Method and system for evaluating service life of power semiconductor device

Abstract

The invention discloses a service life evaluation method and system for a power semiconductor device, and belongs to the technical field of semiconductors, and the method specifically comprises the steps: obtaining a to-be-evaluated device, carrying out the power cycle test under different single junction temperature fluctuations, and recording the failure data, including the cycle index, the junction temperature fluctuation range and the electrical performance parameters; thirdly, performing statistical analysis on failure data, calculating a junction temperature fluctuation amplitude correlation fatigue factor and a power cycle frequency accumulation fatigue factor, and combining the junction temperature fluctuation mode factor to obtain a composite fatigue factor; calculating an accumulated fatigue factor according to the composite fatigue factor and the dynamic weight by adopting a weighted accumulation method; inputting the accumulated fatigue factor into a pre-constructed service life evaluation model, and calculating the service life of the device; when the accumulated fatigue factor exceeds a threshold value, triggering an early warning mechanism, outputting early warning information, and prompting that the service life of the device is about to expire or expired; accurate evaluation and early warning of the service life of the power semiconductor device are realized.

Description

Technical Field

[0001] The present invention belongs to the field of semiconductor technology, and in particular to a service life evaluation method and system for power semiconductor devices. Background Art

[0002] With the rapid development of power electronics technology, power semiconductor devices have been widely used in various power systems, industrial control, new energy vehicles and other fields. The reliability of these devices during long-term operation is crucial. Once a failure occurs, it may cause the paralysis of the entire system and bring huge economic losses. However, there are still deficiencies in the current accurate assessment methods for the service life of power semiconductor devices. Traditional methods often only consider a single factor and cannot fully reflect the actual life of the device under complex working conditions.

[0003] For example, the patent with the authorization announcement number CN118275848B discloses a method, device, and semiconductor device for evaluating the service life of power semiconductor devices, including: obtaining the number of cycle tests of the power semiconductor device to be evaluated under different junction temperature fluctuation combination conditions until failure; calculating the cumulative fatigue factor based on the number of cycle tests and the composite fatigue factor; and evaluating the service life of the power semiconductor device to be evaluated based on the cumulative fatigue factor. This technical solution simulates different service conditions by taking different junction temperature fluctuations to perform separate tests and coupled tests on power semiconductor devices with different junction temperature fluctuations, obtains the fatigue factor of the power semiconductor device under the coupling of different service conditions, realizes the multi-condition service life status evaluation of the power semiconductor device, and improves the accuracy of the service life evaluation of the power semiconductor device.

[0004] The above existing technologies have the following problems: although it mentions testing through different junction temperature fluctuation combination conditions, it does not explain in detail how to combine and the scope of the combination, which may result in the test conditions failing to fully cover the various complex situations that may be encountered in actual service; although the concepts of composite fatigue factor and cumulative fatigue factor are mentioned, it does not explain in detail how to calculate these factors based on the junction temperature fluctuation combination conditions, which may simplify the complexity of the fatigue accumulation process and thus affect the accuracy of the evaluation; there is a lack of specific steps and details. Summary of the invention

[0005] In view of the deficiencies in the prior art, the present invention proposes a service life assessment method and system for power semiconductor devices. The method comprises the following steps: obtaining a device to be assessed, performing power cycle tests under different single junction temperature fluctuations, and recording failure data, including the number of cycles, junction temperature fluctuation range, and electrical performance parameters; then, statistically analyzing the failure data, calculating a fatigue factor related to the junction temperature fluctuation amplitude and a cumulative fatigue factor of the power cycle frequency, and combining the junction temperature fluctuation mode factor to obtain a composite fatigue factor; using a weighted accumulation method, calculating the cumulative fatigue factor based on the composite fatigue factor and a dynamic weight; inputting the cumulative fatigue factor into a pre-constructed service life assessment model to calculate the service life of the device; when the cumulative fatigue factor exceeds a threshold, triggering an early warning mechanism, outputting early warning information, and prompting that the device life is about to expire or has expired; thus, accurate assessment and early warning of the service life of power semiconductor devices are achieved.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A method for evaluating the service life of a power semiconductor device, comprising:

[0008] Step S1: obtaining a power semiconductor device to be evaluated, performing power cycle tests on the power semiconductor device under different single junction temperature fluctuations, and recording failure data, wherein the failure data includes the number of cycles when the device fails under each junction temperature fluctuation test, the junction temperature fluctuation range, and electrical performance parameters;

[0009] Step S2: By statistically analyzing the failure data under different junction temperature fluctuation amplitudes, the fatigue factor related to the junction temperature fluctuation amplitude is calculated. At the same time, the cumulative fatigue factor of the power cycle frequency is calculated by fitting the experimental data and the power cycle frequency. The composite fatigue factor corresponding to each power cycle test is calculated based on the fatigue factor related to the junction temperature fluctuation amplitude and the cumulative fatigue factor of the power cycle frequency, combined with the junction temperature fluctuation mode factor.

[0010] Step S3: Calculate the cumulative fatigue factor by using a weighted accumulation method according to the composite fatigue factor and the dynamic weight allocated based on the magnitude of the junction temperature fluctuation;

[0011] Step S4: Input the cumulative fatigue factor value into the pre-trained service life assessment model to calculate the service life of the device. When the cumulative fatigue factor exceeds the preset threshold, the early warning mechanism is triggered and an early warning message is output, indicating that the device is about to reach or has exceeded its expected service life.

[0012] Specifically, the specific steps of step S2 include:

[0013] S2.1: Obtain the failure data in step S1. For a single junction temperature fluctuation scenario, according to the failure data, use the formula Calculation of fatigue factor related to junction temperature fluctuation amplitude , where N represents the number of cycles, Indicates the junction temperature fluctuation range, It represents the effect function of junction temperature fluctuation on plastic strain range, and k represents the empirical coefficient;

[0014] S2.2: Conduct power cycle experiments, record failure data at different power cycle frequencies, and fit the failure data. Based on the fitting results, calculate the cumulative fatigue factor of the power cycle frequency. , where p represents the power cycle frequency, c represents the constant coefficient, represents the reference frequency, b represents the fitting parameter, represents the adjustment coefficient, represents the logarithmic function, Indicates the reference frequency point.

[0015] Specifically, the specific steps of step S2 also include:

[0016] S2.3: Get the junction temperature fluctuation range , and according to Calculate the time required for the junction temperature to rise from the minimum value to the maximum value or to drop from the maximum value to the minimum value. At the same time, according to the number of junction temperature fluctuations per unit time, the fluctuation frequency is obtained ,in, Indicates the maximum junction temperature, represents the minimum junction temperature, and ;

[0017] S2.4: Based on , , , combined with the waveform shape, calculate the junction temperature fluctuation mode factor ,in, Indicates the reference amplitude, represents the reference frequency, represents the waveform shape factor, , , , represents the weight coefficient;

[0018] S2.5: Fatigue factor based on the correlation of junction temperature fluctuation amplitude and power cycle frequency cumulative fatigue factor , combined with the junction temperature fluctuation mode factor , calculate the composite fatigue factor corresponding to each power cycle test ,in, , and Represents the weight coefficient.

[0019] Specifically, the specific steps of step S3 include:

[0020] S3.1: Composite fatigue factor for the i-th power cycle test calculated according to S2.3 ;

[0021] S3.2: For each power cycle test, record and measure the corresponding junction temperature fluctuation amplitude ;

[0022] S3.3: Based on the magnitude of the junction temperature fluctuation, use the formula Assign a dynamic weight to each test ;

[0023] S3.4: Calculate the cumulative fatigue factor using the weighted accumulation method , where n represents the number of power cycle tests.

