Power module junction temperature estimation method and device, electronic equipment and storage medium

Abstract

This invention provides a method, apparatus, electronic device, and storage medium for estimating the junction temperature of a power module. The method includes acquiring the current operating condition data and remaining lifetime of the current cycle of a target power module; inputting the switching frequency, voltage data, current data, and temperature data from the current operating condition data into a preset electrothermal network model, so that the electrothermal network model outputs the current junction temperature; the thermal impedance value in the electrothermal network model is variable; iteratively training the current junction temperature estimation neural network model based on the current operating condition data, remaining lifetime of the current cycle, and current junction temperature to obtain a new trained junction temperature estimation neural network model; and updating the junction temperature estimation neural network model of the corresponding vehicle end of the target power module to perform power module junction temperature estimation. This improves the reliability and accuracy of the electrothermal network model and the junction temperature estimation neural network model, reduces the computational load of vehicle-side power module junction temperature estimation, and reduces the load rate of the vehicle ECU.

Description

Technical Field

[0001] This application relates to the field of vehicle power module technology, specifically to a method, apparatus, electronic device, and storage medium for estimating the junction temperature of a power module. Background Technology

[0002] New energy vehicles have experienced rapid development in recent years. Power devices, as the core of motor control systems, determine the efficient operation of the entire system based on their reliability and safety. IGBT (Insulated Gate Bipolar Transistor) modules, also known as power modules, are characterized by high input impedance, high on-state current, and high withstand voltage, thus playing a crucial role in motor control systems. While increasing power ratings and switching frequencies, the reliability of IGBT modules is increasingly affected by junction temperature. Therefore, junction temperature is a key parameter for IGBT module condition monitoring.

[0003] Chinese patent CN107025364B discloses a method for predicting the junction temperature of an IGBT module. By establishing a high-order thermal capacity and thermal resistance model of the IGBT module, the power loss within a certain cycle is calculated to determine the junction temperature of the IGBT module within that cycle. However, this method does not consider the change in the thermal resistance of the IGBT module, resulting in insufficient reliability in the IGBT module junction temperature estimation. Furthermore, its high-order thermal capacity and thermal resistance model involves a large amount of computation, which can easily consume a large amount of computing resources in the vehicle and increase the load rate of the vehicle's ECU (Electronic Control Unit). Summary of the Invention

[0004] In view of the shortcomings of the prior art described above, this application provides a method, apparatus, electronic device and storage medium for estimating the junction temperature of a power module, in order to solve the technical problems of the above-mentioned related thermal network technology not considering the change of thermal impedance, the insufficient reliability of IGBT module junction temperature estimation, and the large computational resource consumption of thermal network technology, which increases the load rate of the vehicle ECU.

[0005] This application provides a method for estimating the junction temperature of a power module. The method includes: acquiring the current operating condition data and remaining lifetime of the current cycle of a target power module, wherein the operating condition data includes switching frequency, voltage data, current data, and temperature data; inputting the switching frequency, voltage data, current data, and temperature data from the current operating condition data into a preset electrothermal network model, so that the electrothermal network model outputs the current junction temperature, wherein the electrothermal network model is established based on the thermal impedance value of the target power module, and the thermal impedance value is variable; iteratively training the current junction temperature estimation neural network model based on the current operating condition data, the remaining lifetime of the current cycle, and the current junction temperature to obtain a new trained junction temperature estimation neural network model, and updating the junction temperature estimation neural network model of the corresponding vehicle end of the target power module to perform power module junction temperature estimation.

[0006] In one embodiment of this application, the switching frequency, voltage data, current data, and temperature data from the current operating condition data are input into a preset electrothermal network model. After the electrothermal network model outputs the current junction temperature, the power module junction temperature estimation method includes: determining the actual remaining lifetime of the target power module based on the historical junction temperature output by the electrothermal network model and the current junction temperature; calculating the lifetime degradation degree of the target power module based on the current cycle remaining lifetime and the actual remaining lifetime; if the lifetime degradation degree reaches a preset threshold, sending the actual remaining lifetime and the new junction temperature estimation neural network model to the vehicle end, so that the vehicle end determines the actual remaining lifetime as the remaining lifetime of the next cycle, and estimates the junction temperature of the target power module in the next cycle using the new junction temperature estimation neural network model.

[0007] In one embodiment of this application, determining the actual remaining lifetime of the target power module based on the historical junction temperature output by the electrothermal network model and the current junction temperature includes: acquiring multiple historical junction temperatures output by the electrothermal network model; fitting the multiple historical junction temperatures and the current junction temperature to obtain a junction temperature curve, and statistically analyzing the junction temperature curve using the rainflow counting method to obtain at least one set of thermal stresses and the number of power cycles under each set of thermal stresses, wherein the thermal stresses include junction temperature fluctuation and average junction temperature; inputting the junction temperature fluctuation and average junction temperature in each set of thermal stresses into the lifetime model so that the lifetime model outputs the number of failure power cycles under each set of thermal stresses, wherein the lifetime model is established based on the junction temperature fluctuation and average junction temperature; using the ratio of the number of power cycles under each set of thermal stresses to the number of failure power cycles as the fatigue damage value corresponding to each set of thermal stresses, and calculating the actual remaining lifetime based on the fatigue damage value corresponding to each set of thermal stresses.

[0008] In one embodiment of this application, if the lifetime degradation reaches a preset threshold, the power module junction temperature estimation method further includes: calculating a new thermal impedance value based on the current power loss, the temperature data in the current operating condition data, and the average junction temperature in each group of thermal stresses, to update the thermal impedance value in the electrothermal network model, wherein the current power loss is obtained based on the switching frequency, voltage data, and current data in the current operating condition data; or, calculating a new thermal impedance value based on the input power, temperature data, and the average junction temperature in each group of thermal stresses in the current operating condition data, to update the thermal impedance value in the electrothermal network model, wherein the operating condition data also includes the input power.

[0009] In one embodiment of this application, the establishment of the electrothermal network model includes: establishing a power loss model for calculating power loss based on switching frequency, voltage data, and current data; establishing a thermal network model for calculating junction temperature based on power loss, thermal impedance value, and temperature data; and coupling the power loss model and the thermal network model to obtain the electrothermal network model for junction temperature calculation.

[0010] In one embodiment of this application, a new junction temperature estimation neural network model is iteratively trained based on the current operating condition data, the current remaining lifetime of the current cycle, and the current junction temperature to obtain the trained neural network model. This includes: taking the current remaining lifetime of the current cycle and the pump speed, switching frequency, bus voltage, collector current, and cooling water temperature from the current operating condition data as input values, and the current junction temperature as the output value, and iteratively training the current junction temperature estimation neural network model to obtain the new junction temperature estimation neural network model. The operating condition data also includes the pump speed, the voltage data includes the bus voltage, and the current data includes the collector current.

