A Method, System, Device and Medium for Inverting the Temperature of a High-Voltage Power Module Chip

Abstract

The present application discloses a method, system, device and medium for inverting the chip temperature of a high-voltage power module. By constructing a high-fidelity finite element model of the high-voltage power module, simulations are carried out under boundary conditions such as different current levels and different radiator temperatures to obtain the chip temperature and the case temperature, and a data set is constructed based on the obtained simulation data. A multi-input multi-output deep neural network is built, and a non-linear mapping relationship between the measurable point temperature and the chip temperature is obtained through training with the simulation data set. An experiment is carried out to obtain a small-sample experimental data set of the chip temperature and the case temperature under actual working conditions, and the trained deep neural network is fine-tuned with the small-sample experimental data set to correct the non-linear relationship and improve the accuracy of chip temperature inversion. Thus, the problem in the prior art of difficultly and inefficiently obtaining the chip temperature of a high-voltage power module is solved.

Description

Technical Field

[0001] This application relates to the technical field of high - voltage power module state analysis, and particularly to a method, system, device and medium for inverting the chip temperature of a high - voltage power module. Background Technique

[0002] High - voltage power modules play an important role in modern power electronics applications and are widely used in multiple fields such as DC power transmission, railway traction, and electric vehicles. With the continuous progress of technology and the increasing demand for higher power density and integrated converters, the application of multi - chip power modules is becoming increasingly popular. However, these modules face the problem of high - temperature failure during operation, which is one of the main reasons for their failure. The chip is located inside the high - voltage power module and is encapsulated by insulating materials, making it difficult to accurately obtain its temperature through conventional means. Quickly and accurately inverting the chip temperature of the power module is crucial for effective thermal management, life estimation, and condition monitoring.

[0003] Traditional methods for obtaining the chip temperature of modules mainly include direct measurement method, thermal circuit calculation method, etc. The direct measurement method is likely to damage the module itself and is difficult to be widely promoted; the thermal circuit calculation method is complex, the acquisition of the thermal circuit model is difficult, and the calculation results often deviate greatly from the actual results, making it difficult to widely implement engineering applications. Summary of the Invention

[0004] This application provides a method, system, device and medium for inverting the chip temperature of a high - voltage power module, which is used to solve the problem that it is difficult to accurately and efficiently obtain the chip temperature of a high - voltage power module in the prior art.

[0005] In view of this, in the first aspect of this application, a method for inverting the chip temperature of a high - voltage power module is provided, and the method includes:

[0006] Construct a high - fidelity finite - element model of the high - voltage power module to be inverted, and verify the correctness of the high - fidelity finite - element model;

[0007] Based on the high - fidelity finite - element model that has passed the correctness verification, and according to the preset boundary conditions, perform simulations to obtain the chip temperature and the case temperature for constructing a data set, where the preset boundary conditions include: different current levels and different radiator temperatures;

[0008] Build a multi - input multi - output deep neural network, and train the multi - input multi - output deep neural network with the pre - processed data set to obtain a deep neural network for characterizing the non - linear mapping relationship between the measurable point temperature and the chip temperature;

[0009] After building a two-level inverter using a high-voltage power module and setting the control algorithm and working parameters, the load current is controlled by adjusting the load resistance, and at the same time, several sets of sample experimental data sets of chip temperature and case temperature are obtained through detection.

[0010] Based on the transfer learning theory, the parameters of the pre-trained deep neural network are fine-tuned using the sample experimental data set, so as to correct the non-linear relationship and obtain the final deep neural network for inverting the chip temperature of the high-voltage power module.

[0011] Optionally, the construction of the high-fidelity finite element model of the high-voltage power module to be inverted specifically includes:

[0012] Obtain the geometric dimensions of the high-voltage power module to be inverted and the material property parameters of each component of the high-voltage power module to be inverted. Among them, the geometric dimensions include: chip size, radiator structure, welding layer thickness, and encapsulation material thickness.

[0013] According to the geometric dimensions and the material property parameters, a high-fidelity finite element model of the high-voltage power module to be inverted is constructed using finite element analysis software.

