A power module temperature hybrid model prediction method

Abstract

The application discloses a power module temperature hybrid model prediction method, and relates to the technical field of power electronic device reliability, and comprises the following steps: a power module finite element thermal model is established based on the geometric size and internal characteristics of the power module; a temperature field sampling snapshot is obtained based on the power module finite element thermal model; a stable evolution trend and an unstable evolution trend of the power module temperature field in the sampling snapshot are extracted based on a dynamic modal decomposition method; an iteration method is used to predict the stable trend, and a deep learning model is used to predict the unstable trend. The predicted stable trend and unstable trend are superimposed to obtain a predicted snapshot of the temperature field. The method can improve the calculation efficiency as much as possible while ensuring the simulation accuracy, and can obtain the chip temperature and internal temperature data of the power module, reduce the number of required sampling snapshots and reduce the modeling period.

Description

Technical Field

[0001] This invention relates to the field of power electronic device reliability technology, and specifically to a method for predicting the temperature of a power module using a hybrid model. Background Technology

[0002] Power modules, due to their high power density and high efficiency, have become core components in electric vehicle power converters. Research shows that the lifespan of a power module is closely related to its maximum junction temperature and junction temperature fluctuations, which can be quantified using a lifespan prediction model. Accurate prediction of the power module's temperature field is crucial for effectively predicting its lifespan and improving the overall reliability of power electronic systems.

[0003] The Finite Element Method (FEM) is a numerical solution method based on materials and structures. It can be used to accurately solve the temperature and stress characteristics of power modules. However, due to the large amount of computational resources consumed by the finite element analysis method and the long analysis time, it is difficult to use for transient analysis of junction temperature of power modules.

[0004] Therefore, a hybrid model prediction method for the temperature field of power modules is needed to improve computational efficiency while ensuring computational accuracy. Summary of the Invention

[0005] The problem to be solved by this invention is to provide a method for predicting the temperature of a power module using a hybrid model, which can significantly improve computational efficiency while ensuring computational accuracy, and is a fast method for calculating the temperature field of a power module.

[0006] This invention adopts the following technical solution: a method for predicting the temperature of a power module using a hybrid model, comprising the following steps:

[0007] Step 1: Build a three-dimensional finite element thermal model of the power module and define material properties and boundary conditions;

[0008] Step 2: Obtain the junction-shell thermal resistance response curve of the power module through finite element simulation, and compare it with the junction-shell thermal resistance curve in the datasheet to verify the accuracy of the three-dimensional finite element thermal model.

[0009] Step 3: Obtain a sampling snapshot of the temperature field of the power module through finite element simulation. The sampling interval is Δt, and the number of sampling snapshots is n.

[0010] Step 4: Based on the dynamic mode decomposition method and the deep learning AE-LSTM model (Autoencoder-Long Short-Term Memory), establish a mixed prediction model for the temperature field of the power module;

[0011] Step 5: Extract the evolution mode of the temperature field using dynamic mode decomposition, screen stable and unstable modes based on the frequency and growth rate of the evolution mode, and obtain the stable trend under the stable mode and the unstable trend under the unstable mode.

[0012] Step 6: Use the recursive method to predict future stable trend snapshots, and use the deep learning AE-LSTM model to predict future unstable trend snapshots.

[0013] Step 7: Overlay the future stable trend snapshot and the future unstable trend snapshot to obtain a future temperature field snapshot.

[0014] Preferably, in step 1, the construction of the three-dimensional finite element thermal model of the power module is carried out through the following sub-steps:

[0015] Step 1.1: Based on the internal structural features and geometric dimensions of the power module, use 3D drawing software to create a 3D physical model of the power module at the actual scale;

[0016] Step 1.2: Mesh the 3D physical model of the power module and set the boundary conditions and material properties of the 3D physical model;

[0017] Step 1.3: Set the power module chip as a heat source and set the boundary condition as power loss.