[0011] In one embodiment of this application, before establishing the electrothermal network model, the power module junction temperature estimation method includes: obtaining the voltage drop of the target power module, where the voltage drop characterizes the aging degree of the power module; conducting an aging experiment on a sample power module until the voltage drop of the sample power module reaches the voltage drop of the target power module, wherein the model of the target power module is the same as that of the sample power module; calculating the thermal impedance value of the sample power module based on the junction temperature, casing temperature, and input power of the sample power module, and using it as the thermal impedance value of the target power module to establish the electrothermal network model.

[0012] In one embodiment of this application, a power module junction temperature estimation device is also provided. The power module junction temperature estimation device includes: an acquisition module, used to acquire the current operating condition data and the remaining lifetime of the current cycle of a target power module, the operating condition data including switching frequency, voltage data, current data, and temperature data; a calculation module, used to input the switching frequency, voltage data, current data, and temperature data from the current operating condition data into a preset electrothermal network model, so that the electrothermal network model outputs the current junction temperature, the electrothermal network model being established based on the thermal impedance value of the target power module, wherein the thermal impedance value is variable; a training module, used to iteratively train the current junction temperature estimation neural network model based on the current operating condition data, the remaining lifetime of the current cycle, and the current junction temperature to obtain a new trained junction temperature estimation neural network model; and an update module, used to update the junction temperature estimation neural network model of the vehicle end corresponding to the target power module to perform power module junction temperature estimation.

[0013] In one embodiment of this application, an electronic device is also provided, the electronic device comprising: one or more processors; and a storage device for storing one or more programs, which, when executed by the one or more processors, cause the electronic device to implement the power module junction temperature estimation method as described above.

[0014] In one embodiment of this application, a computer-readable storage medium is also provided, on which a computer program is stored, which, when executed by a computer's processor, causes the computer to perform the power module junction temperature estimation method as described above.

[0015] The beneficial effects of this invention are as follows: This invention provides a method, device, electronic device, and storage medium for estimating the junction temperature of a power module. The method considers the impact of lifetime decay on thermal resistance and junction temperature. It pre-establishes an electrothermal network model based on the changing thermal resistance value, and calculates the junction temperature using this changing model, thus improving the reliability of the electrothermal network model and ensuring its accuracy. It iteratively trains the current junction temperature estimation neural network model based on current operating data, remaining lifetime in the current cycle, and the current junction temperature output by the electrothermal network model, and updates the junction temperature estimation neural network model at the vehicle end, further improving its reliability. This ensures the accuracy of junction temperature calculation for the target power module at the vehicle end, reduces the computational load on the vehicle-side power module junction temperature estimation, and lowers the load rate of the vehicle's ECU.

[0016] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0017] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application. It is obvious that the drawings described below are merely some embodiments of this application, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort. In the drawings:

[0018] Figure 1 This is a schematic diagram illustrating the implementation environment of a power module junction temperature estimation method, as shown in an exemplary embodiment of this application.

[0019] Figure 2 This is a flowchart illustrating a power module junction temperature estimation method in an exemplary embodiment of this application;

[0020] Figure 3 This is a flowchart illustrating a specific embodiment of the present application for estimating the junction temperature of an IGBT module at the vehicle end;

[0021] Figure 4 This is a schematic diagram of the structure of a junction temperature estimation neural network model shown in a specific embodiment of this application;

[0022] Figure 5 This is a flowchart illustrating the vehicle-server interaction according to a specific embodiment of this application;

[0023] Figure 6 This is a flowchart illustrating the calculation of actual remaining lifetime, as shown in a specific embodiment of this application;

[0024] Figure 7 This is a flowchart illustrating a specific embodiment of the present application, showing the server-side estimation of IGBT module junction temperature and neural network training.

[0025] Figure 8 This is a block diagram illustrating a power module junction temperature estimation device according to an exemplary embodiment of this application;

[0026] Figure 9 A schematic diagram of the structure of a computer system suitable for implementing the electronic device of the present application is shown. Detailed Implementation

[0027] The embodiments of this application will be described below with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of this application from the content disclosed in this specification. This application can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this application. It should be understood that the preferred embodiments are only for illustrating this application and are not intended to limit the scope of protection of this application.

[0028] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of this application. Therefore, the drawings only show the components related to this application and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0029] It should be noted that in this application, terms such as "first" and "second" are merely for distinguishing similar objects, and do not limit the order or sequence of similar objects. The variations of "including" and "having" indicate that the scope covered by the subject of the word is not exclusive, except for the examples shown by the word.

[0030] It is understood that the various numerical designations, step numbers, and other identifiers recorded in this application are for descriptive convenience and are not intended to limit the scope of this application. The size of the identifiers in this application does not imply the order of execution; the execution order of each process should be determined by its function and internal logic.

[0031] In the following description, numerous details are explored to provide a more thorough explanation of embodiments of the present application. However, it will be apparent to those skilled in the art that embodiments of the present application may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring embodiments of the present application.

[0032] It should be noted that an IGBT module mainly consists of an IGBT chip, solder layer, copper-clad ceramic substrate, and base plate. With variations in load and operating conditions, the temperature of the IGBT chip fluctuates significantly, and the heat dissipated by the IGBT chip is conducted from top to bottom to the cooling system. Mismatches in the coefficients of thermal expansion between the materials within the IGBT module lead to thermal and mechanical stresses, accelerating the aging process of the module, causing solder layer fatigue, bond wire cracking or even detachment, and ultimately resulting in IGBT module failure. Therefore, the IGBT module junction temperature is a key parameter for IGBT module condition monitoring. The IGBT module junction temperature, also known as the power module junction temperature or IGBT junction temperature, refers to the highest operating temperature of the IGBT chip, and the lifespan assessment of the IGBT module depends on accurate junction temperature measurement. Quantitative studies on the relationship between IGBT module lifespan and IGBT module junction temperature mainly employ physical modeling and analytical modeling methods. Physical models are difficult to extract the stress magnitude of each layer within the device in practice, requiring specialized instruments and equipment, and their accuracy is difficult to verify. Analytical modeling, on the other hand, treats the internal structure as a whole, indirectly reflecting the process of internal thermal stress accumulation leading to material creep failure through external parameter measurements. The thermal network method, as the most commonly used analytical model, estimates the junction temperature of the IGBT module by calculating the average power loss of the IGBT module. However, the thermal network method does not consider the fatigue damage caused by load current surges during IGBT module operation; that is, the thermal impedance value in the thermal network is not updated. This leads to the estimated IGBT module junction temperature being lower than the actual result as the IGBT module's lifespan decreases. Furthermore, online estimation of IGBT module junction temperature using the thermal network method requires significant computational resources, increasing the load on the vehicle ECU.

