CN108519547B - SiC-GTO device state monitoring method based on DTW and SVM - Google Patents
SiC-GTO device state monitoring method based on DTW and SVM Download PDFInfo
- Publication number
- CN108519547B CN108519547B CN201810266737.6A CN201810266737A CN108519547B CN 108519547 B CN108519547 B CN 108519547B CN 201810266737 A CN201810266737 A CN 201810266737A CN 108519547 B CN108519547 B CN 108519547B
- Authority
- CN
- China
- Prior art keywords
- sic
- svm
- gto
- dtw
- gto device
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 39
- 238000000034 method Methods 0.000 title claims abstract description 32
- 230000036541 health Effects 0.000 claims abstract description 12
- 238000012360 testing method Methods 0.000 claims abstract description 11
- 238000013461 design Methods 0.000 claims abstract description 4
- 238000012706 support-vector machine Methods 0.000 claims description 24
- 239000011159 matrix material Substances 0.000 claims description 10
- 238000012549 training Methods 0.000 claims description 8
- 238000013507 mapping Methods 0.000 claims description 2
- 230000007246 mechanism Effects 0.000 claims description 2
- 238000004364 calculation method Methods 0.000 abstract description 6
- 230000000903 blocking effect Effects 0.000 description 7
- 230000015556 catabolic process Effects 0.000 description 3
- 230000007547 defect Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000006731 degradation reaction Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000001186 cumulative effect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005684 electric field Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000003862 health status Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/26—Testing of individual semiconductor devices
- G01R31/2642—Testing semiconductor operation lifetime or reliability, e.g. by accelerated life tests
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/26—Testing of individual semiconductor devices
- G01R31/2607—Circuits therefor
- G01R31/2637—Circuits therefor for testing other individual devices
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/245—Classification techniques relating to the decision surface
- G06F18/2451—Classification techniques relating to the decision surface linear, e.g. hyperplane
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Testing Of Individual Semiconductor Devices (AREA)
Abstract
The invention provides a state monitoring method of a SiC-GTO device based on DTW and SVM, which comprises the following steps: in the SiC-GTO device test structure, a 3 rd P region is taken as a functional pin to be led out and defined as a Base pole, a volt-ampere characteristic curve of any two ends of a device is measured under the condition of low pressure by adopting an embedded circuit design method, and the volt-ampere characteristic curve of the device obtained under different environments is taken as a monitoring object. And a new dynamic time warping improvement method is provided to optimize the search path, so that the problems of inaccurate path warping and artificial setting of the search width in the algorithm can be effectively avoided, and meanwhile, the calculation complexity is reduced and the calculation amount is reduced. And finally, classifying the minimum distortion sum serving as a sample characteristic based on the SVM to realize the health state monitoring of the high-voltage SiC-GTO device.
Description
Technical Field
The invention relates to a health state monitoring technology of a SiC-GTO power device, in particular to a state monitoring method of the SiC-GTO power device based on DTW (dynamic time Warping) and SVM (support vector machine).
Background
The SiC-GTO power device (Gate Turn-Off Thyristor) is a power electronic device developed based on third-generation novel wide-bandgap semiconductor SiC, the SiC material has the advantages of large forbidden band width, high critical breakdown electric field, high thermal conductivity and high drift speed, and has wide application prospect in power electronic systems in environments of high power, high temperature, high pressure, strong radiation and the like. At present, mature commercial SiC power devices are available abroad, and the SiC power devices are developed successfully and put into application at home. The reliability of the power device is one of the most important components affecting the reliability of the power electronic system, and the quality of the performance of the power device directly affects the overall reliability of the power electronic system, so that the reliability of the SiC-GTO device is very necessary to be researched. The blocking voltage of the SiC-GTO device can reach more than thousands of volts, and the blocking characteristic and the switching characteristic are important parameter indexes for measuring the health state of the device. The state monitoring method for researching the blocking characteristic and the dynamic switching characteristic of the SiC-GTO device realizes that the monitoring of the health state of the device is finished under the condition of low pressure by an embedded method, and has important significance in practical application.
