SiC MOSFET gate oxide layer degradation comprehensive diagnosis method and system

Abstract

A SiC MOSFET gate oxide layer degradation comprehensive diagnosis method and system, the method comprises the following steps: constructing a data set, and training to obtain a junction temperature estimation function; obtaining actual precursor parameters under actual operating conditions through a measurement circuit, substituting the actual precursor parameters into the junction temperature estimation function to obtain a current actual junction temperature estimation value, substituting the current actual junction temperature estimation value into a linear relationship between the precursor parameters and the junction temperature before the gate oxide layer degrades to obtain the flat band voltage and the threshold voltage of the gate oxide layer without degradation at the junction temperature; calculating the difference between the actual precursor parameters and the precursor parameters at the junction temperature without degradation of the gate oxide layer to obtain a precursor parameter drift amount, and performing comprehensive diagnosis of the gate oxide layer degradation according to the precursor parameter drift amount to obtain a comprehensive diagnosis result. The present application solves the misdiagnosis problem caused by temperature drift in the prior art, improves the accuracy and robustness of the degradation diagnosis, and realizes real-time monitoring and thermal state management of the device working junction temperature state.

Description

Technical Field

[0001] This invention relates to the field of gate oxide diagnostic technology, and in particular to a comprehensive diagnostic method and system for SiC MOSFET gate oxide degradation. Background Technology

[0002] SiC MOSFETs (silicon carbide metal-oxide-semiconductor field-effect transistors) have gradually replaced traditional silicon-based devices and become key core components in power electronic systems due to their superior performance in high-voltage, high-temperature, and high-frequency applications. Despite their significant performance advantages, SiC MOSFETs still face reliability challenges under long-term operation, with gate oxide degradation considered one of the key failure mechanisms limiting their lifetime and stability.

[0003] To assess the degradation state of the gate oxide layer, existing degradation diagnostic studies typically focus solely on evaluating the degree or location of degradation. The degree of degradation is usually assessed using precursor parameters, while the location is primarily characterized by the CV characteristics of the gate deflection capacitance. However, these techniques are largely sensitive to junction temperature. If the influence of temperature is not effectively mitigated, temperature drift errors may be misinterpreted as degradation features, leading to distorted diagnostic results and affecting the accurate identification of gate oxide degradation. Temperature drift errors may result in misjudging severe device degradation, leading to incorrect maintenance decisions or lifetime predictions. This not only wastes resources but may also cause premature device failure, resulting in significant economic losses. More seriously, temperature drift errors may mask true degradation characteristics, preventing timely detection of actual degradation and threatening long-term stability and safety.

[0004] Existing solutions typically involve frequent re-baseline adjustments to the relationship between precursor parameters and degradation to compensate for junction temperature effects, but this approach is usually costly. Therefore, while existing technologies can accurately diagnose the degree and location of gate oxide degradation, effectively mitigating the effects of temperature drift, reducing baseline maintenance costs, and simultaneously ensuring real-time monitoring and thermal management of device junction temperature remain critical technical challenges that need to be addressed. Summary of the Invention

[0005] This invention provides a comprehensive diagnostic method and system for SiC MOSFET gate oxide degradation to address the technical problems mentioned in the background section. The "junction temperature monitoring" described in this specification refers to the real-time monitoring, anomaly detection, or thermal status management of the actual junction temperature state of the device using the estimated junction temperature value output by the junction temperature estimation function.

[0006] To achieve the above objectives, the technical solution of the present invention is implemented as follows:

[0007] This invention provides a comprehensive diagnostic method for gate oxide degradation in SiC MOSFETs, comprising the following steps:

[0008] S1. Construct a calibration dataset, and then use the calibration dataset and the Gaussian process regression model (GPR) to train and obtain the junction temperature estimation function;

[0009] S2. Obtain the actual flat band voltage and actual threshold voltage under actual operating conditions through the measurement circuit, and substitute the actual flat band voltage and actual threshold voltage into the junction temperature estimation function to obtain the current estimated value of the actual junction temperature.

