A method for testing surge performance of silicon carbide MOSFET

Abstract

The present application relates to the field of semiconductor device testing technology, and specifically to a method for testing the surge performance of a silicon carbide MOSFET, comprising: performing a surge performance test on the silicon carbide MOSFET at different surge current peaks to obtain each source-drain voltage vector; extracting each IMF modal component of the source-drain voltage vector, analyzing the irregular complexity of each IMF modal component, determining the time domain interference degree based on the mutation points within each IMF modal component, obtaining the frequency domain interference degree based on the characteristics of each IMF modal component in the frequency domain, calculating the time-frequency interference degree of each IMF modal component, screening the target component in the source-drain voltage vector, reconstructing the waveform using a wavelet fusion algorithm, and obtaining the maximum surge current of the silicon carbide MOSFET by comparing the reconstructed waveforms of each source-drain voltage. This application can improve the accuracy of the surge performance test results of silicon carbide MOSFETs.

Description

Technical Field

[0001] The present application relates to the field of semiconductor device testing technology, and in particular to a method for testing surge performance of a silicon carbide MOSFET. Background Art

[0002] In the development process of semiconductor materials and semiconductor devices, semiconductor devices based on silicon carbide materials are the third generation of semiconductor devices. Compared with traditional semiconductor devices based on silicon materials, the electron saturation drift rate of silicon carbide materials is higher than that of silicon materials, which makes MOSFETs based on silicon carbide materials have higher current density and faster switching speed.

[0003] Although silicon carbide semiconductor devices have many advantages, their surge performance has certain problems due to the limitations of the silicon carbide MOSFET gate oxide process. Surge performance is an important indicator for evaluating the reliability of silicon carbide semiconductor devices. Therefore, it is necessary to perform surge performance testing on silicon carbide MOSFETs to improve the surge performance of silicon carbide semiconductor devices. For example, Chinese invention patent application publication number CN113466649A discloses a method for determining the cause of SiC MOSFET failure during surge current testing. The method uses an LC resonant circuit to generate a surge current to perform a surge performance test on the silicon carbide MOSFET, and the maximum surge current capability before the silicon carbide MOSFET fails is obtained, which can realize the surge performance test of the silicon carbide MOSFET. However, since the LC resonant circuit is prone to generating high-frequency electromagnetic fields during operation, and the electromagnetic waves in the high-frequency electromagnetic fields propagate in the silicon carbide MOSFET surge performance test environment, they will cause certain electromagnetic interference to the surge performance test results, resulting in inaccurate surge performance test results of the silicon carbide MOSFET.

[0004] Therefore, when using an LC resonant circuit to perform surge testing on silicon carbide MOSFETs, how to avoid the impact of electromagnetic interference on the surge performance test results and improve the accuracy of the surge performance test results of silicon carbide MOSFETs has become one of the technical problems to be solved. Summary of the Invention

[0005] In order to solve the above technical problems, the present application provides a surge performance testing method for silicon carbide MOSFET to solve the existing problems.

[0006] The present invention provides a method for testing the surge performance of a silicon carbide MOSFET using the following technical solutions:

[0007] An embodiment of the present application provides a method for testing surge performance of a silicon carbide MOSFET, comprising the following steps:

[0008] Perform surge performance tests on silicon carbide MOSFETs at different surge current peaks, and sample the source-drain voltage waveform of the silicon carbide MOSFET during each test to obtain the source-drain voltage vector during each test.

[0009] Extracting each IMF modal component of the source-drain voltage vector, analyzing the difference in the change of the modal value within each IMF modal component and the degree of chaos of the change of the modal value within each IMF modal component, determining the irregular complexity of each IMF modal component, and obtaining the mutation distribution distance of each IMF modal component based on the number of mutation points in each IMF modal component and the distance relationship between the mutation points. Combined with the irregular complexity, the time domain interference degree of each IMF modal component is determined;

[0010] The frequency domain interference degree of each IMF modal component is obtained by the average level of the frequency domain energy of all frequencies in the frequency domain and the change of the frequency domain energy. Combined with the time domain interference degree of each IMF modal component, the time-frequency interference degree of each IMF modal component is determined.

[0011] Based on the time-frequency interference of all IMF modal components of each source-drain voltage vector, the target component in each source-drain voltage vector is screened, and the waveform is reconstructed using the wavelet fusion algorithm to obtain the reconstructed source-drain voltage waveform for each test process. By comparing the reconstructed waveforms of each source-drain voltage, the maximum surge current of the silicon carbide MOSFET is obtained.