[0024] Specifically, the specific steps of step S4 include:

[0025] S4.1: Load the pre-trained service life assessment model and obtain the cumulative fatigue factor in S3.4 ;

[0026] S4.2: Input the cumulative fatigue factor value into the pre-trained service life assessment model. The service life assessment model calculates and outputs the service life prediction value Y of the device according to the input cumulative fatigue factor value;

[0027] S4.3: Set cumulative fatigue factor threshold ;

[0028] like , then continue monitoring;

[0029] like , the early warning mechanism is immediately triggered, and the system generates early warning information according to the preset template. The early warning information includes the device number, the current cumulative fatigue factor value, the estimated remaining reliable operation time, and the recommended maintenance measures;

[0030] S4.4: The warning information is transmitted to the monitoring terminal by wireless communication, and the monitoring terminal is the display screen in the central control room.

[0031] Specifically, the training process of the service life assessment model in S4.1 includes:

[0032] S4.11: Collect experimental data of power semiconductor devices and collect operating data of devices under actual working conditions, wherein the experimental data includes aging of devices under different operating temperatures, current densities, voltage levels, switching frequencies and mechanical stress environments;

[0033] S4.12: preprocessing the collected experimental data and operation data, and extracting features from the preprocessed experimental data and operation data to obtain a feature data set;

[0034] S4.13: Load a pre-built support vector regression model, and input the feature data set into the pre-built support vector regression model for training to generate a trained service life assessment model.

[0035] A service life assessment system for power semiconductor devices, comprising: a data acquisition module, a fatigue factor calculation module, and a service life assessment module;

[0036] The data acquisition module is used to acquire power cycle test data of the power semiconductor device to be evaluated under different single junction temperature fluctuations;

[0037] The fatigue factor calculation module is used to calculate the junction temperature fluctuation amplitude correlation fatigue factor and the power cycle frequency cumulative fatigue factor according to the statistical analysis of the test data, and calculate the composite fatigue factor in combination with the junction temperature fluctuation mode factor. At the same time, the cumulative fatigue factor is calculated by using the weighted accumulation method;

[0038] The service life assessment module is used to input the cumulative fatigue factor into a pre-built service life assessment model, calculate the service life of the device, and trigger an early warning mechanism when necessary.

[0039] Specifically, the fatigue factor calculation module includes: a fatigue factor calculation unit and a weight allocation unit;

[0040] The fatigue factor calculation unit is used to calculate the junction temperature fluctuation amplitude correlation fatigue factor, the power cycle frequency cumulative fatigue factor, the junction temperature fluctuation mode factor, the composite fatigue factor and the cumulative fatigue factor calculation unit;

[0041] The weight allocation unit is used to allocate a weight to each test according to the magnitude of the junction temperature fluctuation.

[0042] Specifically, the service life assessment module includes: a service life calculation unit, an early warning mechanism triggering unit, and an early warning information output unit;

[0043] The service life calculation unit is used to calculate the service life of the device according to the cumulative fatigue factor and the service life assessment model;

[0044] The early warning mechanism triggering unit is used to trigger the early warning mechanism when the accumulated fatigue factor exceeds a preset threshold;

[0045] The warning information output unit is used to output warning information to prompt that the device is about to reach or has exceeded its expected service life.

[0046] An electronic device comprises a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of a service life assessment method for a power semiconductor device when executing the computer program.

[0047] Compared with the prior art, the present invention has the following beneficial effects:

[0048] 1. The present invention proposes a service life assessment method for power semiconductor devices. By comprehensively considering the failure data under different single junction temperature fluctuations, the fatigue factor related to the junction temperature fluctuation amplitude, the power cycle frequency cumulative fatigue factor and the composite fatigue factor are accurately calculated, thereby realizing a comprehensive simulation and accurate assessment of the fatigue accumulation process of the device under multiple working conditions, which not only improves the accuracy of the assessment, but also helps to deeply understand the failure mechanism of the device under different service conditions.