[0033] To address the aforementioned problems, embodiments of this application propose a power module junction temperature estimation method, a power module junction temperature estimation device, an electronic device, a computer-readable storage medium, and a computer program product, which will be described in detail below.

[0034] Please see Figure 1 , Figure 1 This is a schematic diagram illustrating the implementation environment of a power module junction temperature estimation method, as shown in an exemplary embodiment of this application.

[0035] like Figure 1As shown, the implementation environment can include vehicle-side 101 and server-side 102. Server-side 102 can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms. No restrictions are imposed here. Vehicle-side 101 collects the current operating condition data of the target power module through sensors and other means, and sends the current operating condition data and the remaining lifetime of the current cycle of the target power module to server-side 102 so that server-side 102 can iteratively train the current junction temperature estimation neural network model. It should be noted that the remaining lifetime of the current cycle of the target power module is stored in vehicle-side 101.

[0036] For example, server 102 obtains the current operating condition data and remaining lifetime of the current cycle of the target power module. The operating condition data includes switching frequency, voltage data, current data, and temperature data. The switching frequency, voltage data, current data, and temperature data in the current operating condition data are input into a preset electrothermal network model so that the electrothermal network model outputs the current junction temperature. The electrothermal network model is established based on the thermal impedance value of the target power module, and the thermal impedance value is variable. Based on the current operating condition data, the remaining lifetime of the current cycle, and the current junction temperature, the current junction temperature estimation neural network model is iteratively trained to obtain a new junction temperature estimation neural network model. The junction temperature estimation neural network model of the vehicle terminal 101 corresponding to the target power module is updated to perform junction temperature estimation of the power module. As can be seen, the technical solution of this application embodiment pre-establishes an electrothermal network model on the server side based on the changing thermal impedance value, and performs junction temperature calculation through the changing electrothermal network model, thereby improving the reliability of the electrothermal network model and ensuring the accuracy of the junction temperature calculation. Based on the current operating condition data, the remaining life of the current cycle, and the current junction temperature output by the electrothermal network model, the current junction temperature estimation neural network model is iteratively trained, and the junction temperature estimation neural network model on the vehicle side is updated, thereby improving the reliability of the junction temperature estimation neural network model, thereby ensuring the accuracy of the junction temperature calculation of the target power module on the vehicle side, and reducing the calculation amount of junction temperature estimation of the power module on the vehicle side, thus reducing the load rate of the vehicle ECU.

[0037] It should be noted that the power module junction temperature estimation method provided in this application embodiment is generally executed by the server 102, and the corresponding power module junction temperature estimation device is generally set in the server 102.

[0038] Please see Figure 2 , Figure 2This is a flowchart illustrating a power module junction temperature estimation method according to an exemplary embodiment of this application. This method can be applied to... Figure 1 The implementation environment shown is specifically executed by server 102 within that implementation environment. It should be understood that this method can also be applied to other exemplary implementation environments and specifically executed by devices in other implementation environments. This embodiment does not limit the implementation environment to which this method is applicable.

[0039] like Figure 2 As shown, in an exemplary embodiment, the power module junction temperature estimation method includes at least steps S210 to S230, which are described in detail below:

[0040] Step S210: Obtain the current operating condition data and remaining lifespan of the target power module for the current cycle.

[0041] In one embodiment of this application, the vehicle-side acquires current operating condition data of the target power module via sensors or other means. This operating condition data includes switching frequency, voltage data, current data, and temperature data. The voltage data includes at least one of the on-state voltage drop and bus voltage; the current data includes at least one of the collector current and load current; and the temperature data includes at least one of the target power module's casing temperature and coolant temperature. The vehicle-side can actively upload the acquired current operating condition data and the stored remaining lifespan of the current cycle to the server in real time or periodically. Alternatively, the server can send acquisition requests to the vehicle-side in real time or periodically to obtain the current operating condition data and the remaining lifespan of the current cycle; no limitation is imposed here. It should be noted that coolant, also called coolant fluid, can specifically be pure water, or other liquid media such as oil or ethylene glycol; no limitation is imposed here.

[0042] Step S220: Input the switching frequency, voltage data, current data and temperature data from the current operating condition data into the preset electrothermal network model so that the electrothermal network model outputs the current junction temperature.

[0043] In one embodiment of this application, a pre-defined electrothermal network model is provided in the server. This model is established based on the thermal impedance value of the target power module, and the thermal impedance value is variable, increasing as the lifespan of the target power module decreases. Typically, the junction temperature of an IGBT module fluctuates due to variations in load current. Current impact on the IGBT module generates thermal stress that interacts between the various structural layers, repeatedly causing fatigue damage and ultimately leading to device failure. During this process, the thermal impedance value of the IGBT module also increases with its lifespan. If this change in thermal impedance value is not considered, the calculated junction temperature of the IGBT module will be too low, deviating from the actual situation. Therefore, this embodiment of the application pre-establishes an electrothermal network model based on the changing thermal impedance value. The junction temperature is calculated using the electrothermal network model, switching frequency, voltage data, current data, and temperature data, ensuring the reliability of the electrothermal network model and thus the accuracy of the junction temperature calculation.

[0044] In one embodiment of this application, the establishment of the electrothermal network model includes: establishing a power loss model for calculating power loss based on switching frequency, voltage data, and current data; establishing a thermal network model for calculating junction temperature based on power loss, thermal impedance value, and temperature data; and coupling the power loss model and the thermal network model to obtain the electrothermal network model for junction temperature calculation.

[0045] In this embodiment, a power loss model for the digital twin IGBT module corresponding to the target power module is established. This power loss model is used to calculate the on-state power loss and switching power loss of the target power module, thereby obtaining the power loss. The on-state power loss is related to the on-state voltage drop in the voltage data and the collector current in the current data. The switching power loss is related to the switching frequency, the bus voltage in the voltage data, and the load current in the current data. The power loss consists of the on-state power loss and the switching power loss. Therefore, the equation for the power loss model is as follows:

[0046]

[0047] Among them, P loss For power loss, P cond P is the on-state power loss. sw For switching power loss, U ce For the on-state voltage drop, I c Where is the collector current and m is the modulation ratio. For power factor, R T f is the internal resistance of the IGBT. sw E is the switching frequency. on To activate the energy, E off To shut off energy, U dc U is the bus voltage.nom For the rated voltage, I m I is the load current. nom This refers to the rated current. It should be noted that the modulation ratio, power factor, and IGBT internal resistance are relatively fixed parameters that do not change significantly in practical applications. These parameters can be obtained through calibration tests or the IGBT datasheet corresponding to the target power module. Turn-on energy, turn-off energy, rated voltage, and rated current can be obtained from the IGBT datasheet corresponding to the target power module.