[0014] Optionally, the verification of the correctness of the high-fidelity finite element model specifically includes:

[0015] The chip, ceramic, and copper substrate in the high-fidelity finite element model are respectively meshed using network units of different sizes.

[0016] The average temperature and maximum temperature on the chip surface obtained by testing with an infrared camera and the finite element simulation calculation results are used to verify the correctness of the high-fidelity finite element model.

[0017] Optionally, the construction of the multi-input multi-output deep neural network specifically includes:

[0018] A multi-input multi-output deep neural network is constructed using a multi-layer perceptron structure as the network structure. Among them, the network inputs include: several measurable point temperatures, and the network outputs include several chip temperatures.

[0019] The second aspect of the present application provides a high-voltage power module chip temperature inversion system, and the system includes:

[0020] A construction unit for constructing a high-fidelity finite element model of the high-voltage power module to be inverted and verifying the correctness of the high-fidelity finite element model;

[0021] A simulation unit, configured to perform a simulation based on the high-fidelity finite element model after passing the correctness verification, and obtain the chip temperature and the case temperature according to preset boundary conditions for constructing a data set, where the preset boundary conditions include: different current levels and different heatsink temperatures;

[0022] A training unit, configured to build a multi-input multi-output deep neural network, and train the multi-input multi-output deep neural network with the preprocessed data set to obtain a deep neural network for characterizing the non-linear mapping relationship between the measurable point temperature and the chip temperature;

[0023] An acquisition unit, configured to build a two-level inverter using a high-voltage power module, set control algorithms and operating parameters, and then control the load current magnitude by adjusting the load resistance, and at the same time perform detections to obtain a sample experimental data set of several groups of chip temperatures and case temperatures;

[0024] An adjustment unit, configured to perform parameter fine-tuning on the trained deep neural network using the sample experimental data set based on the transfer learning theory, so as to correct the non-linear relationship and obtain a final deep neural network for inverting the chip temperature of the high-voltage power module.

[0025] Optionally, the construction of the high-fidelity finite element model of the high-voltage power module to be inverted specifically includes:

[0026] Obtain the geometric dimensions of the high-voltage power module to be inverted and the material property parameters of each component of the high-voltage power module to be inverted, where the geometric dimensions include: chip size, heatsink structure, welding layer thickness, and package material thickness;

[0027] According to the geometric dimensions and the material property parameters, use finite element analysis software to construct a high-fidelity finite element model of the high-voltage power module to be inverted.

[0028] Optionally, the verification of the correctness of the high-fidelity finite element model specifically includes:

[0029] Use network units of different sizes to perform meshing on the chip, ceramic, and copper substrate in the high-fidelity finite element model respectively;

[0030] Use the average temperature and maximum temperature of the chip surface obtained by testing with an infrared camera and the finite element simulation calculation results to verify the correctness of the high-fidelity finite element model.

[0031] Optionally, the construction of the multi-input multi-output deep neural network specifically includes:

[0032] Adopt a multi-layer perceptron structure as the network structure to build a deep neural network with multiple inputs and multiple outputs. Among them, the network inputs include: the temperatures of several measurable points, and the network outputs include the temperatures of several chips.

[0033] The third aspect of this application provides a device for the method of inverting the chip temperature of a high-voltage power module. The device includes a processor and a memory:

[0034] The memory is used to store program code and transmit the program code to the processor;

[0035] The processor is used to execute the steps of the method of inverting the chip temperature of the high-voltage power module as described in the first aspect above according to the instructions in the program code.

[0036] The fourth aspect of this application provides a computer-readable storage medium. The computer-readable storage medium is used to store program code, and the program code is used to execute the method of inverting the chip temperature of the high-voltage power module as described in the first aspect above.

[0037] It can be seen from the above technical solutions that this application has the following advantages:

[0038] For the method of inverting the chip temperature of a high-voltage power module provided by this application, first, a high-fidelity finite element model of the high-voltage power module is constructed, and simulations are carried out under boundary conditions such as different current levels and different radiator temperatures to obtain the chip temperature and the case temperature, and a data set is constructed based on the obtained simulation data. Then, a deep neural network with multiple inputs and multiple outputs is built, and a non-linear mapping relationship between the measurable point temperature and the chip temperature is obtained through training using the simulation data set. Finally, experiments are carried out to obtain a small-sample experimental data set of the chip temperature and the case temperature under actual working conditions, and the trained deep neural network is fine-tuned using the small-sample experimental data set to correct the non-linear relationship and improve the accuracy of chip temperature inversion.