[0049] 2. The present invention proposes a service life assessment method for power semiconductor devices. By inputting the cumulative fatigue factor into a pre-built service life assessment model, the service life of the device can be directly obtained, and the early warning mechanism is triggered when the cumulative fatigue factor exceeds a preset threshold, and early warning information is output in time. This function not only facilitates users to monitor and manage the service status of the device in real time, but also effectively avoids potential safety risks caused by device failure, thereby improving the reliability and stability of the overall system. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 A schematic diagram of a service life assessment method for a power semiconductor device according to the present invention;

[0051] Figure 2 This is an algorithm flow chart of a service life assessment method for a power semiconductor device of the present invention;

[0052] Figure 3 This is a framework diagram of a service life assessment system for a power semiconductor device according to the present invention. DETAILED DESCRIPTION

[0053] Example 1

[0054] See also Figure 1 and Figure 2 , an embodiment of the present invention provides: a method for evaluating the service life of a power semiconductor device, comprising the following steps:

[0055] Step S1: obtaining a power semiconductor device to be evaluated, performing power cycle tests on the power semiconductor device under different single junction temperature fluctuations, and recording failure data, wherein the failure data includes the number of cycles when the device fails under each junction temperature fluctuation test, the junction temperature fluctuation range, and electrical performance parameters;

[0056] Furthermore, the specific steps of step S1 include:

[0057] (1) Obtain the power semiconductor devices to be evaluated and ensure that their quantity is sufficient for multiple tests and statistical analysis;

[0058] (2) Prepare the equipment required for testing, including current source, controller, voltage / current probe, data acquisition system, and instruments for measuring junction temperature, such as thermistor, infrared thermometer, or electrical parameter method based on on-state characteristics;

[0059] (3) Connect the device under test to the test system, ensure that the connection is stable and reliable, and set the test parameters, including the load current, cycle period, and junction temperature fluctuation range;

[0060] (4) Start the test program. The device will perform current on-off operations according to the set parameters to simulate the junction temperature fluctuations of power devices in actual applications. During the test, the device's temperature changes and electrical performance parameters, such as saturation voltage drop and thermal resistance, are monitored in real time.

[0061] (5) Record the number of cycles when the device fails under each junction temperature fluctuation test. The failure standard can be determined based on the device's thermal resistance growth, changes in electrical performance parameters, or abnormal temperature fluctuations;

[0062] (6) After the test is completed, organize and save all failure data, including the number of cycles, junction temperature fluctuation range, and electrical performance parameters of the device when it fails.

[0063] The core principle of the power cycle test is to simulate the junction temperature fluctuation of power devices in actual applications by turning on and off the external load current. When the device is powered on, the temperature of the device rises due to the heat generated by the internal chip; after the power is turned off, the temperature of the device drops. This process is repeated to simulate the thermal stress changes under actual use conditions.

[0064] Step S2: By statistically analyzing the failure data under different junction temperature fluctuation amplitudes, the fatigue factor related to the junction temperature fluctuation amplitude is calculated. At the same time, the cumulative fatigue factor of the power cycle frequency is calculated by fitting the experimental data and the power cycle frequency. The composite fatigue factor corresponding to each power cycle test is calculated based on the fatigue factor related to the junction temperature fluctuation amplitude and the cumulative fatigue factor of the power cycle frequency, combined with the junction temperature fluctuation mode factor.

[0065] Step S3: Calculate the cumulative fatigue factor by using a weighted accumulation method according to the composite fatigue factor and the dynamic weight allocated based on the magnitude of the junction temperature fluctuation;

[0066] Among them, the fatigue factor correlated with the junction temperature fluctuation amplitude is calculated by statistical analysis of failure data under different junction temperature fluctuation amplitudes; the cumulative fatigue factor of power cycle frequency is obtained by fitting experimental data and calculating the power cycle frequency; the junction temperature fluctuation mode factor refers to the junction temperature fluctuation mode experienced by the power semiconductor device to be evaluated, which is obtained by quantifying and analyzing the junction temperature fluctuation characteristics; the composite fatigue factor is calculated by combining the fatigue factor correlated with the junction temperature fluctuation amplitude and the cumulative fatigue factor of power cycle frequency with the junction temperature fluctuation mode factor; the cumulative fatigue factor is obtained by using the weighted addition method, multiplying the composite fatigue factor of each test by its corresponding weight, and then adding all the products.

