal model are inversely designed by the conjugate gradient method, so that the thermal model simulation accuracy meets the requirements. The method can realize automatic correction of the thermal model according to the thermal balance test temperature field, improve the correction accuracy, and reduce the invalid trial and error time. BRIEF DESCRIPTION OF THE DRAWINGS
[0043] Figure 1 It is a module schematic diagram of an inverse design thermal model correction system based on thermal test data of the present invention. DETAILED DESCRIPTION
[0044] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments.
[0045] like Figure 1 As shown, the inverse design thermal model correction system based on thermal test data of the present invention includes the following functional modules:
[0046] Software interface module:
[0047] The interface software provides an interface for the automatic operation of the thermal control analysis software. The main functions of the interface software are: receiving the values of the parameters to be corrected calculated by the conjugate gradient calculation module, passing the values of the parameters to be corrected to the thermal control analysis software, initializing the thermal control analysis software and starting the analysis software; obtaining the results calculated by the thermal control analysis software, and feeding the calculation results back to the conjugate gradient calculation module.
[0048] Satellite thermal model:
[0049] According to the known input conditions such as the current overall layout, heat consumption, working mode of the satellite and the assumed values of the contact heat transfer coefficient to be corrected parameters, a satellite thermal analysis model is established on the thermal control analysis software. In the thermal model, the contact heat transfer coefficient to be corrected parameters need to be parameterized. The parameterization function in the thermal control analysis software sets the parameters to be corrected to variable parameters specified by the code, rather than fixed values.
[0050] Conjugate gradient calculation module:
[0051] According to the temperature field of the thermal test and the temperature field calculated by the satellite thermal model during correction, the conjugate gradient method is used to inversely calculate the parameters to be corrected in the thermal model. The calculation principle is as follows:
[0052] Assume the parameter vector to be corrected is P, define the S function
[0053] S(P)=[YT(P)] T[YT(P)] (1)
[0054] Where Y is the target value of the satellite temperature field, that is, the thermal test temperature field, and T(P) is the temperature field of the satellite when the parameter to be corrected takes the value of P.
[0055] The goal is to find the minimum value of the S function
[0056]
[0057] In each iteration of the conjugate gradient calculation module, new correction parameter values are obtained based on the temperature field of the thermal balance test, the temperature field calculated by the actual received satellite thermal model, the temperature change gradient of the satellite thermal model, and other conditions:
[0058] P k+1 =P k -β k d k (3)
[0059] Among them, β k is the search step length, d k It is the descending direction
[0060]
[0061] where γ k is the conjugate coefficient
[0062]
[0063]
[0064]
[0065]
[0066] A method for correcting an inverse design thermal model based on thermal test data of the present invention is as follows:
[0067] S1. Assume that the initial parameter to be corrected is P 0 , define k=0.
[0068] S2. The software interface module sets the initial parameter value to be corrected P 0 The value of the parameter to be corrected is P. k The satellite temperature field at time T(P k ).
[0069] S3. The software interface module converts T(P k ) is fed back to the conjugate gradient calculation module, which calculates the conjugate gradient according to the convergence criterion S(P k+1)<ε to determine whether it converges. If it converges, stop the correction process; if it does not converge, continue to step S4; where ε is the sum of the satellite temperature field target value and the actual temperature field standard deviation, that is, The standard deviation is set according to the accuracy requirements.
[0070] S4. Calculation parameter conditions: Solve the parameter to be corrected and take the value as P k The temperature field gradient S-function gradient Conjugation coefficient γ k , descending direction d k , search step length β k .
[0071] Among them, since the temperature field solved by this thermal model is nonlinear, It cannot be obtained directly, so it is obtained by numerical solution using the finite difference method.
[0072] Conjugation coefficient γ k The calculation equation is:
[0073]
[0074] Descending direction k The calculation equation is:
[0075]
[0076] Search step length β k The calculation equation is:
[0077]
[0078] in
[0079]
[0080] S5. Obtain a new value P of the parameter to be corrected according to the parameter condition in step S4. k+1 , change the satellite thermal model parameter to be corrected to P k+1 , the new value of the parameter to be corrected P k+1 The calculation equation is:
[0081] P k+1 =P k -β k d k .
[0082] S6. Define k=k+1, and return to step S2 to perform the next correction.