[0039] Compared with the prior art, this application constructs a non-linear relationship between the measurable point temperature and the chip temperature of the high-voltage power module based on a neural network, and uses the measurable point temperature to efficiently and accurately invert the chip temperature, which is easier to promote on a large scale than traditional methods. Thus, the problem that it is difficult to accurately and efficiently obtain the chip temperature of the high-voltage power module in the prior art is solved. Description of the Drawings

[0040] Figure 1 It is a schematic flow chart of a method for inverting the chip temperature of a high-voltage power module provided in an embodiment of this application;

[0041] Figure 2 It is the finite element model of the high-voltage power module provided in an embodiment of this application;

[0042] Figure 3The meshing result of the welded IGBT provided in the embodiment of the present application;

[0043] Figure 4 The verification result of the high-fidelity model provided in the embodiment of the present application;

[0044] Figure 5 The transient response curve of the chip temperature of the high-voltage power module provided in the embodiment of the present application;

[0045] Figure 6 The MLP neural network structure provided in the embodiment of the present application;

[0046] Figure 7 The loss function of the training process provided in the embodiment of the present application;

[0047] Figure 8 The schematic diagram of the two-level inverter experimental platform provided in the embodiment of the present application;

[0048] Figure 9 The load current waveform provided in the embodiment of the present application;

[0049] Figure 10 The schematic diagram of parameter fine-tuning provided in the embodiment of the present application;

[0050] Figure 11 The chip temperature inversion error provided in the embodiment of the present application;

[0051] Figure 12 The schematic diagram of the structure of a chip temperature inversion system for a high-voltage power module provided in the embodiment of the present application. Detailed implementation manners

[0052] In order to enable those skilled in the art to better understand the solution of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present application without creative efforts shall fall within the protection scope of the present application.

[0053] Please refer to Figure 1 , a method for inverting the chip temperature of a high-voltage power module provided in the embodiment of the present application, includes:

[0054] Step 101: Construct a high-fidelity finite element model of the high-voltage power module to be inverted, and verify the correctness of the high-fidelity finite element model.

[0055] In one embodiment, constructing a high-fidelity finite element model of the high-voltage power module to be inverted is as follows:

[0056] Obtain the geometric dimensions of the high-voltage power module to be inverted and the material property parameters of each component of the high-voltage power module to be inverted. Among them, the geometric dimensions include: chip size, heat sink structure, welding layer thickness, and encapsulation material thickness; According to the geometric dimensions and material property parameters, use finite element analysis software to construct a high-fidelity finite element model of the high-voltage power module to be inverted.

[0057] It should be noted that when constructing the finite element model of the high-voltage power module, it is first necessary to accurately measure the geometric dimensions of the module. These dimensions include chip size, heat sink structure, welding layer thickness, encapsulation material thickness, etc. Using high-precision measurement equipment, such as a coordinate measuring machine (CMM) or a laser scanner, can ensure the accuracy of the geometric dimensions. In addition, in order to construct a high-fidelity model, it is necessary to obtain the material property parameters of each component of the high-voltage power module. This includes the thermal conductivity, specific heat capacity, density, etc. of the chip material silicon, the thermal conductivity and thermal expansion coefficient of the encapsulation materials (such as ceramics, copper), and the thermal conductivity and shear modulus of the welding material. These parameters can be obtained through literature search, data provided by manufacturers, or through experimental tests. After obtaining the geometric dimensions and material parameters, use finite element analysis software (COMSOL Multiphysics) to establish a high-fidelity finite element model. The geometric shape and material properties of each component need to be considered in the model, and appropriate boundary conditions need to be applied, including the current rating and the heat sink temperature. In this application, a commercial high-voltage power module (FF150R12ME3G) is taken as an example. The dimensions and materials of this module are obtained through the data sheet. The high-fidelity model is obtained by drawing through finite element analysis software as Figure 2 shown.