[0018] Preferably, in step 4, the temperature field mixing prediction model of the power module is constructed through the following sub-steps:

[0019] Step 4.1: Obtain a sampling snapshot of the power module's temperature field based on finite element simulation, and construct a temperature data matrix T as the input to the prediction model:

[0020] T = [T1 T2 ... T] i ...T n ]

[0021] Among them, T i This represents the i-th temperature field snapshot data, where i = 1, 2, ..., n;

[0022] Step 4.2: Obtain the dynamic mode matrix of the temperature field using the dynamic mode decomposition method based on the temperature data matrix T. and mode coefficient matrix λ:

[0023]

[0024] λ=[λ1,λ2,...,λ r ]

[0025] Where, λ j , These represent the j-th coefficient and the dynamic mode, respectively.

[0026] Preferably, in step 5, the frequency f of each temperature field evolution mode is calculated based on the mode coefficient matrix. j and growth rate g j Choose frequency f j and growth rate g j The mode where all values ​​are 0 is considered a stable mode.

[0027]

[0028] A stable trend matrix is ​​obtained through an iterative method based on sampling snapshots and stable patterns.

[0029]

[0030] Based on temperature data matrix T, instability trend matrix Available:

[0031]

[0032] in, Let λ represent the k-th stable mode. sj T1 represents the coefficient corresponding to the j-th dominant mode, and T1 represents the initial temperature field snapshot.

[0033] Preferably, in step 6, a recursive method is used to predict future stable trends. Predicting future unstable trends using deep learning AE-LSTM models A future temperature field snapshot T is obtained by overlaying future stable trend snapshots and future unstable trend snapshots. pred .

[0034] Preferably, the deep learning AE-LSTM model is constructed using the following sub-steps:

[0035] Step 6.1: Calculate the unstable trend matrix. As training data for the AE model, the compressed feature matrix X of the unstable trend matrix is ​​obtained using the trained AE model;

[0036] Step 6.2: Use the compressed feature matrix X as the training data for the LSTM model, and obtain the trained LSTM model through iterative training.

[0037] Step 6.3: Using the trained LSTM model, predict the future compressed feature matrix X. pred ;

[0038] Step 6.4: Use the trained AE model to compress the future feature matrix X. predRestore to a snapshot of future unstable trend predictions

[0039] The present invention also provides: an electronic device, comprising:

[0040] One or more processors;

[0041] A storage device on which one or more programs are stored;

[0042] When the one or more programs are executed by the one or more processors, the one or more processors implement any of the power module temperature hybrid model prediction methods described above.

[0043] The present invention also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps in any of the above-mentioned power module temperature hybrid model prediction methods.

[0044] Compared with the prior art, the present invention, employing the above technical solution, has the following technical effects:

[0045] 1. The power module temperature field hybrid model prediction method proposed in this invention can improve computational efficiency as much as possible while ensuring simulation accuracy by adopting a hybrid model prediction method.

[0046] 2. The power module temperature field hybrid model prediction method proposed in this invention can predict the temperature field of the power module, including the inside of the module package, and not just the junction temperature.

[0047] 3. The power module temperature field hybrid model prediction method proposed in this invention can reduce the number of required sampling snapshots and shorten the modeling cycle. Attached Figure Description

[0048] Figure 1 This is a flowchart of the power module temperature field hybrid model prediction method of the present invention;

[0049] Figure 2 This is a schematic diagram illustrating the transient temperature response calculation according to an embodiment of the present invention;

[0050] Figure 3 This is a schematic diagram of temperature field calculation according to an embodiment of the present invention. Detailed Implementation

[0051] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0052] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0053] In one embodiment of the present invention, please refer to Figure 1 This paper presents a method for predicting the temperature field of a power module using a hybrid model.

[0054] In this embodiment, firstly, a finite element model of the power module is built using Ansys software, and material properties and boundary conditions are defined. The finite element model discretizes the structure of the power module into multiple grid points, and combines the thermal parameters of the material and the boundary conditions to solve for each grid point using the heat conduction equation.

[0055] Then, using numerical analysis, the finite element model can simulate the heat transfer process inside the power module, thereby calculating the transient changes in the power module's temperature field. Sampling snapshots of the power module's temperature field are obtained through finite element simulation, with a sampling interval of Δt and a number of sampling snapshots of n.

[0056] Next, a mixed prediction model of the temperature field of the power module is built. The temperature field evolution mode is extracted from the obtained sampling snapshots using a dynamic mode decomposition model. The stable mode is selected according to the mode frequency and growth rate. The stable trend and unstable trend are obtained by using the stable mode and the sampling snapshots.