The state monitoring of the SiC-GTO device mainly comprises the steps of collecting current and voltage signals at two ends of the device and obtaining characteristic parameters of device failure or defects. The common power device monitoring methods mainly comprise methods of wavelet analysis, Frechet distance diagnosis, approximate entropy theory, dynamic time warping and the like, and each method has the characteristics and the application range, wherein the dynamic time warping algorithm has high identification precision, is obvious in difference of time sequence differences, can clearly distinguish whether the device has defects and the type of the device, and is an ideal choice for monitoring the state of the SiC-GTO device.
The Dynamic Time Warping (DTW) algorithm is a nonlinear warping algorithm combining a dynamic time warping technique and a distance measure calculation technique, and is one of effective methods for monitoring the state of a power device. The dynamic time warping algorithm can match time sequences with different data lengths, and finds out the optimal path between two time sequence curves through specific constraint conditions, so that the distance between the two time sequences is shortest. The algorithm can solve the problem of inconsistent sequence data length, has good robustness, is insensitive to measurement noise, and has remarkable advantages in characteristic research of SiC-GTO devices.
Although the dynamic time warping algorithm has higher accuracy in identifying different time sequences, when a point on one time sequence and a point on another time sequence are mapped with each other with a longer interval time during a search path, the dynamic time warping algorithm is prone to cause ill-conditioned warping and has higher calculation complexity. In the improved dynamic time warping algorithm, an Itakura parallelogram window is too narrow for the search paths of the starting point and the end point of the path, which easily causes the inaccurate path warping of the starting point and the end point, and although the Sakoe-Chiba window can realize the equal-width search globally, the search width needs to be set according to the size of an actual matrix. Therefore, the current health state monitoring effect combined with the dynamic time warping algorithm cannot meet the current situation requirement.
Disclosure of Invention
The invention provides a state monitoring method of a SiC-GTO device based on DTW and SVM for overcoming the technical defects, and the method optimizes the constraint condition, can effectively avoid the problems of inaccurate path regulation and manual setting of search width, and simultaneously reduces the calculation complexity and the calculation amount.
The technical scheme of the invention is as follows:
the method for monitoring the state of the SiC-GTO device based on the DTW and the SVM comprises the following specific steps:
firstly, leading out a 3 rd P region as a functional pin in a device structure for monitoring the health of a SiC-GTO device, and defining the P region as a Base electrode (B);
secondly, under the conditions of high temperature, high humidity, high power and the like, an embedded circuit is adopted to design and measure volt-ampere characteristic curves of any two ends of the device after normal or degradation, such as an AK end, a GB end and the like;
thirdly, establishing a matching distance matrix D of two time sequences based on a dynamic time warping algorithmn×mAnd global constraint conditions of the dynamic time warping path;
fourthly, searching a path with the minimum accumulated distance between elements, and obtaining the sum gamma (i, j) of the minimum distortion degree;
fifthly, training the SVM classifier by taking the minimum distortion sum as the sample characteristic of the SiC-GTO device to obtain an optimal classification hyperplane;
and sixthly, classifying the samples of the SiC-GTO device to realize the monitoring of the state of the SiC-GTO device.
As shown in fig. 1, the specific method steps of the present invention are as follows:
step one, leading out a 3 rd P area as a functional pin in a device structure for monitoring the health of the SiC-GTO device, and defining the P area as a Base pole (B);
step two, adopting an embedded circuit design to measure volt-ampere characteristic curves of any two ends of the device after normal or degradation under the condition of low voltage under the conditions of high temperature, high humidity, high power and the like, such as an AK end, a GB end and the like;
step three, establishing a matching distance matrix D of two time sequences based on a dynamic time warping algorithmn×mAnd global constraint conditions of the dynamic time warping path;
step four, searching a path with the minimum accumulated distance between elements, and obtaining the sum gamma (i, j) of the minimum distortion degree;
step five, taking the minimum distortion sum as a sample characteristic of the SiC-GTO device to train a Support Vector Machine (SVM) classifier to obtain an optimal classification hyperplane;
and step six, classifying the samples of the SiC-GTO device to realize the monitoring of the state of the SiC-GTO device.
The invention is expanded on the basis of a dynamic time warping algorithm, optimizes the device monitoring structure and the parameter acquisition method of the SiC-GTO power device, and has obvious advantages in the health state monitoring aspect of the novel SiC-GTO power device based on the dynamic warping algorithm.