[0048] A thermal network model of the digital twin IGBT module corresponding to the target power module is established. Since the junction temperature of the target power module is related to the power loss, thermal impedance, and casing temperature, the equations of the thermal network model are as follows:

[0049] T j =P loss Z th +T c Equation (2)

[0050] Among them, T j For the junction temperature, P loss For power loss, Z th T is the thermal resistance value. c This refers to the outer casing temperature.

[0051] The power loss model and the thermal network model are coupled to establish a corresponding electrothermal network model. This electrothermal network model is then used to estimate the junction temperature of the target power module. For example, the switching frequency, voltage, current, and temperature data from the current operating conditions are input into the electrothermal network model. The power loss model calculates the current conduction loss power based on the on-state voltage drop and collector current from the current operating conditions data, and calculates the current switching loss power based on the switching frequency, bus voltage, and load current from the current operating conditions data. The sum of the current conduction loss power and the current switching loss power is determined as the current power loss. The thermal network model calculates the current junction temperature based on the thermal impedance value, the casing temperature from the current operating conditions data, and the current power loss output by the power loss model.

[0052] Furthermore, since there is a linear relationship between the shell temperature and the cooling water temperature, a thermal network model can be established based on the power loss, thermal resistance value, and cooling water temperature. This allows the thermal network model to calculate the current junction temperature based on the thermal resistance value, the cooling water temperature in the current operating data, and the current power loss output by the power loss model.

[0053] In one embodiment of this application, before establishing the electrothermal network model, the power module junction temperature estimation method includes: obtaining the voltage drop of the target power module, where the voltage drop characterizes the aging degree of the power module; conducting an aging experiment on a sample power module until the voltage drop of the sample power module reaches the voltage drop of the target power module, wherein the model of the target power module is the same as that of the sample power module; and calculating the thermal impedance value of the sample power module based on the junction temperature, casing temperature, and input power of the sample power module, using it as the thermal impedance value of the target power module to establish the electrothermal network model.

[0054] In this embodiment, an aging experiment is conducted on the sample power module through power cycling. Heating is stopped when the temperature reaches the upper limit and starts when it reaches the lower limit. The cyclic thermal stress continuously impacts the sample power module to achieve the aging purpose. During the aging experiment, the voltage drop of the sample power module is observed. If the voltage drop of the sample power module reaches the voltage drop of the target power module, heating is stopped. The junction temperature, shell temperature, and input power of the sample power module are measured during the cooling process. The input power of the sample power module is taken as the power loss of the sample power module. Based on the power loss, the junction temperature and shell temperature of the sample power module, the thermal impedance value of the sample power module is calculated according to equation (2) and taken as the thermal impedance value of the target power module. An electrothermal network model is established based on the thermal impedance value. It should be noted that when the IGBT module is heated with a fixed current and reaches thermal equilibrium, the power loss is the same as the input power. In addition, the transient thermal impedance value under different aging degrees can be obtained according to the magnitude of the cumulative damage, that is, the thermal impedance value under different aging degrees, so as to obtain the remaining lifetime-thermal impedance value curve.

[0055] Step S230: Based on the current operating condition data, the remaining life of the current cycle, and the current junction temperature, the current junction temperature estimation neural network model is iteratively trained to obtain a new junction temperature estimation neural network model after training. The junction temperature estimation neural network model of the vehicle end corresponding to the target power module is then updated to perform junction temperature estimation of the power module.

[0056] In one embodiment of this application, since the IGBT module's lifetime decay affects the IGBT module junction temperature, and the accuracy of the neural network depends on a large amount of differentially distributed training data, the neural network model trained with parameters not under actual driving conditions suffers from underfitting. Therefore, the remaining lifetime, switching frequency, voltage data, current data, and temperature data can be used together as feature parameters of the neural network, with the junction temperature as the output of the neural network. The neural network is then trained, and the trained neural network is determined as the junction temperature estimation neural network model. The current junction temperature estimation neural network model is then iteratively trained using the remaining lifetime of the current cycle, the current junction temperature, and the switching frequency, voltage data, current data, and temperature data from the current operating conditions. This embodiment of the application considers the impact of the target power module's lifetime decay on the target power module's junction temperature and combines it with actual automotive operating condition data. The large and differentially distributed data sample size allows for more accurate training of the junction temperature estimation neural network model, improving its reliability. This ensures the accuracy of the vehicle-side junction temperature calculation for the target power module and reduces the computational load on the vehicle-side power module junction temperature estimation, thus lowering the load rate of the vehicle's ECU.

[0057] In this embodiment, the server can periodically send the trained new junction temperature estimation neural network model to the vehicle, enabling the vehicle to estimate the junction temperature of the power module using the new model. For example, every day, the server uploads the new model to an OTA (Over-the-Air) platform, and pushes an upgrade package to the vehicle via the OTA platform to update the model.

[0058] For the vehicle side, in the current cycle, the junction temperature of the power module is estimated using the current junction temperature estimation neural network model. In the next cycle, the junction temperature of the power module is estimated using a new junction temperature estimation neural network model. This can reduce the amount of computation on the vehicle side and reduce the computational load on the automotive chips.

[0059] In another embodiment of this application, the current junction temperature estimation neural network model is iteratively trained based on the current operating condition data, the remaining lifetime of the current cycle, and the current junction temperature to obtain a new junction temperature estimation neural network model. This includes: taking the remaining lifetime of the current cycle, the switching frequency, bus voltage, collector current, and cooling water temperature in the current operating condition data as input values, and the current junction temperature as the output value, and iteratively training the current junction temperature estimation neural network model to obtain a new junction temperature estimation neural network model. The operating condition data also includes the water pump speed, the voltage data includes the bus voltage, and the current data includes the collector current.

[0060] In this embodiment, the pump speed refers to the rotational speed of the cooling water pump in the cooling system of the target power module. Since the pump speed also affects the IGBT module junction temperature, the switching frequency, bus voltage, collector current, cooling water temperature, and remaining lifetime can be used as feature parameters of the neural network, along with the pump speed. The junction temperature is used as the output of the neural network for training. The trained neural network is then determined as the junction temperature estimation neural network model. The model is then iteratively trained using the remaining lifetime of the current cycle, the current junction temperature, and the switching frequency, bus voltage, collector current, cooling water temperature, and pump speed from the current operating data. This embodiment uses the pump speed as an input for iterative training, further improving the accuracy of the new junction temperature estimation neural network model.