[0010] S3. Substitute the current estimated junction temperature into the linear relationship between the flat-band voltage and threshold voltage before gate oxide degradation and the junction temperature to obtain the flat-band voltage and threshold voltage when the gate oxide is not degraded at that junction temperature. Then, calculate the difference between the actual flat-band voltage and actual threshold voltage and the flat-band voltage and threshold voltage when the gate oxide is not degraded at that junction temperature to obtain the flat-band voltage drift and threshold voltage drift. Finally, perform a comprehensive diagnosis of gate oxide degradation based on the flat-band voltage drift and threshold voltage drift to obtain the comprehensive diagnosis result.

[0011] Furthermore, step S1 specifically includes the following steps:

[0012] S11. Under a set gate bias voltage and a set test temperature, the gate oxide layer of the SiC MOSFET is subjected to accelerated degradation treatment to obtain the device state at different degradation stages; subsequently, at each degradation stage, the device parameters are measured at different test temperatures, and the test temperature is the junction temperature T of the device. j A calibration dataset for training the Gaussian process regression model (GPR) was thus collected; the gate bias voltage was set to +20 V to +30 V; and the test temperature was set to 125℃ to 175℃.

[0013] Each sample in the calibration dataset contains: flat band voltage V FB Threshold voltage V TH Degradation time t BTI With junction temperature T j ;

[0014] S12. Input the calibration dataset into the Gaussian process regression model GPR, and use multiple candidate kernel functions to fit and train the Gaussian process regression model GPR. By comparing the standard deviation (STD) and root mean square error (RMSE) of the fitting results of each kernel function, select the single kernel function or combination of kernel functions with the best fitting performance to construct a model for fitting the precursor parameters and junction temperature T. j The junction temperature estimation function is used for subsequent actual junction temperature estimation and comprehensive diagnosis of gate oxide degradation. The precursor parameters include the flat-band voltage V. FB Threshold voltage VTH .

[0015] Furthermore, the candidate kernel function is selected from one or more combinations of Gaussian radial basis function (RBF), Matern kernel, and linear kernel. The kernel function is used to fit the precursor parameters and junction temperature T. j The linear variation trend and nonlinear degradation characteristics between them are considered, while taking into account the linear drift and local nonlinear fluctuation characteristics of the device under actual working conditions.

[0016] Furthermore, the formula for calculating the junction temperature estimation function in S12 is:

[0017] ;

[0018] Where Φ represents the junction temperature estimation function obtained by training the Gaussian process regression model GPR, used to estimate the junction temperature based on the input flat-band voltage V. FB and threshold voltage V TH Estimate junction temperature T j .

[0019] Furthermore, step S2 specifically includes the following steps:

[0020] S21. Obtain the actual precursor parameters under actual operating conditions through a measurement circuit. The actual precursor parameters include the actual flat-band voltage. and actual threshold voltage ;

[0021] S22, The measured actual flat band voltage and actual threshold voltage Substituting these values ​​into the joint mapping junction temperature estimation function yields the current estimated actual junction temperature. ;

[0022] S23. Determine the current estimated actual junction temperature. If the expected temperature is met, proceed to S3; otherwise, trigger active thermal state management measures for the SiC MOSFET gate oxide layer and return to S21 to remeasure the actual precursor parameters, so as to realize real-time monitoring, abnormal warning and dynamic thermal management of the working junction temperature state of the SiC MOSFET gate oxide layer.

[0023] Furthermore, the estimated current actual junction temperature in S22 The formula for calculation is:

[0024] .

[0025] Furthermore, step S3 specifically includes the following steps:

[0026] S31. Based on the calibration dataset obtained in S1, fit the flat-band voltage V of the SiC MOSFET before gate oxide degradation. FB With junction temperature T j The linear relationship and the threshold voltage V before gate oxide degradation TH With junction temperature T j The linear relationship;

[0027] S32, Estimated current actual junction temperature Substituting into the flat-band voltage V before gate oxide degradation FB With junction temperature T j From the linear relationship, the flat-band voltage at this junction temperature without gate oxide degradation is obtained. ;

[0028] The current estimated actual junction temperature Substitute into the threshold voltage V before gate oxide degradation TH With junction temperature T j From the linear relationship, the threshold voltage at which the gate oxide layer does not degrade at this junction temperature is obtained. ;

[0029] S33, The actual flat band voltage Actual threshold voltage Respectively with flat band voltage Threshold voltage The difference is calculated to obtain the flat-band voltage drift. Threshold voltage drift ;

[0030] S34, via threshold voltage drift Reflects the degree of degradation of the gate oxide layer, and is expressed by the flat-band voltage drift. and threshold voltage drift Further determine the degradation location of the gate oxide layer, including the flat-band voltage drift. Reflects the degradation state of the gate oxide layer above the JFET region, threshold voltage drift. This reflects the degradation state of the gate oxide layer in the channel region; thus, the degree and location of degradation of the gate oxide layer of the SiC MOSFET are obtained.