[0058] In one embodiment, the correctness of the high-fidelity finite element model is verified as follows:

[0059] Use network elements of different sizes to respectively divide the chip, ceramic, and copper substrate in the high-fidelity finite element model; The average temperature and maximum temperature on the chip surface obtained by testing with an infrared camera and the finite element simulation calculation results are used to verify the correctness of the high-fidelity finite element model.

[0060] It should be noted that before analyzing the high-fidelity model of the high-voltage power module, it needs to be meshed. The finite element method requires dividing the model into elements with simple geometric shapes. Reasonable mesh division is beneficial to improving the accuracy and convergence of the results. Since the chip size is relatively small compared to other components and has a great influence on the calculation results, smaller mesh elements are used for sweeping, and the maximum element size is set to 0.5 mm. The thermal conductivities of the ceramic and copper substrates are relatively large, and their temperature distributions are relatively uniform. Therefore, larger mesh elements can be used for division. After meshing, there are a total of 807,656 mesh elements and 199,540 vertices. The meshing results of the finite element simulation model of the high-voltage power module are as Figure 3 shown.

[0061] To verify the accuracy of the high-fidelity model, the average temperature and maximum temperature of the chip surface obtained by testing with an infrared camera in this application and the results of finite element simulation calculations are listed in Table 1. The error between the average temperature of the surfaces of the three parallel chips at the lower tube obtained by finite element calculation and the measured value is less than 0.5 °C; the maximum deviation between the calculated results of the maximum temperature of the surfaces of the three parallel chips and the measured value is 1.7 °C. The bonding wires on the chip surface affect the measured surface maximum temperature result. The surface temperature measured by the infrared camera is affected by the spraying effect of the heat-reflecting black paint. The surface temperature measured by the infrared camera is as Figure 4 shown. Therefore, it is considered that under the heat sink and heat dissipation conditions used in this application, the high-fidelity model of the high-voltage power module is accurate and reasonable.

[0062] Table 1 Comparison of chip surface temperatures

[0063]

[0064] Step 102: Based on the high-fidelity finite element model that has passed the correctness verification, and according to the preset boundary conditions, perform simulations to obtain the chip temperature and the case temperature for constructing a data set. Among them, the preset boundary conditions include: different current levels and different heat sink temperatures.

[0065] It should be noted that since a large amount of data is required to train a neural network to obtain its non-linear mapping function. Therefore, common working current levels and heat sink temperatures of the high-voltage power module are selected, and a data set is constructed under these conditions.

[0066] In this application, the working current levels are selected as 25A, 50A, 75A, 100A, 125A, and 150A, and the radiator temperatures are 0°C, 5°C, 10°C, 15°C, 20°C, 25°C, and 30°C. With the combination of these two working conditions, a total of 36 groups of working condition simulation results are obtained. Since the transient temperature rise curve is simulated, each group of working condition simulation results can be resampled to obtain M groups, and the value of M can be determined according to the size of the transient temperature rise time. In this application, M is selected as 8000, that is, each group of working conditions can obtain 8000 groups of training data. Thus, a total of 36×8000 = 288000 groups of training data can be obtained finally. Select the simulation results with a current of 50A and a radiator temperature of 25°C, as Figure 5 shown.

[0067] To facilitate finding the non-linear mapping relationship, 12 substrate positions corresponding to the high-voltage power module chips are selected as measurable points, and a structure that maps 12 measurable points to 12 chip temperatures is constructed, that is, a one-dimensional array of 24 bits. The first 12 bits represent the temperatures of the measurable points, and the last 12 bits represent the chip temperatures.

[0068] Step 103: Build a multi-input multi-output deep neural network, and train the multi-input multi-output deep neural network with the preprocessed data set to obtain a deep neural network for characterizing the non-linear mapping relationship between the temperatures of the measurable points and the chip temperatures.

[0069] In one embodiment, building a multi-input multi-output deep neural network is as follows:

[0070] Adopt a multi-layer perceptron structure as the network structure to build a multi-input multi-output deep neural network. Among them, the network input includes: several temperatures of the measurable points, and the network output includes several chip temperatures.