[0067] Step S4: Input the cumulative fatigue factor value into the pre-trained service life assessment model to calculate the service life of the device. When the cumulative fatigue factor exceeds the preset threshold, the early warning mechanism is triggered and an early warning message is output, indicating that the device is about to reach or has exceeded its expected service life.

[0068] The specific steps of step S2 include:

[0069] S2.1: Obtain the failure data in step S1. For a single junction temperature fluctuation scenario, according to the failure data, use the formula Calculation of fatigue factor related to junction temperature fluctuation amplitude , where N represents the number of cycles, Indicates the junction temperature fluctuation range, It represents the effect function of junction temperature fluctuation on plastic strain range, and k represents the empirical coefficient;

[0070] S2.2: Conduct power cycle experiments, record failure data at different power cycle frequencies, and fit the failure data. Based on the fitting results, calculate the cumulative fatigue factor of the power cycle frequency. , where p represents the power cycle frequency, c represents the constant coefficient, represents the reference frequency, b represents the fitting parameter, represents the adjustment coefficient, represents the logarithmic function, Indicates the reference frequency point;

[0071] Furthermore, the specific steps of S2.2 include:

[0072] (1) Select power cycle test system, temperature monitoring equipment, and determine power semiconductor device samples;

[0073] (2) Install the power semiconductor device on the test system and connect the relevant test circuits. Perform the experiment according to the set power cycle frequency and temperature change range to simulate the power cycle and temperature changes in the actual working environment.

[0074] (3) During the experiment, pay attention to the temperature response and performance of power semiconductor devices and record failure data. Failure data includes the number of cycles at each power cycle frequency when the device fails, the temperature variation range, and electrical performance parameters;

[0075] (4) Organize and analyze the recorded failure data to find out the relationship between power cycle frequency and failure cycle number. The failure data is fitted to obtain a fitting equation between the power cycle frequency and the number of failure cycles, where d and r represent the coefficients of the fitting equation;

[0076] (5) According to the fitting equation, calculate the power cycle frequency cumulative fatigue factor.

[0077] S2.3: Get the junction temperature fluctuation range , and according to Calculate the time required for the junction temperature to rise from the minimum value to the maximum value or to drop from the maximum value to the minimum value. At the same time, according to the number of junction temperature fluctuations per unit time, the fluctuation frequency is obtained ,in, Indicates the maximum junction temperature, represents the minimum junction temperature, and ;

[0078] S2.4: Based on , , , combined with the waveform shape, calculate the junction temperature fluctuation mode factor ,in, Indicates the reference amplitude, represents the reference frequency, represents the waveform shape factor, , , , represents the weight coefficient;

[0079] S2.5: Fatigue factor based on the correlation of junction temperature fluctuation amplitude and power cycle frequency cumulative fatigue factor , combined with the junction temperature fluctuation mode factor , calculate the composite fatigue factor corresponding to each power cycle test ,in, , and Represents the weight coefficient.

[0080] The specific steps of step S3 include:

[0081] S3.1: Composite fatigue factor for the i-th power cycle test calculated according to S2.3 ;

[0082] S3.2: For each power cycle test, record and measure the corresponding junction temperature fluctuation amplitude ;

[0083] S3.3: Based on the magnitude of the junction temperature fluctuation, use the formula Assign a dynamic weight to each test ;

[0084] S3.4: Calculate the cumulative fatigue factor using the weighted accumulation method , where n represents the number of power cycle tests.