[0061] Please see Figure 3 , Figure 3 This is a flowchart illustrating the estimation of IGBT module junction temperature at the vehicle end, as shown in a specific embodiment of this application. Figure 3 As shown, the process for estimating the junction temperature of the IGBT module at the vehicle end is as follows: The vehicle end acquires the switching frequency f under real-time operating conditions. sw Collector current I c Bus voltage U dc Motor speed n, cooling water temperature T cool , will f sw I c U dc n, T cool The current IGBT module lifetime c1 is input to the vehicle-side junction temperature estimation neural network model, which is the current junction temperature estimation neural network model for estimating the junction temperature T of the IGBT module. j And output.

[0062] Please see Figure 4 , Figure 4 This is a schematic diagram illustrating the structure of a junction temperature estimation neural network model according to a specific embodiment of this application. Figure 4 As shown, the input layer of the junction temperature calculation neural network is the switching frequency f. sw Collector current I c Bus voltage U dc Motor speed n, cooling water temperature T cool And the current lifetime c1 of the IGBT module, the junction temperature T output by the electrothermal network model of the output layer. j The switching frequency f under real-time operating conditions sw Collector current I c Bus voltage U dc Motor speed n, cooling water temperature T coolUsing the current lifetime c1 of the IGBT module as a feature parameter of the neural network, the junction temperature T calculated based on the electrothermal network model is... j As the output of the neural network, the neural network under the current lifetime, i.e. the current junction temperature estimation neural network model, is iteratively trained to obtain a new round of IGBT module junction temperature estimation neural network model, i.e., the new junction temperature estimation neural network model.

[0063] In one embodiment of this application, after step S220, the power module junction temperature estimation method includes: determining the actual remaining lifetime of the target power module based on the historical junction temperature and the current junction temperature output by the electrothermal network model; calculating the lifetime degradation degree of the target power module based on the current cycle remaining lifetime and the actual remaining lifetime; if the lifetime degradation degree reaches a preset threshold, sending the actual remaining lifetime and a new junction temperature estimation neural network model to the vehicle end, so that the vehicle end determines the actual remaining lifetime as the remaining lifetime of the next cycle, and estimates the junction temperature of the target power module in the next cycle through the new junction temperature estimation neural network model.

[0064] In this embodiment, when the lifetime degradation of the target power module is relatively minor, the junction temperature estimation error between the new and old junction temperature estimation neural network models is relatively small. However, when the lifetime degradation of the target power module is severe, the estimation error between the two is large. Frequent updates to the junction temperature estimation neural network model on the vehicle side can easily lead to long-term bandwidth resource occupation. Therefore, it is possible to determine whether a new junction temperature estimation neural network model needs to be pushed to the vehicle side based on the degree of lifetime degradation of the target power module. The actual remaining lifetime of the target power module can be determined based on the rainflow counting method and the lifetime model, thereby determining the degree of lifetime degradation of the target power module. The difference between the remaining lifetime of the current cycle and the actual remaining lifetime can be used as the degree of lifetime degradation. When the degree of lifetime degradation reaches a preset threshold, a new junction temperature estimation neural network model is pushed to the vehicle side, and the remaining lifetime of the next cycle is also pushed to the vehicle side, so that the vehicle side can perform junction temperature estimation of the target power module in the next cycle based on the new junction temperature estimation neural network model and the remaining lifetime of the next cycle. It should be noted that the duration of the current cycle and the next cycle are not necessarily the same, but depend on the degree of lifetime degradation. When the degree of lifetime degradation reaches the preset threshold, it indicates that the junction temperature estimation of the target power module by the vehicle side has moved from the current cycle to the next cycle. The preset threshold can be 2% or other values, and is not limited here. This embodiment determines the update time of the junction temperature estimation neural network model at the vehicle end by determining the degree of lifetime degradation of the target power module. This ensures both the reliability and accuracy of the junction temperature estimation of the target power module at the vehicle end, and also reasonably reduces the occupation of bandwidth resources.

[0065] Please see Figure 5 , Figure 5 This is a flowchart illustrating the vehicle-server interaction according to a specific embodiment of this application. Figure 5 As shown, the vehicle-side (vehicle-side) is responsible for sending real-time operating condition data (including current operating condition data and remaining lifespan for the current cycle) to the big data platform (server-side), receiving the new junction temperature estimation neural network model pushed through the OTA platform, and performing offline IGBT module junction temperature calculations. The big data platform is responsible for receiving the real-time operating condition data sent by the vehicle-side, calculating the IGBT module junction temperature for the current lifespan online, updating the current junction temperature estimation neural network model, and determining the degree of IGBT module lifespan degradation. When the lifespan degradation reaches a preset threshold, the big data platform sends a push request to the vehicle-side, which decides whether to update the junction temperature estimation neural network model. If so, the new junction temperature estimation neural network model will be updated to the vehicle-side through the OTA platform, enabling the vehicle-side to estimate the IGBT module junction temperature for the next cycle using the new model.

[0066] In one embodiment of this application, determining the actual remaining lifetime of a target power module based on the historical junction temperature and the current junction temperature output by an electrothermal network model includes: acquiring multiple historical junction temperatures output by the electrothermal network model; fitting the multiple historical junction temperatures and the current junction temperature to obtain a junction temperature curve, and statistically analyzing the junction temperature curve using a rainflow counting method to obtain at least one set of thermal stresses and the number of power cycles under each set of thermal stresses, wherein the thermal stresses include junction temperature fluctuation and average junction temperature; inputting the junction temperature fluctuation and average junction temperature in each set of thermal stresses into a lifetime model so that the lifetime model outputs the number of failure power cycles under each set of thermal stresses, wherein the lifetime model is established based on the junction temperature fluctuation and average junction temperature; using the ratio of the number of power cycles under each set of thermal stresses to the number of failure power cycles as the fatigue damage value corresponding to each set of thermal stresses, and calculating the actual remaining lifetime based on the fatigue damage value corresponding to each set of thermal stresses.

[0067] Please see Figure 6 , Figure 6 This is a flowchart illustrating the calculation of actual remaining lifetime, as shown in a specific embodiment of this application.