[0057] Finally, an iterative method is used to predict the future stable trend, and an AE-LSTM model is used to predict the future unstable trend. The future stable and unstable trends are superimposed to obtain a snapshot of the predicted temperature field. Mesh data from the finite element model of the power module is extracted and combined with the predicted temperature field snapshot to generate a temperature field visualization cloud map.

[0058] The specific implementation steps of the method in this embodiment are as follows:

[0059] Step S1: Construct the finite element model of the power module. The sub-steps are as follows:

[0060] (1) Based on the internal structural features and geometric dimensions of the power module, a three-dimensional physical model of the power module is constructed using three-dimensional drawing software at the actual scale;

[0061] (2) Mesh the established three-dimensional model and configure the boundary conditions and material property parameters of the model in the finite element analysis software Ansys;

[0062] (3) Set the power module chip as a heat source and set the boundary condition as power loss.

[0063] Step S2: Build a power module temperature field prediction model. The sub-steps are as follows:

[0064] (1) In MATLAB software, the temperature data from the first to the nth sampling snapshots are used to construct a temperature data matrix T as the input to the prediction model:

[0065] T = [T1 T2 T3 ... T] i ...T n ]

[0066] (2) Obtain the dynamic mode matrix of the temperature field using dynamic mode decomposition method and its mode coefficient matrix λ:

[0067]

[0068] λ=[λ1,λ2,...,λ r ]

[0069] Step S3: Extract the evolution mode of the temperature field using dynamic mode decomposition. Based on the frequency and growth rate of the evolution mode, screen stable and unstable modes, and obtain the stable trend within the stable mode and the unstable trend within the unstable mode. The sub-steps are as follows:

[0070] (1) Obtain the mode frequency f based on the mode coefficients. j and growth rate g j :

[0071]

[0072] Here, the Re function represents the real part of the complex number, and the Im function represents the imaginary part of the complex number.

[0073] (2) Select frequency f j and growth rate g j The mode with a value of 0 is considered a stable mode:

[0074] (3) Obtain stable trends based on stable patterns and sampling snapshots using iterative methods. and unstable trends

[0075]

[0076]

[0077] Step S4: Predict future stable trends using recursive methods. Predicting unstable trends using AE-LSTM models

[0078]

[0079] Step S5: Overlay future stable trend snapshots and future unstable trend snapshots to obtain future temperature field snapshot T. pred .

[0080]

[0081] Specifically, in this embodiment, the steps for building the AE-LSTM model are as follows:

[0082] (1) Unstable trend matrix As training data for the AE model, the compressed feature matrix X of the unstable trend matrix is ​​obtained using the trained AE model.

[0083] (2) The unstable trend compression feature matrix is ​​used as the training data of the LSTM model, and the LSTM model is obtained through iterative training.

[0084] (3) Predict the future compressed feature matrix X using the trained LSTM model. pred。

[0085] (4) Use the trained AE model to restore the future compressed feature matrix to a future unstable trend prediction snapshot.

[0086] In this embodiment, the transient temperature response at the maximum temperature point is obtained, such as... Figure 2 As shown, temperature fluctuations and temperature ranges can be accurately obtained; the temperature field calculation results are as follows: Figure 3 As shown, the temperature data of the entire power module can be obtained, including the temperature of components such as chips, substrates, and solder.

[0087] In this embodiment of the invention, an electronic device is also provided, comprising: one or more processors; a storage device having one or more programs stored thereon; when the one or more programs are executed by the one or more processors, the one or more processors implement the power module temperature field hybrid model prediction method of any of the above embodiments.

[0088] In this embodiment of the invention, a computer-readable storage medium is also provided, on which a computer program is stored. When the program is executed by a processor, it implements the steps in the power module temperature field hybrid model prediction method of the above embodiments.

[0089] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention.