[0031] S35. Based on the degree and location of SiC MOSFET gate oxide degradation, assess the health status or lifetime prediction of the SiC MOSFET gate oxide to obtain a comprehensive diagnostic result.

[0032] Furthermore, the flat band voltage The specific calculation formula is as follows:

[0033] ;

[0034] The threshold voltage The specific calculation formula is as follows:

[0035] ;

[0036] In the formula, a, b, c, and d are the temperature coefficients obtained during the calibration process under the condition that the gate oxide layer of the SiC MOSFET is not degraded.

[0037] Furthermore, the flat-band voltage drift in S33 The formula for calculation is:

[0038] ;

[0039] Threshold voltage drift in S33 The formula for calculation is:

[0040] .

[0041] In another aspect, the present invention provides a comprehensive diagnostic system for SiC MOSFET gate oxide degradation, including a device-side component such as a junction temperature state monitoring module. The device-side component performs a comprehensive diagnostic assessment of the degradation status of the SiC MOSFET gate oxide layer based on a comprehensive diagnostic method for SiC MOSFET gate oxide layer degradation. The junction temperature state monitoring module is used to output an estimated value of the current actual junction temperature based on a junction temperature estimation function. Real-time monitoring and thermal status management of the operating junction temperature of SiC MOSFETs.

[0042] The beneficial effects of this invention are:

[0043] 1. By decoupling temperature and degradation effects, diagnostic accuracy is improved:

[0044] This invention constructs a multivariate joint mapping relationship by introducing precursor parameters (flat-band voltage, threshold voltage), junction temperature, and degradation level into the training samples, and learns its coupling characteristics using a Gaussian process regression model (GPR). This joint mapping relationship is used to accurately estimate the current actual junction temperature during diagnosis. Combined with the linear relationship between the precursor parameters and junction temperature obtained through experimental calibration and fitting of the gate oxide layer in a degradation-free state, the theoretical baseline precursor parameter values ​​(i.e., the flat-band voltage and threshold voltage of the gate oxide layer without degradation at this junction temperature) are derived. By comparing the measured actual precursor parameters with the theoretical baseline precursor parameter values, the flat-band voltage drift and threshold voltage drift caused by degradation are obtained, thereby effectively decoupling the junction temperature and degradation effect. This allows for quantitative characterization of the degradation degree and location of the gate oxide layer, effectively avoiding misdiagnosis caused by temperature drift, improving the accuracy and robustness of gate oxide layer degradation diagnosis, and enabling real-time monitoring of the device junction temperature state and online early warning of thermal anomalies, further enhancing system safety and reliability.

[0045] 2. This invention has outstanding comprehensive diagnostic capabilities:

[0046] Compared to traditional methods that can only assess the degree or location of degradation, this invention can not only assess the severity of gate oxide degradation, but also accurately locate the degradation location (gate oxide above the channel region or JFET region), thus providing a more comprehensive assessment of the gate oxide health status.

[0047] 3. Low cost and easy to implement:

[0048] Compared with traditional methods, the method provided by this invention does not require periodic recalibration of the relationship between precursor parameters and junction temperature, thus reducing cost and operational complexity. Attached Figure Description

[0049] Figure 1 This is a flowchart of the comprehensive diagnostic method for SiC MOSFET gate oxide degradation in this invention. Detailed Implementation

[0050] To facilitate understanding of the present invention, a more complete description will be given below with reference to the accompanying drawings. Preferred embodiments of the invention are shown in the drawings. However, the invention can be implemented in many other different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided to provide a thorough and complete understanding of the disclosure of the invention.