[0012] Preferably, the method for obtaining the source-drain voltage vector in each test process is:

[0013] The source-drain voltages of the source-drain voltage waveform are sampled in each test process respectively, and a vector consisting of all the source-drain voltages sampled in each test process is arranged in chronological order as the source-drain voltage vector in each test process.

[0014] Preferably, the calculation method of the irregular complexity of each IMF modal component is:

[0015] R d =(Xu d +Xv d ) / (|Pu d -Pv d |+∈), where R d is the irregular complexity of the d-th IMF modal component, Xu d and Xv d are the element means of the first-order difference vector and the element means of the second-order difference vector of the d-th IMF modal component, Pu d and Pv dare the information entropy of all elements in the first-order difference vector and the information entropy of all elements in the second-order difference vector of the d-th IMF modal component, respectively, and ∈ is a constant to avoid the denominator being 0.

[0016] Preferably, the calculation method of the mutation distribution distance of each IMF modal component is:

[0017] Where, F d is the mutation distribution distance of the d-th IMF modal component, N d is the number of mutation points in the d-th IMF modal component, dist() is the Euclidean distance function, Y j and Y j-1 are the j-th and j-1-th mutation points in the d-th IMF modal component respectively.

[0018] Preferably, the time domain interference degree of each IMF modal component is a normalized result of the sum of the irregular complexity and the mutation distribution distance of each IMF modal component.

[0019] Preferably, the frequency domain interference degree of each IMF modal component is calculated as follows:

[0020] Where G d is the frequency domain interference degree of the d-th IMF modal component, norm[] is the exponential normalization function, M is the number of elements in the frequency domain energy vector of the d-th IMF modal component, wherein each IMF modal component is transformed into the frequency domain to construct the frequency domain energy vector of each IMF modal component, Xc d is the mean value of the frequency domain energy vector of the d-th IMF modal component, K d,g and K d,g-1 are the slopes of the gth and g-1th elements in the frequency domain energy vector of the dth IMF modal component, respectively, and ∈ is a constant to avoid the denominator being zero.

[0021] Preferably, the method for obtaining the frequency domain energy vector is: a vector composed of the frequency domain energies of all frequencies in each IMF modal component arranged in ascending order is used as the frequency domain energy vector of each IMF modal component.

[0022] Preferably, the time-frequency interference degree of each IMF modal component is the sum of the time domain interference degree and the frequency domain interference degree of each IMF modal component.

[0023] Preferably, screening the target component in each source-drain voltage vector further includes:

[0024] For the time-frequency interference degrees of all IMF modal components of each source-drain voltage vector, the remaining IMF modal components except the IMF modal component corresponding to the maximum time-frequency interference degree are used as the target components in each source-drain voltage vector.

[0025] Preferably, obtaining the maximum surge current of the silicon carbide MOSFET further comprises:

[0026] If the source-drain voltage reconstructed waveform of the current test process shows abnormal deformation, and the source-drain voltage reconstructed waveform of the previous test process does not show abnormal deformation, the surge current when the abnormal deformation occurs will be used as the maximum surge current of the silicon carbide MOSFET;

[0027] If no abnormal deformation occurs in the source-drain voltage reconstructed waveform up to the current test process, continue to increase the surge current peak value for testing until abnormal deformation occurs in the source-drain voltage reconstructed waveform during the test process. The surge current when abnormal deformation occurs is taken as the maximum surge current of the silicon carbide MOSFET.

[0028] This application has at least the following beneficial effects:

[0029] This application takes into account the irregular and complex characteristics and mutation point distribution characteristics within the electromagnetic interference components. By accurately measuring the irregular and complex characteristics and mutation point distribution characteristics of each IMF modal component, and combining these two characteristics to construct the time domain interference degree, it is used to more accurately describe the electromagnetic interference characteristics of each IMF modal component in the future, eliminate the electromagnetic interference components in the IMF modal components, and improve the accuracy of surge performance testing of silicon carbide MOSFETs.