Claims

1. A method for predicting the temperature of a power module using a hybrid model, characterized in that, Includes the following steps: Step 1: Build a three-dimensional finite element thermal model of the power module and define material properties and boundary conditions; Step 2: Obtain the junction-shell thermal resistance response curve of the power module through finite element simulation, and compare it with the junction-shell thermal resistance curve in the datasheet to verify the accuracy of the three-dimensional finite element thermal model. Step 3: Obtain a sampling snapshot of the temperature field of the power module through finite element simulation. The sampling interval is Δt, and the number of sampling snapshots is n. Step 4: Based on the dynamic mode decomposition method and the deep learning AE-LSTM model, build a mixed prediction model for the temperature field of the power module; Step 5: Extract the evolution mode of the temperature field using dynamic mode decomposition, screen stable and unstable modes based on the frequency and growth rate of the evolution mode, and obtain the stable trend under the stable mode and the unstable trend under the unstable mode. The frequency f of each temperature field evolution mode is calculated based on the mode coefficient matrix. j and growth rate g j Choose frequency f j and growth rate g j The mode where all values ​​are 0 is considered a stable mode. : ； A stable trend matrix is ​​obtained through an iterative method based on sampling snapshots and stable patterns. : ； ； Based on the temperature data matrix T, obtain the unstable trend matrix. : ； in, This represents the k-th stable mode. This represents the coefficient corresponding to the j-th dominant mode. This represents a snapshot of the initial temperature field. The frequency f of the temperature field evolution mode j and growth rate g j Calculated using mode coefficients: ； in, The function represents the real part of a complex number. The function represents the imaginary part of a complex number; Step 6: Use the recursive method to predict future stable trend snapshots, and use the deep learning AE-LSTM model to predict future unstable trend snapshots. Step 7: Overlay the future stable trend snapshot and the future unstable trend snapshot to obtain a future temperature field snapshot.

2. The power module temperature hybrid model prediction method according to claim 1, characterized in that, In step 1, the construction of the three-dimensional finite element thermal model of the power module is carried out through the following sub-steps: Step 1.1: Based on the internal structural features and geometric dimensions of the power module, use 3D drawing software to create a 3D physical model of the power module at the actual scale; Step 1.2: Mesh the 3D physical model of the power module and set the boundary conditions and material properties of the 3D physical model; Step 1.3: Set the power module chip as a heat source and set the boundary condition as power loss.

3. The power module temperature hybrid model prediction method according to claim 1, characterized in that, In step 4, the temperature field hybrid prediction model of the power module is constructed through the following sub-steps: Step 4.1: Obtain a sampling snapshot of the power module's temperature field based on finite element simulation, and construct a temperature data matrix T as the input to the prediction model: ； in, This represents the i-th temperature field snapshot data, where i = 1, 2, ..., n; Step 4.2: Obtain the dynamic mode matrix of the temperature field using the dynamic mode decomposition method based on the temperature data matrix T. and mode coefficient matrix : ； ； in, These represent the j-th coefficient and the dynamic mode, respectively.

4. The power module temperature hybrid model prediction method according to claim 3, characterized in that, In step 6, the recursive method is used to predict the future stable trend. , ； ； Predicting future unstable trends using deep learning AE-LSTM models : 。 5. The power module temperature hybrid model prediction method according to claim 4, characterized in that, In step 7, a future temperature field snapshot is obtained by overlaying a future stable trend snapshot and a future unstable trend snapshot. : 。 6. The power module temperature hybrid model prediction method according to claim 4, characterized in that, The construction steps for the deep learning AE-LSTM model are as follows: Step 6.1: Calculate the unstable trend matrix. As training data for the AE model, the compressed feature matrix X of the unstable trend matrix is ​​obtained using the trained AE model; Step 6.2: Use the compressed feature matrix X as the training data for the LSTM model, and obtain the trained LSTM model through iterative training. Step 6.3: Using the trained LSTM model, predict the future compressed feature matrix X. pred ; Step 6.4: Use the trained AE model to compress the future feature matrix X. pred Restore to a snapshot of future unstable trend predictions .

7. An electronic device, characterized in that, include: One or more processors; A storage device on which one or more programs are stored; When the one or more programs are executed by the one or more processors, the one or more processors implement the power module temperature hybrid model prediction method as described in any one of claims 1 to 6.

8. A computer-readable storage medium, characterized in that, It stores a computer program that, when executed by a processor, implements the steps in the power module temperature hybrid model prediction method according to any one of claims 1 to 6.

Images