[0085] The specific steps of step S4 include:

[0086] S4.1: Load the pre-trained service life assessment model and obtain the cumulative fatigue factor in S3.4 ;

[0087] S4.2: Input the cumulative fatigue factor value into the pre-trained service life assessment model. The service life assessment model calculates and outputs the service life prediction value Y of the device according to the input cumulative fatigue factor value;

[0088] Furthermore, the specific steps of S4.2 include:

[0089] (1) Ensure that the service life assessment model has been fully trained and has good prediction performance. The model should be able to receive the cumulative fatigue factor as input and output the service life prediction value of the device;

[0090] (2) Preprocess the cumulative fatigue factor values ​​obtained from the experiment, such as data cleaning and normalization, to ensure the quality of the input data;

[0091] (3) passing the preprocessed cumulative fatigue factor value as input to the service life assessment model;

[0092] (4) The service life assessment model uses the learned mapping relationship or algorithm to perform calculations based on the input cumulative fatigue factor value. This process may involve complex mathematical operations and logical judgments, and ultimately derives a predicted service life value for the device. The mapping relationship or algorithm within the model is the prior art content in this field and is not an inventive solution of the present application, and is not elaborated here.

[0093] S4.3: Set cumulative fatigue factor threshold ;

[0094] like , then continue monitoring;

[0095] like , the early warning mechanism is immediately triggered, and the system generates early warning information according to the preset template. The early warning information includes the device number, the current cumulative fatigue factor value, the estimated remaining reliable operation time, and the recommended maintenance measures;

[0096] S4.4: The warning information is transmitted to the monitoring terminal by wireless communication, and the monitoring terminal is the display screen in the central control room.

[0097] The training process of the service life assessment model in S4.1 includes:

[0098] S4.11: Collect experimental data of power semiconductor devices and collect operating data of devices under actual working conditions, wherein the experimental data includes aging of devices under different operating temperatures, current densities, voltage levels, switching frequencies and mechanical stress environments;

[0099] S4.12: preprocessing the collected experimental data and operation data, and extracting features from the preprocessed experimental data and operation data to obtain a feature data set;

[0100] In the present invention, a statistical feature extraction method is used for feature extraction, wherein the statistical feature extraction method is the prior art content in the field, is not the inventive solution of the present application, and is not described in detail here.

[0101] S4.13: Load a pre-built support vector regression model, and input the feature data set into the pre-built support vector regression model for training to generate a trained service life assessment model, wherein the support vector regression model is the prior art content in this field and is not the inventive solution of this application, and will not be elaborated here.

[0102] Example 2

[0103] See also Figure 3 Another embodiment provided by the present invention is a service life assessment system for a power semiconductor device, comprising:

[0104] Data acquisition module, fatigue factor calculation module, service life assessment module;

[0105] A data acquisition module, used to acquire power cycle test data of the power semiconductor device to be evaluated under different single junction temperature fluctuations;

[0106] The fatigue factor calculation module is used to calculate the junction temperature fluctuation amplitude correlation fatigue factor and the power cycle frequency cumulative fatigue factor based on the statistical analysis of the test data, and calculate the composite fatigue factor in combination with the junction temperature fluctuation mode factor. At the same time, the weighted accumulation method is used to calculate the cumulative fatigue factor;

[0107] The service life assessment module is used to input the cumulative fatigue factor into the pre-built service life assessment model, calculate the service life of the device, and trigger the early warning mechanism when necessary.

[0108] The data acquisition module includes: a test execution unit and a data recording unit;

[0109] A test execution unit, for performing a power cycle test, including setting test conditions, such as a junction temperature fluctuation range, a power level, etc., and running the test;

[0110] The data recording unit is used to record the failure data during the test process, including the number of cycles when the device fails under each junction temperature fluctuation test, the junction temperature fluctuation range and the electrical performance parameters.

[0111] The fatigue factor calculation module includes: a fatigue factor calculation unit and a weight distribution unit;

[0112] Fatigue factor calculation unit, used to calculate the junction temperature fluctuation amplitude correlation fatigue factor, power cycle frequency cumulative fatigue factor, junction temperature fluctuation mode factor, composite fatigue factor and cumulative fatigue factor;

[0113] The weight allocation unit is used to allocate a weight to each test according to the magnitude of the junction temperature fluctuation.