Further, the process of researching the forward blocking characteristic of the SiC-GTO device in the step one shows that the blocking characteristic of the device is mainly dependent on the second N +/P node. Therefore, in the health monitoring process of the SiC-GTO device, the 3 rd P region in the device structure can be led out as a functional pin and is defined as a Base pole (B), as shown in FIG. 2.
Further, the second step specifically comprises the following steps: under the condition of low voltage (<10V) by building an embedded circuit, gradually increasing the voltage value applied to any two ends (such as an AK end, a GB end and the like) of the device from 0V at an interval smaller than 0.1V, and synchronously acquiring the voltage and current values at the two ends of the device to obtain a volt-ampere characteristic curve; the device is tested under normal, high temperature, high pressure, high power and other test environments, and volt-ampere characteristic curves of the device in different states are obtained, as shown in fig. 3.
Further, the third step specifically comprises the following steps: defining two time sequences as R, T, respectively, and calculating the two sequencesMatching distance matrix D ofn×m(ii) a Projecting the sequences R and T into a two-dimensional rectangular plane, wherein R is the horizontal axis, T is the vertical axis, and the grid point (i, j) represents the element RiAnd tjThe intersection point of (a); obtaining a time curved path P consisting of a series of lattice points in a plane; the search constraint range of the dynamic time-warping path P is established as shown in fig. 4.
Further, the fourth step specifically comprises the following steps: the sum of distortion degrees of the two time series is defined as γ (i, j), γ (1,1) is made to be 0, and from the point (1,1), the sum is repeatedly recurred to γ (n, m) according to the search path P, which is the minimum sum of distortion degrees of the two time series.
Further, the step five specifically comprises the following steps: calculating the sum of the minimum distortion degrees of the SiC-GTO device under the conditions of normal, high temperature, high humidity, high power and the like according to the steps, taking the sum as a sample characteristic, and dividing the sample characteristic into a training sample and a test sample; and mapping the input vector into a high-dimensional feature vector space based on a support vector machine aiming at the training sample, and constructing an optimal classification hyperplane f (x) in the feature space to divide two different types of samples so as to enable the interval between the two types to be maximum.
Further, the sixth step specifically comprises the following steps: and classifying the test samples based on the optimal classification hyperplane to realize the monitoring of the state of the SiC-GTO device.
The invention has the beneficial effects that:
according to the method, a 3 rd P area in the structure of the SiC-GTO device is taken as a functional pin to be led out and defined as a B pole, an embedded circuit is adopted to obtain normal or degraded volt-ampere characteristic curves at any two ends of the device under the environments of high temperature, high pressure, high humidity, high power and the like, the global constraint condition of a regular path in the device is improved based on a dynamic time regular algorithm, the minimum distortion sum obtained by the algorithm is taken as a sample characteristic, part of samples in the device are taken as training samples to train an SVM classifier to obtain an optimal classification hyperplane, and a test sample is classified based on a trained support vector machine to realize the state monitoring of the SiC-GTO device.
The invention is expanded on the basis of a dynamic time warping algorithm, optimizes the device monitoring structure and the parameter acquisition method of the SiC-GTO power device, and has obvious advantages in the health state monitoring aspect of the novel SiC-GTO power device based on the dynamic warping algorithm.
Drawings
FIG. 1 is a schematic flow diagram of the present invention;
FIG. 2 is a schematic view of a device structure for monitoring the health status of a SiC-GTO device according to the present invention;
FIG. 3 is a schematic view of a voltage-current characteristic curve at the GB end of a SiC-GTO device monitored under different conditions according to the invention;
FIG. 4 is a diagram illustrating a dynamic time-warping path search constraint range according to the present invention.
Detailed Description
Preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings.
FIG. 1 is a flow chart of the method of the present invention, which includes the steps of:
in the process of researching the forward blocking characteristic of the SiC-GTO device in the first step, the blocking characteristic of the device is mainly determined by the second N +/P node. Therefore, in the health monitoring process of the SiC-GTO device, the 3 rd P region in the device structure can be led out as a functional pin and is defined as a Base pole (B), as shown in FIG. 2.
The second step specifically comprises the following steps: under the condition of low voltage (<10V) by building an embedded circuit, gradually increasing the voltage value applied to any two ends (such as an AK end, a GB end and the like) of the device from 0V at an interval smaller than 0.1V, and synchronously acquiring the voltage and current values at the two ends of the device to obtain a volt-ampere characteristic curve; the device is tested under normal, high temperature, high pressure, high power and other test environments, and volt-ampere characteristic curves of the device in different states are obtained, as shown in fig. 3.