[0051] Reference Figure 1 This application provides a comprehensive diagnostic method for SiC MOSFET gate oxide degradation, comprising the following steps:

[0052] S1. Construct a calibration dataset, and then use the calibration dataset and the Gaussian Process Regression (GPR) model to train and obtain the junction temperature estimation function;

[0053] S2. Obtain the actual flat-band voltage and actual threshold voltage under actual operating conditions through the measurement circuit. Substitute the actual flat-band voltage and actual threshold voltage into the junction temperature estimation function to obtain the estimated value of the current actual junction temperature (i.e., the junction temperature inside the device). This estimated junction temperature value can be used as input for subsequent degradation diagnosis and can also be fed back in real time for online monitoring and thermal management judgment of the device junction temperature status.

[0054] S3. Substitute the current estimated junction temperature into the linear relationship between the flat-band voltage and threshold voltage before gate oxide degradation and the junction temperature to obtain the flat-band voltage and threshold voltage when the gate oxide is not degraded at that junction temperature. Then, calculate the difference between the actual flat-band voltage and actual threshold voltage and the flat-band voltage and threshold voltage when the gate oxide is not degraded at that junction temperature to obtain the flat-band voltage drift and threshold voltage drift. Finally, perform a comprehensive diagnosis of gate oxide degradation based on the flat-band voltage drift and threshold voltage drift to obtain the comprehensive diagnosis result.

[0055] This invention constructs a joint mapping relationship between precursor parameters (including flat-band voltage and threshold voltage) and junction temperature and degradation level, namely, the linear relationship between flat-band voltage and junction temperature before gate oxide degradation and the linear relationship between threshold voltage and junction temperature before gate oxide degradation. By using the Gaussian process regression model (GPR), the temperature and degradation effects are decoupled. This decoupling effectively isolates the influence of junction temperature changes on diagnostic results, thereby solving the problem of misdiagnosis caused by temperature drift in the prior art and improving the accuracy and robustness of degradation diagnosis.

[0056] Furthermore, the method provided by this invention is based on the mapping relationship between precursor parameters, junction temperature, and degradation level, and combined with the Gaussian process regression model (GPR), which can accurately identify gate oxide degradation characteristics at different operating temperatures.

[0057] In some embodiments, S1 specifically includes the following steps:

[0058] S11. Under a set gate bias voltage and a set test temperature, the gate oxide layer of the SiC MOSFET is subjected to accelerated degradation treatment to obtain the device state at different degradation stages; subsequently, at each degradation stage, the device parameters are measured at different test temperatures, and the test temperature is the junction temperature T of the device. jThe calibration dataset for training the Gaussian process regression model (GPR) was thus collected; the gate bias voltage was set to +20 V ~ +30 V; the test temperature was set to 125℃ ~ 175℃; the test temperature should be higher than the normal operating temperature of the device to accelerate the accumulation of oxide layer defects, and generally should not exceed the maximum junction temperature of the device (usually around 175℃) to avoid thermal breakdown and interference from non-target degradation mechanisms.

[0059] Each sample in the calibration dataset contains: flat band voltage V FB Threshold voltage V TH Degradation time t BTI With junction temperature T j ;

[0060] S12. Input the calibration dataset into the Gaussian process regression model GPR, and use multiple candidate kernel functions to fit and train the Gaussian process regression model GPR. By comparing the standard deviation (STD) and root mean square error (RMSE) of the fitting results of each kernel function, select the single kernel function and kernel function combination with the best fitting performance to construct a model for fitting the precursor parameters and junction temperature T. j The junction temperature estimation function is used for subsequent actual junction temperature estimation and comprehensive diagnosis of gate oxide degradation. The precursor parameters include the flat-band voltage V. FB Threshold voltage V TH .

[0061] In some embodiments, the candidate kernel function is selected from one or more of the following: Radial Basis Function (RBF), Matern kernel, and linear kernel. The kernel function is used to fit the precursor parameters and the junction temperature T. j The model considers both the linear variation trend and nonlinear degradation characteristics between the parameters, while also taking into account the linear drift and local nonlinear fluctuations under actual device operating conditions. Typical combinations of kernel functions include the superposition of a linear kernel and a Gaussian radial basis function (RBF) kernel, the superposition of a linear kernel and a Matern kernel, or the superposition of a Gaussian radial basis function (RBF) kernel and a Matern kernel. The selection of kernel function combinations and parameter settings can be combined with the actual degradation characteristics of the SiC MOSFET gate oxide layer, device structural characteristics, and temperature drift patterns to optimize the fitting effect, ensure that the estimation results conform to physical characteristics, and reduce the risk of overfitting.