[0114] The service life assessment module includes: service life calculation unit, early warning mechanism triggering unit, and early warning information output unit;

[0115] A service life calculation unit, used to calculate the service life of the device based on the cumulative fatigue factor and the service life assessment model;

[0116] An early warning mechanism triggering unit is used to trigger the early warning mechanism when the accumulated fatigue factor exceeds a preset threshold;

[0117] The warning information output unit is used to output warning information to prompt that the device is about to reach or has exceeded its expected service life.

[0118] Example 3

[0119] An electronic device includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, steps of a method for evaluating the service life of a power semiconductor device are implemented. For details, please refer to the above method embodiment, which will not be repeated here.

[0120] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0121] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0122] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation modes, which are merely illustrative rather than restrictive. Under the guidance of the present invention, ordinary technicians in the field may also change, modify, replace and modify the above-mentioned embodiments without departing from the purpose and scope of protection of the present invention, and all of these are within the protection of the present invention.

Claims

1. A method for evaluating the service life of a power semiconductor device, characterized in that: include: Step S1: obtaining a power semiconductor device to be evaluated, performing power cycle tests on the power semiconductor device under different single junction temperature fluctuations, and recording failure data, wherein the failure data includes the number of cycles when the device fails under each junction temperature fluctuation test, the junction temperature fluctuation range, and electrical performance parameters; Step S2: By statistically analyzing the failure data under different junction temperature fluctuation amplitudes, the fatigue factor related to the junction temperature fluctuation amplitude is calculated. At the same time, the cumulative fatigue factor of the power cycle frequency is calculated by fitting the experimental data and the power cycle frequency. The composite fatigue factor corresponding to each power cycle test is calculated based on the fatigue factor related to the junction temperature fluctuation amplitude and the cumulative fatigue factor of the power cycle frequency, combined with the junction temperature fluctuation mode factor. Step S3: Calculate the cumulative fatigue factor by using a weighted accumulation method according to the composite fatigue factor and the dynamic weight allocated based on the magnitude of the junction temperature fluctuation; Step S4: Input the cumulative fatigue factor value into the pre-trained service life assessment model to calculate the service life of the device. When the cumulative fatigue factor exceeds the preset threshold, the early warning mechanism is triggered and an early warning message is output, indicating that the device is about to reach or has exceeded its expected service life.

2. A method for evaluating the service life of a power semiconductor device as claimed in claim 1, characterized in that: The specific steps of step S2 include: S2.1: Obtain the failure data in step S1. For a single junction temperature fluctuation scenario, according to the failure data, use the formula Calculation of fatigue factor related to junction temperature fluctuation amplitude , where N represents the number of cycles, Indicates the junction temperature fluctuation range, It represents the effect function of junction temperature fluctuation on plastic strain range, and k represents the empirical coefficient; S2.2: Conduct power cycle experiments, record failure data at different power cycle frequencies, and fit the failure data. Based on the fitting results, calculate the cumulative fatigue factor of the power cycle frequency. , where p represents the power cycle frequency, c represents the constant coefficient, represents the reference frequency, b represents the fitting parameter, represents the adjustment coefficient, represents the logarithmic function, Indicates the reference frequency point.

3. A method for evaluating the service life of a power semiconductor device as claimed in claim 2, characterized in that: The specific steps of step S2 also include: S2.3: Get the junction temperature fluctuation range , and according to Calculate the time required for the junction temperature to rise from the minimum value to the maximum value or to drop from the maximum value to the minimum value. At the same time, according to the number of junction temperature fluctuations per unit time, the fluctuation frequency is obtained ,in, Indicates the maximum junction temperature, represents the minimum junction temperature, and ; S2.4: Based on , , , combined with the waveform shape, calculate the junction temperature fluctuation mode factor ,in, Indicates the reference amplitude, represents the reference frequency, represents the waveform shape factor, , , , represents the weight coefficient; S2.5: Fatigue factor based on the correlation of junction temperature fluctuation amplitude and power cycle frequency cumulative fatigue factor , combined with the junction temperature fluctuation mode factor , calculate the composite fatigue factor corresponding to each power cycle test ,in, , and Represents the weight coefficient.