[0071] It should be noted that a multi-input multi-output (MIMO) deep neural network is designed. The network input includes 12 temperatures of the measurable points, and the network output is 12 chip temperatures. The network structure adopts a multi-layer perceptron (MLP) structure, as Figure 6 shown.

[0072] Then, preprocess the simulation data, including normalization, denoising, etc. Normalization can make the data on the same scale, which is convenient for the training of the neural network; denoising can improve the quality and training effect of the data. Use the preprocessed simulation data set to train the deep neural network. During the training process, the cross-validation method can be adopted to prevent overfitting. Select an appropriate loss function (such as mean square error MSE) and optimization algorithm (such as Adam), and adjust the network parameters to ensure that the network converges to the optimal solution. The loss function values during the training process are as Figure 7 shown. After the training is completed, a deep neural network that can characterize the non-linear mapping relationship is obtained.

[0073] Step 104: After building a two-level inverter with a high-voltage power module and setting the control algorithm and working parameters, control the load current by adjusting the load resistance, and at the same time perform detection to obtain several sets of sample experimental data sets of chip temperature and case temperature.

[0074] It should be noted that a two-level inverter is built using a high-voltage power module of model FF150R12ME3G, as Figure 8 shown. The inverter adopts a sine pulse width modulation control algorithm, the DC voltage Udc is 200V, the control switching frequency fsw is 4kHz, the carrier frequency f0 is 1Hz, and the load inductance L is 5mH. The silicone of the high-voltage power module needs to be removed to measure the chip temperature, and the module cannot work under high voltage, so the DC voltage is maintained at 200V. Control the load current by adjusting the load resistance. The load resistance R selected in this application is 0.25Ω, and the modulated load current is as Figure 9 shown, which can make the maximum current reach about 80A.

[0075] 100 sets of experimental data are obtained by measurement using sensors and infrared thermometers to construct a small sample data set for the experiment.

[0076] Step 105: Based on the transfer learning theory, use the sample experimental data set to fine-tune the parameters of the pre-trained deep neural network, so as to correct the non-linear relationship and obtain the final deep neural network for inverting the chip temperature of the high-voltage power module.

[0077] It should be noted that transfer learning is a machine learning method that transfers the knowledge of a pre-trained model to a new task. For this project, the deep neural network model trained on simulation data can be used, and fine-tuned through a small amount of experimental data to improve the prediction accuracy of the model under actual working conditions. Load the deep neural network model trained on simulation data and freeze some parameters (weights of the first few layers), and only fine-tune the last few layers, as Figure 10 shown.

[0078] In this way, the basic structure and most of the knowledge of the model can be retained, and at the same time, the model can be adaptively adjusted through small sample experimental data. Fine-tune and train the deep neural network using small sample experimental data. Select appropriate learning rates and training strategies to ensure the convergence of the model on small sample data. Through the feedback of experimental data, gradually adjust the model parameters to improve the accuracy of chip temperature inversion. In this application, the chip inversion error can be reduced from 7.5% to 2.5%, as Figure 11 shown, with remarkable effects.

[0079] A method for inverting the chip temperature of a high-voltage power module provided by this application first constructs a high-fidelity finite element model of the high-voltage power module, conducts simulations under boundary conditions such as different current levels and different radiator temperatures to obtain the chip temperature and the case temperature, and constructs a data set based on the obtained simulation data. Then, a multi-input multi-output deep neural network is built, and the non-linear mapping relationship between the measurable point temperature and the chip temperature is obtained through training with the simulation data set. Finally, an experiment is carried out to obtain a small-sample experimental data set of the chip temperature and the case temperature under actual working conditions, and the trained deep neural network is fine-tuned using the small-sample experimental data set to correct the non-linear relationship and improve the accuracy of chip temperature inversion.

[0080] The above is a method for inverting the chip temperature of a high-voltage power module provided in the embodiments of this application. The following is a system for inverting the chip temperature of a high-voltage power module provided in the embodiments of this application.

[0081] Please refer to Figure 12 , a system for inverting the chip temperature of a high-voltage power module provided in the embodiments of this application, includes:

[0082] A construction unit 201, configured to construct a high-fidelity finite element model of the high-voltage power module to be inverted and verify the correctness of the high-fidelity finite element model.