[0068] like Figure 6 As shown, the calculation process for actual remaining lifespan is as follows:

[0069] 1. The rainflow counting method is used to statistically analyze the fatigue load (i.e., the number of power cycles under each group of thermal stresses) of the historical junction temperature curve (junction temperature curve). Using the rainflow counting method, the historical junction temperature data (multiple historical junction temperatures and the current junction temperature) are analyzed according to the basic elements of power cycles: average stress (average junction temperature T). jm ) and stress fluctuation (junction temperature fluctuation ΔT) jThe stress is categorized and statistically analyzed to determine various stresses at different time points. Based on the corresponding rainfall flow and time, a cycle number is calculated, resulting in a stress-cycle number curve, which provides data for fatigue damage research. The specific process of the rainflow counting method is as follows:

[0070] 1)T j Indicates the junction temperature, ΔT j T represents the junction temperature fluctuation. jm This represents the average junction temperature. The "rainflow" starts from the starting point and flows downwards along the curve from the maximum points of the historical junction temperature curve.

[0071] 2) The "rain flow" starts from a certain extreme point and falls when it reaches the extreme point, until it encounters an extreme point larger than its starting point and stops flowing.

[0072] 3) During the dripping process of the "rain stream", it stops flowing once it encounters the previous "rain stream". At this time, the trajectory of the "rain stream" forms a loop. Record the number of loops and the amplitude of each loop.

[0073] 4) Count all the loops that meet the requirements, and concatenate the remaining data. Repeat the above three steps.

[0074] 5) Calculate the ΔT under each stress condition (each group of thermal stresses). j and T jm The number of power cycles.

[0075] 2. A Less lifetime model is established based on power cycling experiments and the damage accumulation rule. Junction temperature fluctuation ΔT is considered. j Average junction temperature T jm The impact of two parameters on IGBT modules was investigated. Based on power cycling experiments and the damage accumulation rule, a Less lifetime model was established. The calculation formula for the lifetime model is as follows:

[0076]

[0077] Where, N f Let ΔT be the failure cycle number, A and α be the fitting parameters, and ΔT be the value of ΔT. j E represents the junction temperature fluctuation. α The activation energy of the material is 9.89 × 10⁻⁶. -20 J, k B The Boltzmann constant is 1.38 × 10⁻⁶. -23 J.K. -1 T jm This represents the average junction temperature.

[0078] ΔT under different stress conditions j and T jm Substituting into equation (3), we obtain the stress conditions ΔT.j and T jm The number of failure power cycles when the IGBT fails.

[0079] 3. Calculate the actual remaining life c2. Apply the statistically obtained fatigue load to the life model to calculate the actual remaining life of the target power module. First, summarize the temperature fluctuation ΔT. j and average junction temperature T jm The corresponding failure power cycle number is used to obtain a three-dimensional histogram, where the Z-axis represents the failure power cycle number N. f The X-axis represents the junction temperature fluctuation ΔT. j The Y-axis represents the average junction temperature T. jm The ΔT under the statistically obtained junction temperature history curve j and T jm Corresponding to this lifetime model, calculate the number of power cycles N under each stress condition. cyc And the number of failure power cycles N under this stress condition f The ratios between these ratios are summed to obtain the cumulative fatigue damage value D, which is used to calculate the actual remaining life. The formula for calculating the actual remaining life is as follows:

[0080]

[0081] Where c2 is the actual remaining life, D is the cumulative fatigue damage value, and N is the total fatigue life. cyc N represents the number of power cycles. f This represents the number of cycles for the failure power.

[0082] In one embodiment of this application, if the degree of lifetime degradation reaches a preset threshold, the power module junction temperature estimation method further includes: calculating a new thermal impedance value based on the current power loss, the temperature data in the current operating condition data, and the average junction temperature in each group of thermal stresses, so as to update the thermal impedance value in the electrothermal network model. The current power loss is obtained based on the switching frequency, voltage data, and current data in the current operating condition data.

[0083] In this embodiment, even when the IGBT module operates under the same conditions, the calculated junction temperature of the IGBT module will vary due to the change in thermal impedance value caused by the accumulation of fatigue damage. Considering the impact of the target power module's lifespan decay on its thermal impedance value, when the lifespan decay reaches a preset threshold, the thermal impedance value in the electrothermal network model is updated. This ensures the reliability and accuracy of the junction temperature estimation in the electrothermal network model during the continuous decay of the target power module's lifespan. Specifically, the update method is to perform an average or weighted average calculation based on the average junction temperature in each group of thermal stresses to obtain the average value of all average junction temperatures, which is used as the current cycle average junction temperature. The new thermal impedance value is calculated according to equation (2) based on the current cycle average junction temperature, the current power loss, and the shell temperature in the current operating condition data. The thermal impedance value in the electrothermal network model is then updated to the new thermal impedance value, completing the update of the electrothermal network model. The current power loss can be calculated according to equation (1) based on the switching frequency, voltage data, and current data in the current operating condition data. The current casing temperature can also be estimated based on the linear relationship between the cooling water temperature and the casing temperature in the current operating data. The new thermal resistance value can be calculated according to Equation (2) based on the current cycle average junction temperature, the current power loss, and the estimated current casing temperature.

[0084] In another embodiment of this application, if the degree of lifetime degradation reaches a preset threshold, the power module junction temperature estimation method further includes: calculating a new thermal impedance value based on the input power, temperature data and the average junction temperature in each group of thermal stresses in the current operating condition data, so as to update the thermal impedance value in the electrothermal network model. The operating condition data also includes the input power.

[0085] In this embodiment, the input power in the current operating data can also be used as the current loss power to calculate the new thermal impedance value according to the method in the above embodiment.

[0086] In another embodiment of this application, if the degree of lifetime degradation does not reach a preset threshold, neither the junction temperature estimation neural network model at the vehicle end nor the thermal impedance value will be updated.

[0087] The technical solution of this application adds the degree of lifetime degradation of the target power module as a dimension for calculating the junction temperature of the target power module, thereby improving the accuracy of junction temperature calculation. Secondly, this method uses a neural network to perform offline junction temperature calculation of the target power module at the vehicle end, significantly reducing the computational load on the onboard chip. Simultaneously, the actual remaining lifetime of the target power module is calculated, and the current junction temperature estimation neural network model is updated and trained online in real time. Then, the new junction temperature estimation neural network model is pushed to the vehicle end via OTA to conduct a new round of power module junction temperature calculation.