4. A method for evaluating the service life of a power semiconductor device as claimed in claim 3, characterized in that: The specific steps of step S3 include: S3.1: Composite fatigue factor for the i-th power cycle test calculated according to S2.3 ; S3.2: For each power cycle test, record and measure the corresponding junction temperature fluctuation amplitude ; S3.3: Based on the magnitude of the junction temperature fluctuation, use the formula Assign a dynamic weight to each test ; S3.4: Calculate the cumulative fatigue factor using the weighted accumulation method , where n represents the number of power cycle tests.

5. A method for evaluating the service life of a power semiconductor device as claimed in claim 4, characterized in that: The specific steps of step S4 include: S4.1: Load the pre-trained service life assessment model and obtain the cumulative fatigue factor in S3.4 ; S4.2: Input the cumulative fatigue factor value into the pre-trained service life assessment model. The service life assessment model calculates and outputs the service life prediction value Y of the device according to the input cumulative fatigue factor value; S4.3: Set cumulative fatigue factor threshold ; like , then continue monitoring; like , the early warning mechanism is immediately triggered, and the system generates early warning information according to the preset template. The early warning information includes the device number, the current cumulative fatigue factor value, the estimated remaining reliable operation time, and the recommended maintenance measures; S4.4: The warning information is transmitted to the monitoring terminal by wireless communication, and the monitoring terminal is the display screen in the central control room.

6. A method for evaluating the service life of a power semiconductor device as claimed in claim 5, characterized in that: The training process of the service life assessment model in S4.1 includes: S4.11: Collect experimental data of power semiconductor devices and collect operating data of devices under actual working conditions, wherein the experimental data includes aging of devices under different operating temperatures, current densities, voltage levels, switching frequencies and mechanical stress environments; S4.12: preprocessing the collected experimental data and operation data, and extracting features from the preprocessed experimental data and operation data to obtain a feature data set; S4.13: Load a pre-built support vector regression model, and input the feature data set into the pre-built support vector regression model for training to generate a trained service life assessment model.

7. A service life assessment system for a power semiconductor device, which is used to implement a service life assessment method for a power semiconductor device according to any one of claims 1 to 6, characterized in that: include: Data acquisition module, fatigue factor calculation module, service life assessment module; The data acquisition module is used to acquire power cycle test data of the power semiconductor device to be evaluated under different single junction temperature fluctuations; The fatigue factor calculation module is used to calculate the junction temperature fluctuation amplitude correlation fatigue factor and the power cycle frequency cumulative fatigue factor according to the statistical analysis of the test data, and calculate the composite fatigue factor in combination with the junction temperature fluctuation mode factor. At the same time, the cumulative fatigue factor is calculated by using the weighted accumulation method; The service life assessment module is used to input the cumulative fatigue factor into a pre-built service life assessment model, calculate the service life of the device, and trigger an early warning mechanism when necessary.

8. A power semiconductor device service life assessment system as claimed in claim 7, characterized in that: The fatigue factor calculation module includes: a fatigue factor calculation unit and a weight allocation unit; The fatigue factor calculation unit is used to calculate the junction temperature fluctuation amplitude correlation fatigue factor, the power cycle frequency cumulative fatigue factor, the junction temperature fluctuation mode factor, the composite fatigue factor and the cumulative fatigue factor calculation unit; The weight allocation unit is used to allocate a weight to each test according to the magnitude of the junction temperature fluctuation.

9. A service life assessment system for power semiconductor devices as claimed in claim 8, characterized in that: The service life assessment module includes: a service life calculation unit, an early warning mechanism triggering unit, and an early warning information output unit; The service life calculation unit is used to calculate the service life of the device according to the cumulative fatigue factor and the service life assessment model; The early warning mechanism triggering unit is used to trigger the early warning mechanism when the accumulated fatigue factor exceeds a preset threshold; The warning information output unit is used to output warning information to prompt that the device is about to reach or has exceeded its expected service life.

10. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method for evaluating the service life of a power semiconductor device according to any one of claims 1 to 6 are implemented.

Images