[0083] The above contents are further detailed descriptions of the present invention in combination with specific preferred embodiments, and it cannot be determined that the specific implementation of the present invention is limited to these descriptions. For ordinary technicians in the technical field to which the present invention belongs, several simple deductions or substitutions can be made without departing from the concept of the present invention, which should be regarded as falling within the protection scope of the present invention.
Claims
1. An inverse design thermal model correction system based on thermal test data, characterized in that: include: Software interface module: provides an interface for automatic operation of thermal control analysis software; Satellite thermal model: According to the known input conditions of the satellite's current overall layout, heat consumption, working mode and the assumed values of the parameters to be corrected of the contact heat transfer coefficient, a satellite thermal analysis model is established on the thermal control analysis software; in the thermal model, the parameters to be corrected are parameterized; Conjugate gradient calculation module: Based on the temperature field of the thermal test and the temperature field calculated by the satellite thermal model during correction, the conjugate gradient method is used to inversely calculate the parameters to be corrected in the thermal model; The software interface module receives the parameter value to be corrected calculated by the conjugate gradient calculation module, transmits the parameter value to be corrected to the thermal control analysis software, initializes the thermal control analysis software and starts the analysis software; Obtain the calculation results of the thermal control analysis software and feed the calculation results back to the conjugate gradient calculation module; Characterized in that the operation process of the conjugate gradient calculation module includes: Assume the parameter vector to be corrected is P, define the function S S(P)=[Y-T(P)] T [Y-T(P)] (1) Where Y is the target value of the satellite temperature field, that is, the thermal test temperature field, and T(P) is the temperature field of the satellite when the parameter to be corrected is P; The goal is to find the minimum value of function S In each iteration of the conjugate gradient calculation module, a new correction parameter value is obtained according to conditions including the temperature field of the thermal balance test, the temperature field calculated by the actual received satellite thermal model, and the temperature change gradient of the satellite thermal model: P.S k+1 JP k -β k d k (3) Among them, β k is the search step length, d k is the descending direction; where γ k is the conjugate coefficient; Among them, P k Indicates the value of the parameter to be corrected, P k+1 represents the new value of the parameter to be corrected, k represents the kth iteration; i represents the i-th temperature point, T i represents the calculated temperature of the i-th point of the satellite, Y i is the target temperature of the i-th point of the satellite; The parameter to be corrected is taken as P k The temperature field gradient at The parameter to be corrected is taken as P k The gradient of the S function when .
2. A method for correcting an inverse design thermal model based on thermal test data, characterized in that: The following steps are involved: S1. Assume the initial parameter to be corrected is P 0 , define k = 0; S2. The software interface module sets the initial parameter value to be corrected P 0 The value of the parameter to be corrected is P. k The satellite temperature field at time T(P k ); S3. The software interface module converts T(P k ) is fed back to the conjugate gradient calculation module, which calculates the conjugate gradient according to the convergence criterion S(P k+1 )<ε to determine whether it converges. If it converges, stop the correction process; if it does not converge, continue to step S4; where ε is the sum of the satellite temperature field target value and the actual temperature field standard deviation; S4. Calculation parameter conditions: Solve the parameter to be corrected and take the value as P k The temperature field gradient S-function gradient Conjugation coefficient γ k , descending direction d k , search step length β k ; S5. Obtain a new value P of the parameter to be corrected according to the parameter condition in step S4. k+1 , change the satellite thermal model parameter to be corrected to P k+1 ; S6. Define k=k+1, and return to execute step S2; In step S4, the parameter to be corrected is solved to be P k The temperature field gradient S function gradient ▽S(P k ) is obtained by numerical approximation using the finite difference method; The conjugate coefficient γ in step S4 k The calculation equation is: In step S4, the descending direction d k The calculation equation is: In step S4, the search step length β k The calculation equation is: in 3. The inverse design thermal model correction method based on thermal test data according to claim 2 is characterized in that: The new correction parameter P in step S5 k+1 The calculation equation is: P.S k+1 JP k -β k d k 。 。
Citation Information
Patent Citations
Satellite on-orbit transient temperature simulation and prediction system
CN103942429A
Spacecraft transient thermal analysis model inversion correction method
CN105930676A
Thermal analysis software external call interface-based real-time satellite temperature field calculation method
CN108664693A
Small satellite dynamic thermal analysis modeling method
CN111753375A
Thermal sense calculation device, thermal sense calculation method, air conditioner, and program
JP2019023528A