[0088] Please see Figure 7 , Figure 7 This is a flowchart illustrating a specific embodiment of the present application, showing the server-side estimation of IGBT module junction temperature and neural network training. (See attached flowchart.) Figure 7 As shown, the server obtains the switching frequency f under the real-time operating conditions of the vehicle. sw Collector current I c Load current I m Bus voltage U dc , conduction voltage drop U ce Motor speed n, cooling water temperature T cool Casing temperature T c The current lifespan c1 of the IGBT module is determined. A high-precision power loss model and a high-precision thermal network model of the digital twin IGBT module corresponding to the vehicle-side IGBT module are established. These two models are coupled to obtain an electrothermal network model, which is used to calculate the IGBT module junction temperature T. j f sw I c U dc n, T cool Using c1 as input and c1 as output, the current junction temperature estimation neural network model is iteratively trained to obtain a new, trained model, which updates the server-side junction temperature estimation neural network model. Based on the junction temperature output by the electrothermal network model, the actual remaining lifespan of the IGBT module is calculated, thus determining the IGBT module's lifespan degradation level. When the lifespan degradation reaches 2%, the new junction temperature estimation neural network model is uploaded to the OTA platform, enabling the OTA platform to push an OTA upgrade package including the new model to the vehicle, updating the vehicle-side junction temperature estimation neural network model and the thermal impedance value, thereby updating the thermal network model (high-precision thermal network model) within the electrothermal network model. If the lifespan degradation is less than 2%, neither the vehicle-side junction temperature estimation neural network model nor the thermal impedance value is updated. Figure 7 For detailed procedures, please refer to the descriptions in the foregoing embodiments; they will not be repeated here. The technical solution in this embodiment employs a neural network model that considers lifetime decay when estimating junction temperature. This reduces the online computational load on the vehicle and the computational burden on the onboard chip. Furthermore, by considering lifetime decay when calculating junction temperature and combining it with real-time vehicle operating data, the large and varied sample size of the data allows for more accurate training of the junction temperature estimation neural network. Simultaneously, the vehicle data is uploaded to a big data platform for iterative interaction with lifetime assessment, enabling better safety, health, and fault warnings for the vehicle.

[0089] It should be noted that the IGBT module in each specific embodiment of this application, the current lifespan c1 of the IGBT module under real-time operating conditions, and the switching frequency f are all different. sw Collector current I c Load current I m Bus voltage U dc , conduction voltage drop U ce Motor speed n, cooling water temperature T cool Casing temperature T c These correspond sequentially to the target power module, remaining lifetime of the current cycle, switching frequency, collector current, load current, bus voltage, on-state voltage drop, water pump speed, cooling water temperature, and casing temperature in the various embodiments of this application.

[0090] Please see Figure 8 , Figure 8 This is a block diagram illustrating a power module junction temperature estimation device according to an exemplary embodiment of this application. The device can be applied to... Figure 1 The implementation environment shown is specifically configured in server 102. This device can also be applied to other exemplary implementation environments and specifically configured in other devices. This embodiment does not limit the implementation environment to which this device is applicable.

[0091] like Figure 8 As shown, this exemplary power module junction temperature estimation device includes:

[0092] The acquisition module 810 is configured to acquire the current operating condition data and remaining lifetime of the current cycle of the target power module. The operating condition data includes switching frequency, voltage data, current data, and temperature data. The calculation module 820 is configured to input the switching frequency, voltage data, current data, and temperature data from the current operating condition data into a preset electrothermal network model so that the electrothermal network model outputs the current junction temperature. The electrothermal network model is established based on the thermal impedance value of the target power module, and the thermal impedance value is variable. The training module 830 is configured to iteratively train the current junction temperature estimation neural network model based on the current operating condition data, the remaining lifetime of the current cycle, and the current junction temperature to obtain a new junction temperature estimation neural network model after training. The update module 840 is configured to update the junction temperature estimation neural network model of the vehicle end corresponding to the target power module to perform junction temperature estimation of the power module.

[0093] It should be noted that the power module junction temperature estimation device and the power module junction temperature estimation method provided in the above embodiments belong to the same concept. The specific operation methods of each module and unit have been described in detail in the method embodiments and will not be repeated here. In practical applications, the power module junction temperature estimation device provided in the above embodiments can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. This is not a limitation here.

[0094] Embodiments of this application also provide an electronic device, including: one or more processors; and a storage device for storing one or more programs, which, when executed by the one or more processors, cause the electronic device to implement the power module junction temperature estimation method provided in the above embodiments.

[0095] Please see Figure 9 , Figure 9 A schematic diagram of a computer system suitable for implementing the embodiments of this application is shown. It should be noted that... Figure 9 The computer system 900 of the electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0096] like Figure 9 As shown, the computer system 900 includes a Central Processing Unit (CPU) 901, which can perform various appropriate actions and processes based on programs stored in Read-Only Memory (ROM) 902 or programs loaded from Storage Unit 908 into Random Access Memory (RAM) 903, such as performing the methods described in the above embodiments. The RAM 903 also stores various programs and data required for system operation. The CPU 901, ROM 902, and RAM 903 are interconnected via a bus 904. An Input / Output (I / O) interface 905 is also connected to the bus 904.

[0097] The following components are connected to I / O interface 905: an input section 906 including a keyboard, mouse, etc.; an output section 907 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 908 including a hard disk, etc.; and a communication section 909 including a network interface card such as a LAN (Local Area Network) card, modem, etc. The communication section 909 performs communication processing via a network such as the Internet. A drive 910 is also connected to I / O interface 905 as needed. Removable media 911, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., are installed on drive 910 as needed so that computer programs read from them can be installed into storage section 908 as needed.

[0098] Specifically, according to embodiments of this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program including a computer program for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 909, and / or installed from removable medium 911. When the computer program is executed by central processing unit (CPU) 901, it performs various functions defined in the system of this application.

[0099] It should be noted that the computer-readable medium shown in the embodiments of this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, portable compact disc read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this application, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying a computer-readable computer program. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media can also be any computer-readable medium other than computer-readable storage media, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The computer program contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to wireless, wired, etc., or any suitable combination thereof.

[0100] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods, and computer program products according to various embodiments of this application. Each block in a flowchart or block diagram may represent a module, segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, or they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0101] The units described in the embodiments of this application can be implemented in software or hardware, and the described units can also be located in a processor. The names of these units do not necessarily limit the specific unit itself.

[0102] Another aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a computer's processor, causes the computer to perform the power module junction temperature estimation method as described above. This computer-readable storage medium may be included in the electronic device described in the above embodiments, or it may exist independently and not assembled into the electronic device.

[0103] Another aspect of this application provides a computer program product or computer program including computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the power module junction temperature estimation method provided in the various embodiments described above.

[0104] The above embodiments are merely illustrative of the principles and effects of this application and are not intended to limit this application. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of this application. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in this application should still be covered by the claims of this application.