[0062] In some embodiments, the formula for calculating the junction temperature estimation function in S12 is:

[0063] ;

[0064] Where Φ represents the junction temperature estimation function obtained by training the Gaussian process regression model GPR, used to estimate the junction temperature based on the input flat-band voltage V. FB and threshold voltage V TH Estimate junction temperature T j .

[0065] In some embodiments, S2 specifically includes the following steps:

[0066] S21. Obtain the actual precursor parameters under actual operating conditions through a measurement circuit. The actual precursor parameters include the actual flat-band voltage. and actual threshold voltage ;

[0067] S22, The measured actual flat band voltage and actual threshold voltage Substituting these values ​​into the joint mapping junction temperature estimation function yields the current estimated actual junction temperature. ;

[0068] S23. Determine the current estimated actual junction temperature. If the expected temperature is met, proceed to S3; otherwise, trigger active thermal state management measures for the SiC MOSFET gate oxide layer and return to S21 to remeasure the actual precursor parameters, so as to realize real-time monitoring, abnormal warning and dynamic thermal management of the working junction temperature state of the SiC MOSFET gate oxide layer.

[0069] In some embodiments, the current actual junction temperature estimate in S22 The formula for calculation is:

[0070] .

[0071] In some embodiments, S3 specifically includes the following steps:

[0072] S31. Based on the calibration dataset obtained in S1, fit the flat-band voltage V of the SiC MOSFET before gate oxide degradation. FB With junction temperature T j The linear relationship and the threshold voltage V before gate oxide degradation TH With junction temperature T j The linear relationship;

[0073] S32, Estimated current actual junction temperature Substituting into the flat-band voltage V before gate oxide degradation FB With junction temperature T j From the linear relationship, the flat-band voltage at this junction temperature without gate oxide degradation is obtained. ;

[0074] The current estimated actual junction temperature Substitute into the threshold voltage V before gate oxide degradation TH With junction temperature T j From the linear relationship, the threshold voltage at which the gate oxide layer does not degrade at this junction temperature is obtained. ;

[0075] S33, The actual flat band voltage Actual threshold voltage Respectively with flat band voltage Threshold voltage The difference is calculated to obtain the flat-band voltage drift. Threshold voltage drift ;

[0076] S34, via threshold voltage drift Reflects the degree of degradation of the gate oxide layer, and is expressed by the flat-band voltage drift. and threshold voltage drift Further determine the degradation location of the gate oxide layer, including the flat-band voltage drift. Reflects the degradation state of the gate oxide layer above the JFET region, threshold voltage drift. This reflects the degradation state of the gate oxide layer in the channel region; thus, the degree and location of degradation of the gate oxide layer of the SiC MOSFET are obtained.

[0077] S35. Based on the degree and location of SiC MOSFET gate oxide degradation, assess the health status or lifetime prediction of the SiC MOSFET gate oxide to obtain a comprehensive diagnostic result.

[0078] In some embodiments, the flat band voltage The specific calculation formula is as follows:

[0079] ;

[0080] The threshold voltage The specific calculation formula is as follows:

[0081] ;

[0082] In the formula, a, b, c, and d are temperature coefficients obtained during calibration under conditions of no degradation of the SiC MOSFET gate oxide layer. These are temperature coefficients obtained through experimental calibration and fitting, and may vary for different SiC MOSFET models depending on differences in device structure and process parameters. Taking GC3M0075120D, SCT3080KLGC1, and IMW120R060M1H models as examples, the preferred coefficient values ​​are shown in Table 1 below:

[0083] Table 1: Optimal values ​​for temperature coefficients a, b, c, and d for different models;

[0084]

[0085] In some embodiments, the flat-band voltage drift in S33 The formula for calculation is:

[0086] ;

[0087] Threshold voltage drift in S33 The formula for calculation is:

[0088] .