[0030] At the same time, based on the time domain interference index, the present application uses discrete Fourier transform to analyze the frequency domain interference characteristics of each IMF modal component to construct the frequency domain interference index. The frequency domain interference index reflects the characteristics of high intensity and wide frequency range of electromagnetic interference. The time domain interference index and the frequency domain interference index are combined to construct the time-frequency interference index, which is used to more accurately identify the electromagnetic interference component within the IMF modal component, and then eliminate the electromagnetic interference component, thereby improving the accuracy of subsequent surge performance testing of silicon carbide MOSFETs.

[0031] This application accurately identifies the electromagnetic interference component within the IMF modal component through the time-frequency interference degree, and then uses the wavelet fusion algorithm to obtain the source-drain voltage reconstructed waveform after eliminating the electromagnetic interference component. By performing surge performance testing on the source-drain voltage reconstructed waveform after eliminating the electromagnetic interference component, it can avoid the impact of electromagnetic interference on the surge performance test results and improve the accuracy of the surge performance test results of silicon carbide MOSFET. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present application or the prior art, the following is a brief introduction to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0033] Figure 1 This is a flowchart of the steps of a surge performance testing method for a silicon carbide MOSFET provided in this application. DETAILED DESCRIPTION

[0034] To further illustrate the technical means and effectiveness of this application to achieve the intended invention objectives, the following, in conjunction with the accompanying drawings and preferred embodiments, describes in detail a method for testing the surge performance of a silicon carbide MOSFET proposed in this application, including its specific implementation, structure, features, and effectiveness. In the following description, different references to "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.

[0035] Unless otherwise defined, terms such as "comprises," "comprising," or any other variants thereof are intended to encompass non-exclusive inclusion, such that a circuit structure, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such article or device. In the absence of further restrictions, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the article or device comprising the element. In addition, the term "and\or" as used herein includes any and all combinations of one or more related listed items. All technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application pertains.

[0036] The specific scheme of the surge performance testing method of a silicon carbide MOSFET provided by the present application is described in detail below with reference to the accompanying drawings.

[0037] An embodiment of the present application provides a method for testing the surge performance of a silicon carbide MOSFET. For details, please refer to Figure 1 , including the following steps:

[0038] In this embodiment, a surge performance test of a silicon carbide MOSFET is performed by using an LC resonant circuit to provide a surge current, aiming to solve the problem of electromagnetic interference in the electromagnetic field when the LC resonant circuit is working affecting the surge performance test results, and to improve the accuracy of the surge performance test results of the silicon carbide MOSFET.

[0039] Step 1: Perform surge performance tests on the silicon carbide MOSFET under different surge current peaks, and sample the source-drain voltage waveform of the silicon carbide MOSFET during each test to obtain the source-drain voltage vector during each test.

[0040] An LC resonant circuit composed of a capacitor and an inductor is used, and the surge current generated by the LC resonant circuit is input into the silicon carbide MOSFET for surge performance testing. The silicon carbide MOSFET is measured using the probe of the test probe station to obtain the source-drain voltage waveforms under different surge current peaks.

[0041] Through the DC voltage V DC The surge current peak value generated by the LC resonant circuit is adjusted, and the surge current peak value is gradually increased for testing to obtain the source-drain voltage waveforms during N tests, which correspond to the source-drain voltage waveforms under N different surge current peak values.

[0042] However, since the LC resonant circuit easily generates high-frequency electromagnetic fields during operation, the electromagnetic waves in this high-frequency electromagnetic field propagate in the SiC MOSFET surge performance test environment, causing certain electromagnetic interference to the source-drain voltage waveform, resulting in inaccurate surge performance test results for the SiC MOSFET. Therefore, in order to improve the accuracy of the SiC MOSFET surge performance test results, it is necessary to eliminate the electromagnetic interference characteristics in the source-drain voltage waveform.

[0043] In order to facilitate the subsequent elimination of the electromagnetic interference characteristics in the source-drain voltage waveform, the source-drain voltage of the source-drain voltage waveform is sampled separately during each test process, and the vector composed of all the drain-source voltages sampled each time is arranged in chronological order and recorded as the source-drain voltage vector during each test process, where the sampling rate of the source-drain voltage is 5kHz.

[0044] Step 2: Extract each IMF modal component of the source-drain voltage vector, analyze the change difference of the modal value within each IMF modal component and the degree of chaos of the modal value change within each IMF modal component, determine the irregular complexity of each IMF modal component, and obtain the mutation distribution distance of each IMF modal component based on the number of mutation points in each IMF modal component and the distance relationship between the mutation points. Combined with the irregular complexity, determine the time domain interference degree of each IMF modal component.