[0083] A simulation unit 202, configured to perform simulations based on the high-fidelity finite element model after passing the correctness verification and obtain the chip temperature and the case temperature for constructing a data set according to preset boundary conditions, where the preset boundary conditions include: different current levels and different radiator temperatures.

[0084] A training unit 203, configured to build a multi-input multi-output deep neural network and train the multi-input multi-output deep neural network with the preprocessed data set to obtain a deep neural network for characterizing the non-linear mapping relationship between the measurable point temperature and the chip temperature.

[0085] An acquisition unit 204, configured to build a two-level inverter using the high-voltage power module, set the control algorithm and working parameters, control the magnitude of the load current by adjusting the load resistance, and simultaneously detect to obtain a sample experimental data set of several groups of chip temperatures and case temperatures.

[0086] An adjustment unit 205, configured to fine-tune the parameters of the trained deep neural network using the sample experimental data set based on the transfer learning theory, so as to correct the non-linear relationship and obtain a final deep neural network for inverting the chip temperature of the high-voltage power module.

[0087] Furthermore, in the embodiments of this application, a device for inverting the chip temperature of a high-voltage power module is also provided. The device includes a processor and a memory:

[0088] The memory is used to store program code and transmit the program code to the processor;

[0089] The processor is used to execute the steps of the high-voltage power module chip temperature inversion method described in the foregoing method embodiment according to the instructions in the program code.

[0090] Further, an embodiment of the present application also provides a computer-readable storage medium, which is used to store program code, and the program code is used to execute the high-voltage power module chip temperature inversion method described in the foregoing method embodiment.

[0091] Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of the systems and units described above can refer to the corresponding processes in the foregoing method embodiments, and will not be described herein again.

[0092] Terms such as "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects and do not have to be used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of the present application described herein can be implemented in an order different from those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units does not have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0093] It should be understood that in the present application, "at least one (item)" means one or more, and "a plurality" means two or more. "And / or" is used to describe the association relationship of associated objects and indicates that three relationships can exist. For example, "A and / or B" can mean: only A exists, only B exists, and both A and B exist at the same time. Among them, A and B can be singular or plural. The character " / " generally means that the associated objects before and after are in an "or" relationship. "At least one (one) of the following" or its similar expression means any combination of these items, including any combination of single item (one) or plural items (ones). For example, at least one (one) of a, b, or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0094] In several embodiments provided in this application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces, and the indirect couplings or communication connections of the devices or units can be in electrical, mechanical, or other forms.

[0095] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place, or they can be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0096] In addition, in each embodiment of this application, the functional units can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.

[0097] If the above-mentioned integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in each embodiment of this application. And the aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (English full name: Read-Only Memory, English abbreviation: ROM), random access memories (English full name: Random Access Memory, English abbreviation: RAM), magnetic disks, or optical discs that can store program codes.

[0098] As described above, the above embodiments are only used to illustrate the technical solutions of the present application and are not intended to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for inverting the temperature of a high-voltage power module chip, characterized in that Including: Construct a high-fidelity finite element model of the high-voltage power module to be inverted, and verify the correctness of the high-fidelity finite element model; Based on the high-fidelity finite element model after passing the correctness verification, and perform simulations according to the preset boundary conditions to obtain the chip temperature and the case temperature for constructing a data set, where the preset boundary conditions include: different current levels and different radiator temperatures; Build a multi-input multi-output deep neural network, and train the multi-input multi-output deep neural network with the preprocessed data set to obtain a deep neural network for characterizing the non-linear mapping relationship between the measurable temperature and the chip temperature; Use a high-voltage power module to build a two-level inverter, set the control algorithm and working parameters, then control the load current size by adjusting the load resistance, and at the same time perform detections to obtain several groups of sample experimental data sets of chip temperature and case temperature; Based on the transfer learning theory, use the sample experimental data set to fine-tune the parameters of the trained deep neural network, so as to correct the non-linear relationship and obtain the final deep neural network for inverting the chip temperature of the high-voltage power module.