[0089] Compared to traditional methods that can only assess the degree or location of degradation, this invention can not only assess the severity of gate oxide degradation, but also accurately locate the degradation location (gate oxide above the channel region or JFET region), thus providing a more comprehensive assessment of the gate oxide health status.

[0090] Furthermore, compared with traditional methods, the method provided by this invention does not require periodic recalibration of the relationship between precursor parameters and junction temperature, thus reducing cost and operational complexity.

[0091] In another aspect, the present invention provides a comprehensive diagnostic system for SiC MOSFET gate oxide degradation, including a device-side component such as a junction temperature state monitoring module. The device-side component performs a comprehensive diagnostic assessment of the degradation status of the SiC MOSFET gate oxide layer based on a comprehensive diagnostic method for SiC MOSFET gate oxide layer degradation. The junction temperature state monitoring module is used to output an estimated value of the current actual junction temperature based on a junction temperature estimation function. Real-time monitoring and thermal status management of the operating junction temperature of the gate oxide layer of SiC MOSFET.

[0092] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Furthermore, the technical solutions of the various embodiments of the present invention can be combined with each other, but this must be based on the ability of those skilled in the art to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed by the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A comprehensive diagnostic method for SiC MOSFET gate oxide degradation, characterized in that, Includes the following steps: S1. Construct a calibration dataset, and then use the calibration dataset and the Gaussian process regression model (GPR) to train and obtain the junction temperature estimation function; S2. Obtain the actual flat band voltage and actual threshold voltage under actual operating conditions through the measurement circuit, and substitute the actual flat band voltage and actual threshold voltage into the junction temperature estimation function to obtain the current estimated value of the actual junction temperature. S3. Substitute the current estimated junction temperature into the linear relationship between the flat band voltage and threshold voltage before gate oxide degradation and the junction temperature to obtain the flat band voltage and threshold voltage when the gate oxide is not degraded at that junction temperature. Then, calculate the difference between the actual flat band voltage and actual threshold voltage and the flat band voltage and threshold voltage when the gate oxide is not degraded at that junction temperature to obtain the flat band voltage drift and threshold voltage drift. Finally, perform a comprehensive diagnosis of gate oxide degradation based on the flat band voltage drift and threshold voltage drift to obtain the comprehensive diagnosis result.

2. The comprehensive diagnostic method for SiC MOSFET gate oxide degradation according to claim 1, characterized in that, S1 specifically includes the following steps: S11. Under a set gate bias voltage and a set test temperature, the gate oxide layer of the SiC MOSFET is subjected to accelerated degradation treatment to obtain the device state at different degradation stages; subsequently, at each degradation stage, the device parameters are measured at different test temperatures, and the test temperature is the junction temperature T of the device. j A calibration dataset for training the Gaussian process regression model (GPR) was thus collected; the gate bias voltage was set to +20 V to +30 V; and the test temperature was set to 125℃ to 175℃. Each sample in the calibration dataset contains: flat band voltage V FB Threshold voltage V TH Degradation time t BTI With junction temperature T j ; S12. Input the calibration dataset into the Gaussian process regression model GPR, and use multiple candidate kernel functions to fit and train the Gaussian process regression model GPR. By comparing the standard deviation (STD) and root mean square error (RMSE) of the fitting results of each kernel function, select the single kernel function or combination of kernel functions with the best fitting performance to construct a model for fitting the precursor parameters and junction temperature T. j The junction temperature estimation function is used for subsequent actual junction temperature estimation and comprehensive diagnosis of gate oxide degradation. The precursor parameters include the flat-band voltage V. FB Threshold voltage V TH .

3. The comprehensive diagnostic method for SiC MOSFET gate oxide degradation according to claim 2, characterized in that, The candidate kernel function is selected from one or more of the following: Gaussian radial basis function (RBF), Matern kernel, and linear kernel. The kernel function is used to fit the precursor parameters and the junction temperature T. j The linear variation trend and nonlinear degradation characteristics between them are considered, while taking into account the linear drift and local nonlinear fluctuation characteristics of the device under actual working conditions.