[0045] If the silicon carbide MOSFET does not reach the maximum surge current that causes failure, the source-drain voltage waveform often shows the characteristics of first rising and then falling, and the slope of the source-drain voltage change shows a regular change from large to small. However, since the LC resonant circuit is prone to generate high-frequency electromagnetic fields when working, it will produce strong electromagnetic interference to the source-drain voltage changes under normal circumstances, causing irregular interference changes in the source-drain voltage changes under normal circumstances, thereby affecting the subsequent test of the silicon carbide MOSFET surge performance.

[0046] Through the above analysis, the analysis is performed based on the source-drain voltage vector of any test process. In this embodiment, the source-drain voltage vector of the n-th test process is taken as an example for detailed description. Specifically, the source-drain voltage vector of the n-th test process is input into the traditional VMD variational mode decomposition algorithm (Variational Mode Decomposition), wherein the number of modes of VMD decomposition is 4, and the penalty factor is 2000. The various IMF modal components of the source-drain voltage vector are obtained through the traditional VMD variational mode decomposition algorithm, and the data storage form in the IMF modal component is in vector form.

[0047] Since the VMD variational mode decomposition algorithm separates multiple target components and electromagnetic interference components in the source-drain voltage vector, in order to more accurately eliminate the electromagnetic interference components generated by the high-frequency electromagnetic field, it is necessary to analyze the change characteristics of each IMF modal component.

[0048] In order to more accurately extract the target component in the source-drain voltage vector and avoid the impact of electromagnetic interference generated by the high-frequency electromagnetic field on the subsequent surge performance test, this embodiment calculates the first-order difference vector and second-order difference vector of each IMF modal component of the source-drain voltage vector. The first-order difference vector and the second-order difference vector reflect the first-order change difference and second-order change difference of the modal value in the IMF modal component over time. Since the elements in the electromagnetic interference component have irregular changes, if the first-order difference vector and the second-order difference vector of the IMF modal component have a high level of change difference at the same time, and the more similar the degree of confusion between the first-order change difference and the second-order change difference, the more complex irregular changes in the electromagnetic interference component can be reflected, that is, the IMF modal component is more likely to be an electromagnetic interference component.

[0049] Through the above analysis, the irregular complexity of each IMF modal component is calculated:

[0050] R d =(Xu d +Xv d ) / (|Pu d -pv d |+∈), where R dis the irregular complexity of the d-th IMF modal component, Xu d and Xv d are the element means of the first-order difference vector and the element means of the second-order difference vector of the d-th IMF modal component, Pu d and Pv d are the information entropy of all elements in the first-order difference vector and the information entropy of all elements in the second-order difference vector of the d-th IMF modal component, respectively. ∈ is a constant to avoid a denominator of 0, and its value range is 0 to 0.1. In this embodiment, it is 0.001. The calculation of information entropy is a well-known technology and the specific process is not repeated here. The higher the average level of the first-order difference value and the second-order difference value of the IMF modal component, the more chaotic the change of the elements in the IMF modal component is. At the same time, the information entropy technology is used to measure the degree of chaos of all elements in the first-order difference sequence and the second-order difference sequence. If the difference between the information entropies of the two is smaller, the more similar the degree of chaos between the first-order change difference and the second-order change difference of the IMF modal component is, the more complex irregular changes in the electromagnetic interference component can be highlighted, and the greater the irregular complexity.

[0051] At the same time, if a certain IMF modal component is an electromagnetic interference component, due to the irregular characteristics of electromagnetic interference, more mutation points are likely to appear on the IMF modal component, and the distribution range of the mutation points is relatively wide; and if the silicon carbide MOSFET fails during the surge test, that is, the surge current at this time exceeds the maximum surge current that the silicon carbide MOSFET can withstand, a mutation point will also appear near the peak value of the target component in the source-drain voltage vector, and the mutation points at this time are often distributed in a local range near the peak value.