Claims

1. A method for estimating the junction temperature of a power module, characterized in that, The power module junction temperature estimation method includes: Obtain the current operating condition data and remaining lifespan of the target power module for the current cycle. The operating condition data includes switching frequency, voltage data, current data, and temperature data. The switching frequency, voltage data, current data, and temperature data from the current operating condition data are input into a preset electrothermal network model so that the electrothermal network model outputs the current junction temperature. The electrothermal network model is established based on the thermal resistance value of the target power module, and the thermal resistance value is variable. Based on the current operating data, the remaining lifetime of the current cycle, and the current junction temperature, the current junction temperature estimation neural network model is iteratively trained to obtain a new junction temperature estimation neural network model. The junction temperature estimation neural network model at the vehicle end corresponding to the target power module is then updated to estimate the power module junction temperature. Updating the junction temperature estimation neural network model at the vehicle end corresponding to the target power module includes: determining the actual remaining lifetime of the target power module based on the historical junction temperature output by the electrothermal network model and the current junction temperature; calculating the lifetime degradation degree of the target power module based on the remaining lifetime of the current cycle and the actual remaining lifetime; if the lifetime degradation degree reaches a preset threshold, then the actual remaining lifetime and the new junction temperature estimation neural network model are sent to the vehicle end, so that the vehicle end determines the actual remaining lifetime as the remaining lifetime of the next cycle and estimates the junction temperature of the target power module in the next cycle using the new junction temperature estimation neural network model.

2. The power module junction temperature estimation method according to claim 1, characterized in that, The actual remaining lifetime of the target power module is determined based on the historical junction temperature output by the electrothermal network model and the current junction temperature, including: Obtain multiple historical junction temperatures output by the electrothermal network model; The junction temperature curves are obtained by fitting the multiple historical junction temperatures and the current junction temperature, and the junction temperature curves are statistically analyzed using the rainflow counting method to obtain at least one set of thermal stresses and the number of power cycles under each set of thermal stresses. The thermal stresses include junction temperature fluctuations and average junction temperature. The junction temperature fluctuation and average junction temperature in each set of thermal stress are input into the lifetime model so that the lifetime model outputs the failure power cycle number under each set of thermal stress. The lifetime model is established based on the junction temperature fluctuation and average junction temperature. The ratio of the number of power cycles to the number of failure power cycles under each set of thermal stresses is taken as the fatigue damage value corresponding to each set of thermal stresses, and the actual remaining life is calculated based on the fatigue damage value corresponding to each set of thermal stresses.

3. The power module junction temperature estimation method according to claim 2, characterized in that, If the degree of lifetime degradation reaches a preset threshold, the power module junction temperature estimation method further includes: A new thermal impedance value is calculated based on the current power loss, the temperature data in the current operating condition data, and the average junction temperature in each group of thermal stresses, so as to update the thermal impedance value in the electrothermal network model. The current power loss is obtained based on the switching frequency, voltage data, and current data in the current operating condition data. or, A new thermal impedance value is calculated based on the input power and temperature data in the current operating condition data, as well as the average junction temperature in each group of thermal stresses, to update the thermal impedance value in the electrothermal network model. The operating condition data also includes the input power.

4. The method for estimating the junction temperature of a power module according to any one of claims 1-3, characterized in that, The electrothermal network model is established in the following ways: A power loss model for calculating power loss is established based on switching frequency, voltage data, and current data. A thermal network model for calculating junction temperature is established based on power loss, thermal impedance value, and temperature data. The power loss model and the thermal network model are coupled to obtain the electrothermal network model for junction temperature calculation.

5. The method for estimating the junction temperature of a power module according to any one of claims 1-3, characterized in that, Based on the current operating condition data, the remaining lifetime of the current cycle, and the current junction temperature, the current junction temperature estimation neural network model is iteratively trained to obtain a new, trained junction temperature estimation neural network model, including: Using the remaining lifetime of the current cycle, the pump speed, switching frequency, bus voltage, collector current, and cooling water temperature from the current operating data as input values, and the current junction temperature as the output value, the current junction temperature estimation neural network model is iteratively trained to obtain the new junction temperature estimation neural network model. The operating data also includes the pump speed, the voltage data includes the bus voltage, and the current data includes the collector current.

6. The method for estimating the junction temperature of a power module according to any one of claims 1-3, characterized in that, Before establishing the electrothermal network model, the method for estimating the junction temperature of the power module includes: Obtain the voltage drop of the target power module; the voltage drop of the power module characterizes the aging degree of the power module. An aging test was conducted on the sample power module until the voltage drop of the sample power module reached the voltage drop of the target power module, wherein the model of the target power module is the same as that of the sample power module. The thermal impedance value of the sample power module is calculated based on the junction temperature, casing temperature and input power of the sample power module, and is used as the thermal impedance value of the target power module to establish the electrothermal network model.

7. A power module junction temperature estimation device, characterized in that, The power module junction temperature estimation device includes: The acquisition module is used to acquire the current operating condition data and the remaining lifespan of the target power module in the current cycle. The operating condition data includes switching frequency, voltage data, current data and temperature data. The calculation module is used to input the switching frequency, voltage data, current data and temperature data in the current operating condition data into a preset electrothermal network model, so that the electrothermal network model outputs the current junction temperature. The electrothermal network model is established based on the thermal impedance value of the target power module, and the thermal impedance value is variable. The training module is used to iteratively train the current junction temperature estimation neural network model based on the current operating condition data, the remaining lifetime of the current cycle, and the current junction temperature to obtain a new junction temperature estimation neural network model after training. An update module is used to update the junction temperature estimation neural network model of the vehicle end corresponding to the target power module in order to estimate the junction temperature of the power module. Updating the junction temperature estimation neural network model of the vehicle end corresponding to the target power module includes: determining the actual remaining lifetime of the target power module based on the historical junction temperature output by the electrothermal network model and the current junction temperature; calculating the lifetime degradation degree of the target power module based on the current cycle remaining lifetime and the actual remaining lifetime; if the lifetime degradation degree reaches a preset threshold, then sending the actual remaining lifetime and the new junction temperature estimation neural network model to the vehicle end, so that the vehicle end determines the actual remaining lifetime as the remaining lifetime of the next cycle, and estimates the junction temperature of the target power module in the next cycle using the new junction temperature estimation neural network model.

8. An electronic device, characterized in that, The electronic device includes: One or more processors; A storage device for storing one or more programs that, when executed by one or more processors, cause the electronic device to implement the power module junction temperature estimation method as described in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, It stores a computer program that, when executed by the computer's processor, causes the computer to perform the power module junction temperature estimation method as described in any one of claims 1-6.

Images