4. The comprehensive diagnostic method for SiC MOSFET gate oxide degradation according to claim 3, characterized in that, The formula for calculating the junction temperature estimation function in S12 is: ； Where Φ represents the junction temperature estimation function obtained by training the Gaussian process regression model GPR, used to estimate the junction temperature based on the input flat-band voltage V. FB and threshold voltage V TH Estimate junction temperature T j .

5. The comprehensive diagnostic method for SiC MOSFET gate oxide degradation according to claim 4, characterized in that, S2 specifically includes the following steps: S21. Obtain the actual precursor parameters under actual operating conditions through a measurement circuit. The actual precursor parameters include the actual flat-band voltage. and actual threshold voltage ; S22, Measure the actual flat band voltage and actual threshold voltage Substituting these values ​​into the joint mapping junction temperature estimation function yields the current estimated actual junction temperature. ; S23. Determine the current estimated actual junction temperature. If the expected temperature is met, proceed to S3; otherwise, trigger active thermal state management measures for the SiC MOSFET gate oxide layer and return to S21 to remeasure the actual precursor parameters, so as to realize real-time monitoring, abnormal warning and dynamic thermal management of the working junction temperature state of the SiC MOSFET gate oxide layer.

6. The comprehensive diagnostic method for SiC MOSFET gate oxide degradation according to claim 5, characterized in that, The estimated current actual junction temperature in S22 The formula for calculation is: 。 7. The comprehensive diagnostic method for SiC MOSFET gate oxide degradation according to claim 6, characterized in that, S3 specifically includes the following steps: S31. Based on the calibration dataset obtained in S1, fit the flat-band voltage V of the SiC MOSFET before gate oxide degradation. FB With junction temperature T j The linear relationship and the threshold voltage V before gate oxide degradation TH With junction temperature T j The linear relationship; S32, Estimated current actual junction temperature Substituting into the flat-band voltage V before gate oxide degradation FB With junction temperature T j From the linear relationship, the flat-band voltage at this junction temperature without gate oxide degradation is obtained. ; The current estimated actual junction temperature Substitute into the threshold voltage V before gate oxide degradation TH With junction temperature T j From the linear relationship, the threshold voltage at which the gate oxide layer does not degrade at this junction temperature is obtained. ; S33, The actual flat band voltage Actual threshold voltage Respectively with flat band voltage Threshold voltage The difference is calculated to obtain the flat-band voltage drift. Threshold voltage drift ; S34, via threshold voltage drift Reflects the degree of degradation of the gate oxide layer, and is expressed by the flat-band voltage drift. and threshold voltage drift Further determine the degradation location of the gate oxide layer, including the flat-band voltage drift. Reflects the degradation state of the gate oxide layer above the JFET region, threshold voltage drift. This reflects the degradation state of the gate oxide layer in the channel region; thus, the degree and location of degradation of the gate oxide layer of the SiC MOSFET are obtained. S35. Based on the degree and location of SiC MOSFET gate oxide degradation, assess the health status or lifetime prediction of the SiC MOSFET gate oxide to obtain a comprehensive diagnostic result.

8. The comprehensive diagnostic method for SiC MOSFET gate oxide degradation according to claim 7, characterized in that, The flat band voltage The specific calculation formula is as follows: ； The threshold voltage The specific calculation formula is as follows: ； In the formula, a, b, c, and d are the temperature coefficients obtained during the calibration process under the condition that the gate oxide layer of the SiC MOSFET is not degraded.

9. The comprehensive diagnostic method for SiC MOSFET gate oxide degradation according to claim 8, characterized in that, The flat-band voltage drift in S33 The formula for calculation is: ； Threshold voltage drift in S33 The formula for calculation is: 。 10. A comprehensive diagnostic system for SiC MOSFET gate oxide degradation, characterized in that, The device side, including the junction temperature state monitoring module, performs a comprehensive diagnosis of the degradation status of the SiC MOSFET gate oxide layer according to the comprehensive diagnostic method for SiC MOSFET gate oxide layer degradation as described in any one of claims 1 to 9. The junction temperature state monitoring module is used to estimate the current actual junction temperature based on the junction temperature estimation function output. Real-time monitoring and thermal status management of the operating junction temperature of SiC MOSFETs.

Images