[0052] Each IMF modal component of the source-drain voltage vector is input into the traditional Pettitt mutation point detection algorithm, where the significance level is set to 0.05, and all the mutation points in each IMF modal component of the source-drain voltage vector are detected by the traditional Pettitt mutation point detection algorithm;

[0053] Through the above analysis, according to the number of mutation points in each IMF modal component and the distance relationship between the mutation points, the mutation distribution distance of each IMF modal component is calculated:

[0054] Where, F d is the mutation distribution distance of the d-th IMF modal component, N d is the number of mutation points in the d-th IMF modal component, dist() is the Euclidean distance function, Y j and Y j-1 are the jth and j-1th mutation points in the dth IMF modal component, respectively, dist(Y j,Y j-1 ) is the Euclidean distance between the jth mutation point and the j-1th mutation point in the dth IMF modal component. The greater the number of mutation points in an IMF modal component and the larger the Euclidean distance between adjacent mutation points, the more likely it is that the electromagnetic interference component has more mutation points and a wider distribution of mutation points. In other words, the more likely the IMF modal component is an electromagnetic interference component, the larger the mutation distribution distance.

[0055] Through the above time domain analysis of each IMF modal component of the source-drain voltage vector, the irregular complexity and mutation distribution distance of each IMF modal component are extracted. The irregular complexity and mutation distribution distance simultaneously measure the different characteristics of the electromagnetic interference component. The normalized result of the sum of the irregular complexity and mutation distribution distance of each IMF modal component is recorded as the time domain interference degree of each IMF modal component. The greater the time domain interference degree, the more the IMF modal component can reflect the characteristics of the electromagnetic interference component from the time domain perspective, that is, the more likely the IMF modal component is an electromagnetic interference component. It should be noted that there are many normalization methods, which can be selected by the implementer. In this embodiment, the Z-Score normalization method is adopted. The specific process is a well-known technology and will not be described in detail in this embodiment.

[0056] Step 3: The frequency domain interference degree of each IMF modal component is obtained by the average level of the frequency domain energy of each IMF modal component at all frequencies in the frequency domain and the change of the frequency domain energy. Combined with the time domain interference degree of each IMF modal component, the time-frequency interference degree of each IMF modal component is determined.

[0057] To more accurately measure the electromagnetic interference characteristics of each IMF modal component and thus more accurately identify the electromagnetic interference component within each IMF modal component, each IMF modal component of the source-drain voltage vector is input into a conventional discrete Fourier transform (DFT). Frequency-domain energy for all frequencies within each IMF modal component is obtained through the conventional DFT. It should be noted that the specific DFT process is well known and will not be described in detail in this embodiment.

[0058] During the surge performance test of silicon carbide MOSFET, due to the high intensity of electromagnetic interference and the wide frequency range, the frequency domain energy within the frequency range is at a high level, reflecting that the electromagnetic waves generated by the LC resonant circuit have a large electromagnetic interference capability on the surge performance test of silicon carbide MOSFET.

[0059] In order to analyze the frequency domain electromagnetic interference characteristics of each IMF modal component, the frequency domain energies of all frequencies in each IMF modal component are arranged in ascending order to form a vector, which is recorded as the frequency domain energy vector of each IMF modal component. The frequency domain energy vector reflects the change of the frequency domain energy of all frequencies in the IMF modal component from small to large. If the relative change of the frequency domain energy of all frequencies from small to large is smoother and the average level of the frequency domain energy is higher, the more it can reflect the characteristics of greater electromagnetic interference intensity and wider frequency range, and the IMF modal component can better reflect the electromagnetic interference characteristics in the frequency domain.

[0060] Through the above analysis, in this embodiment, the frequency domain interference degree of each IMF modal component is calculated by the average level of the frequency domain energy vector of each IMF modal component and the change of the elements in the frequency domain energy vector. The specific calculation formula is:

[0061] Where G d is the frequency domain interference degree of the d-th IMF modal component, norm[] is the exponential normalization function, M is the number of elements in the frequency domain energy vector of the d-th IMF modal component, Xc d is the mean value of the frequency domain energy vector of the d-th IMF modal component, K d,g and K d,g-1 are respectively the slopes of the g-th and g-1-th element positions in the frequency domain energy vector of the d-th IMF modal component. The calculation of the slope is a well-known technology and the specific process will not be repeated here.

[0062] It can be understood that if the relative change of the frequency domain energy of all frequencies in the IMF modal component from small to large is smoother, that is, the slope change of the adjacent element positions is smaller, and at the same time, the average level of the frequency domain energy of all frequencies in the IMF modal component is higher, then the IMF modal component can better reflect the electromagnetic interference characteristics in the frequency domain, and the frequency domain interference degree after exponential normalization is greater, that is, the greater the frequency domain interference degree, the more the IMF modal component can reflect the characteristics of the electromagnetic interference component from the perspective of the frequency domain, that is, the more likely the IMF modal component is an electromagnetic interference component.