2. The method for inverting the chip temperature of a high-voltage power module according to claim 1, wherein The construction of the high-fidelity finite element model of the high-voltage power module to be inverted specifically includes: Obtain the geometric dimensions of the high-voltage power module to be inverted and the material property parameters of each component of the high-voltage power module to be inverted, where the geometric dimensions include: chip size, radiator structure, solder layer thickness, package material thickness; According to the geometric dimensions and the material property parameters, use finite element analysis software to construct a high-fidelity finite element model of the high-voltage power module to be inverted.

3. The method for inverting the chip temperature of a high-voltage power module according to claim 1, wherein The verification of the correctness of the high-fidelity finite element model specifically includes: Use network elements of different sizes to respectively mesh the chip, ceramic and copper substrate in the high-fidelity finite element model; Use the average temperature and maximum temperature on the chip surface measured by an infrared camera and the finite element simulation calculation results to verify the correctness of the high-fidelity finite element model.

4. The method for inverting the chip temperature of a high-voltage power module according to claim 1, wherein The construction of the multi-input multi-output deep neural network specifically includes: Adopt a multi-layer perceptron structure as the network structure to build a multi-input multi-output deep neural network, where the network input includes: several measurable temperatures, and the network output includes several chip temperatures.

5. A high-voltage power module chip temperature inversion system, characterized in that, Including: A construction unit for constructing a high-fidelity finite element model of the high-voltage power module to be inverted and verifying the correctness of the high-fidelity finite element model; A simulation unit for performing simulations based on the high-fidelity finite element model after passing the correctness verification and according to the preset boundary conditions to obtain the chip temperature and the case temperature for constructing a data set, where the preset boundary conditions include: different current levels and different radiator temperatures; A training unit for building a multi-input multi-output deep neural network and training the multi-input multi-output deep neural network with the preprocessed data set to obtain a deep neural network for characterizing the non-linear mapping relationship between the measurable temperature and the chip temperature; An acquisition unit is configured to build a two-level inverter using a high-voltage power module, set control algorithms and operating parameters, and then control the magnitude of the load current by adjusting the load resistance. Meanwhile, several sets of sample experimental data sets of chip temperature and case temperature are obtained through detection. An adjustment unit is configured to fine-tune the parameters of the pre-trained deep neural network based on the transfer learning theory using the sample experimental data sets, so as to correct the non-linear relationship and obtain a final deep neural network for inverting the chip temperature of the high-voltage power module.

6. The high-voltage power module chip temperature inversion system according to claim 5, characterized in that, The construction of the high-fidelity finite element model of the high-voltage power module to be inverted specifically includes: Obtaining the geometric dimensions of the high-voltage power module to be inverted and the material property parameters of each component of the high-voltage power module to be inverted. Among them, the geometric dimensions include: chip size, radiator structure, welding layer thickness, and encapsulation material thickness. According to the geometric dimensions and the material property parameters, a high-fidelity finite element model of the high-voltage power module to be inverted is constructed using finite element analysis software.

7. The high-voltage power module chip temperature inversion system according to claim 5, characterized in that The verification of the correctness of the high-fidelity finite element model specifically includes: Meshing the chip, ceramic, and copper substrate in the high-fidelity finite element model respectively using network elements of different sizes. The average temperature and maximum temperature on the chip surface obtained by testing with an infrared camera and the finite element simulation calculation results are used to verify the correctness of the high-fidelity finite element model.

8. The high-voltage power module chip temperature inversion system according to claim 5, characterized in that, The construction of the multi-input multi-output deep neural network specifically includes: Using a multi-layer perceptron structure as the network structure to build a multi-input multi-output deep neural network. Among them, the network inputs include: several measurable point temperatures, and the network outputs include several chip temperatures.

9. A high-voltage power module chip temperature inversion device, characterized in that, The device includes a processor and a memory: The memory is used to store program codes and transmit the program codes to the processor. The processor is used to execute the high-voltage power module chip temperature inversion method according to any one of claims 1-4 based on the instructions in the program codes.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store program codes, and the program codes are used to execute the high-voltage power module chip temperature inversion method according to any one of claims 1-4.

Images