[0063] Furthermore, the electromagnetic interference components in all IMF modal components are extracted by combining time-domain and frequency-domain features. The sum of the time-domain interference degree and the frequency-domain interference degree of each IMF modal component is recorded as the time-frequency interference degree of each IMF modal component. The larger the time-frequency interference degree, the more likely the IMF modal component is to be an electromagnetic interference component.

[0064] Step 4: Based on the time-frequency interference of all IMF modal components of each source-drain voltage vector, the target components in each source-drain voltage vector are screened, and the waveform is reconstructed using the wavelet fusion algorithm to obtain the reconstructed waveform of the source-drain voltage for each test process. By comparing the reconstructed waveforms of each source-drain voltage, the maximum surge current of the silicon carbide MOSFET is obtained.

[0065] In order to eliminate the influence of electromagnetic interference on the surge performance test, and thus more accurately test the surge performance of silicon carbide MOSFET, in this embodiment, for the source-drain voltage vector of the nth test process, the IMF modal component corresponding to the maximum time-frequency interference is recorded as the electromagnetic interference component generated when the LC resonant circuit is working, and the remaining IMF modal components are recorded as the target components in the source-drain voltage vector of the nth test process, and the target components in the source-drain voltage vector of each test process are input into a traditional wavelet fusion algorithm, and the waveform of the source-drain voltage vector is reconstructed. The source-drain voltage reconstructed waveform of each test process is obtained by the traditional wavelet fusion algorithm, wherein the wavelet fusion algorithm is a well-known technology and the specific process is not repeated here.

[0066] By comparing the source-drain voltage reconstructed waveforms of each test process, the maximum surge current that the silicon carbide MOSFET can withstand is tested. Specifically, in this embodiment, if the source-drain voltage reconstructed waveform of the current test process shows abnormal deformation, and the source-drain voltage reconstructed waveform of the previous test process does not show abnormal deformation, it indicates that the silicon carbide MOSFET has experienced surge failure during the current surge performance test process, and the surge current when the abnormal deformation occurs is recorded as the maximum surge current that the silicon carbide MOSFET can withstand. If no abnormal deformation occurs in the source-drain voltage reconstructed waveform up to the current test process, the surge current peak value is gradually increased and tested until abnormal deformation occurs in the source-drain voltage reconstructed waveform of the test process, and the maximum surge current that the silicon carbide MOSFET can withstand is obtained. It should be noted that there are many existing technologies for detecting waveform abnormal deformation. In this embodiment, a measurement method based on FFT (Fast Fourier Transform) is used to analyze and detect waveform distortion. The specific process is well known and will not be described in detail in this embodiment.

[0067] It is understood that references to "one embodiment" or "some embodiments" in the present specification mean that one or more embodiments of the present application include a particular feature, structure, or characteristic described in conjunction with that embodiment. Thus, if "in one embodiment," "in some embodiments," "in other embodiments," or "in other embodiments" appear in different places in this specification, they do not necessarily refer to the same embodiment, but rather mean "one or more but not all embodiments," unless otherwise specifically emphasized. The terms "including," "comprising," "having," and their variations all mean "including but not limited to," unless otherwise specifically emphasized.

[0068] It should be noted that the above-mentioned sequence of the embodiments of the present application is for description only and does not represent the advantages and disadvantages of the embodiments. The above description is of a specific embodiment of this specification. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-tasking and parallel processing are also possible or may be advantageous. At the same time, the size of the sequence number of each step in the embodiment does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments in this specification.

[0069] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present application, and should all be included in the scope of protection of the present application.

Claims

1. A method for testing surge performance of a silicon carbide MOSFET, characterized in that: The following steps are involved: Perform surge performance tests on silicon carbide MOSFETs at different surge current peaks, and sample the source-drain voltage waveform of the silicon carbide MOSFET during each test to obtain the source-drain voltage vector during each test. Extracting each IMF modal component of the source-drain voltage vector, analyzing the difference in the change of the modal value within each IMF modal component and the degree of chaos of the change of the modal value within each IMF modal component, determining the irregular complexity of each IMF modal component, and obtaining the mutation distribution distance of each IMF modal component based on the number of mutation points in each IMF modal component and the distance relationship between the mutation points. Determining the time domain interference degree of each IMF modal component in combination with the irregular complexity; The frequency domain interference degree of each IMF modal component is obtained by the average level of the frequency domain energy of all frequencies in the frequency domain and the change of the frequency domain energy. Combined with the time domain interference degree of each IMF modal component, the time-frequency interference degree of each IMF modal component is determined. Based on the time-frequency interference of all IMF modal components of each source-drain voltage vector, the target component in each source-drain voltage vector is screened, and the waveform is reconstructed using a wavelet fusion algorithm to obtain the reconstructed source-drain voltage waveform for each test process. By comparing the reconstructed source-drain voltage waveforms, the maximum surge current of the silicon carbide MOSFET is obtained. The calculation method of the irregular complexity of each IMF modal component is: , where is the irregular complexity of the d-th IMF modal component, and are the element means of the first-order difference vector and the second-order difference vector of the d-th IMF modal component, respectively. and are the information entropy of all elements in the first-order difference vector and the information entropy of all elements in the second-order difference vector of the d-th IMF modal component, respectively. To avoid constants with denominators equal to 0.

2. A method for testing surge performance of a silicon carbide MOSFET according to claim 1, characterized in that: The method for obtaining the source-drain voltage vector during each test is as follows: The source-drain voltages of the source-drain voltage waveform are sampled in each test process respectively, and a vector consisting of all the source-drain voltages sampled in each test process is arranged in chronological order as the source-drain voltage vector in each test process.

3. The method for testing surge performance of a silicon carbide MOSFET according to claim 1, wherein: The calculation method of the mutation distribution distance of each IMF modal component is: , where is the mutation distribution distance of the d-th IMF modal component, is the number of mutation points within the d-th IMF modal component, is the Euclidean distance function, and are the j-th and j-1-th mutation points in the d-th IMF modal component respectively.

4. The method for testing surge performance of a silicon carbide MOSFET according to claim 1, wherein: The time domain interference degree of each IMF modal component is the normalized result of the sum of the irregular complexity and the mutation distribution distance of each IMF modal component.

5. The method for testing surge performance of a silicon carbide MOSFET according to claim 1, wherein: The calculation method of the frequency domain interference degree of each IMF modal component is: , where is the frequency domain interference degree of the d-th IMF modal component, is the exponential normalization function, is the number of elements in the frequency domain energy vector of the d-th IMF modal component, where each IMF modal component is transformed into the frequency domain to construct the frequency domain energy vector of each IMF modal component. is the mean value of the frequency domain energy vector of the d-th IMF modal component, and are the slopes of the gth and g-1th element positions in the frequency domain energy vector of the dth IMF modal component, To avoid constants with denominators equal to 0.

6. A method for testing surge performance of a silicon carbide MOSFET according to claim 5, characterized in that: The frequency domain energy vector is obtained by arranging the frequency domain energies of all frequencies in each IMF modal component in ascending order to form a vector as the frequency domain energy vector of each IMF modal component.

7. A method for testing surge performance of a silicon carbide MOSFET according to claim 1, characterized in that: The time-frequency interference degree of each IMF modal component is the sum of the time domain interference degree and the frequency domain interference degree of each IMF modal component.

8. The method for testing surge performance of a silicon carbide MOSFET according to claim 1, wherein: The screening of target components in each source-drain voltage vector further includes: For the time-frequency interference degrees of all IMF modal components of each source-drain voltage vector, the remaining IMF modal components except the IMF modal component corresponding to the maximum time-frequency interference degree are used as the target components in each source-drain voltage vector.

9. The method for testing surge performance of a silicon carbide MOSFET according to claim 1, wherein: Obtaining the maximum surge current of the silicon carbide MOSFET further includes: If the source-drain voltage reconstructed waveform of the current test process shows abnormal deformation, and the source-drain voltage reconstructed waveform of the previous test process does not show abnormal deformation, the surge current when the abnormal deformation occurs will be used as the maximum surge current of the silicon carbide MOSFET; If no abnormal deformation occurs in the source-drain voltage reconstructed waveform up to the current test process, continue to increase the surge current peak value for testing until abnormal deformation occurs in the source-drain voltage reconstructed waveform during the test process. The surge current when abnormal deformation occurs is taken as the maximum surge current of the silicon carbide MOSFET.

Images