In the third step, taking the GB end as an example, selecting the GB end volt-ampere characteristic curve of the SiC-GTO device under normal conditions to define it as R, and selecting the other volt-ampere characteristic curves to define T:
R＝[r1,r2,Λri,rn]
T＝[t1,t2,Λti,tn]
where n and m are the dimensions of the sequences R and T, respectively, and the distance δ (i, j) between the elements of the two sequences is defined as:
δ(i,j)＝(ri-tj)2
a matching distance matrix D of the two sequences can be establishedn×m：
Finding a path P from the matching distance matrix such that the cumulative distance between elements is minimized, i.e.
In order to avoid the ill-conditioned warping problem of the dynamic time warping algorithm and effectively reduce the computation complexity, a search constraint range of the dynamic time warping path P is established, as shown in fig. 4. The intersection points a, B of the two lines were found with 2 lines with slopes 1/2 and 2, and the start and end points, respectively. And makes a line y parallel to the matrix diagonal based on point A, BOn the upper partAnd yLower partThe range enclosed by the two straight lines and the matrix is the search area of the dynamic regularization algorithm, wherein
yOn the upper part＝(3mx+5mn-2n2-2m2)/(3n)
yLower part＝(3mx+5mn+2n2+2m2)/(3n)
The entire search interval is limited to the following regions:
in step four, the distortion sum of the two time series is defined as γ (i, j), γ (1,1) is made equal to 0, and from the point (1,1), the process is repeated until γ (n, m) according to the search path P, which is the minimum distortion sum of the two time series.
γ(i,j)＝δ(i,j)+min[γ(i-1,j),γ(i,j-1),γ(i-1,j-1)]
In the fifth step, the SVM is based on the mechanism that an optimal classification hyperplane meeting the classification requirement is found, so that the hyperplane can maximize blank areas on two sides of the hyperplane while ensuring the classification precision, and a training sample set (x) is giveni,yi) I ═ 1, 2., l, y ∈ { ± 1} the classification hyperplane is written as (w · x + b) ═ 0, with the following constraints:
introducing Lagrange function to solve to obtain a decision function of the support vector machine as follows:
wherein a is Lagrange multiplier, b is classification threshold, and K (x, x)i) Is a classification kernel function.
The sixth step specifically comprises the following steps: and classifying the test samples based on the optimal classification hyperplane to realize the monitoring of the state of the SiC-GTO device.
Through the steps, the state of the SiC-GTO device can be monitored.
Finally, it is noted that the above-mentioned preferred embodiments illustrate rather than limit the invention, and that, although the invention has been described in detail with reference to the above-mentioned preferred embodiments, it will be understood by those skilled in the art that various changes in form and detail may be made therein without departing from the scope of the invention as defined by the appended claims.
Claims (6)
1. The method for monitoring the state of the SiC-GTO device based on the DTW and the SVM is characterized by comprising the following steps of:
the method comprises the following steps: leading out a 3 rd P region as a functional pin in a device structure for monitoring the health of the SiC-GTO device, and defining the P region as a Base electrode (B);
step two, under different environments, adopting an embedded circuit design to measure volt-ampere characteristic curves of any two ends of a normal or degraded device under a low-voltage condition, wherein the any two ends comprise an AK end and a GB end; under the condition that the low voltage is less than 10V, gradually increasing the voltage values applied to the AK end and the GB end of the device at intervals less than 0.1V from 0V, and synchronously collecting the voltage values and the current values at the two ends of the device to obtain a volt-ampere characteristic curve; testing the device under different test environments of normal, high temperature, high pressure and high power respectively to obtain volt-ampere characteristic curves of the device under different states;
step three, establishing a matching distance matrix D of two time sequences based on a dynamic time warping algorithmn×mAnd global constraint conditions of the dynamic time warping path;
step four, searching a path with the minimum accumulated distance among elements, and obtaining the minimum distortion sum gamma (i, j), wherein i belongs to (0, n), and j belongs to (0, m);
step five, taking the minimum distortion sum as a sample characteristic of the SiC-GTO device to train a Support Vector Machine (SVM) classifier to obtain an optimal classification hyperplane;
and step six, classifying the samples of the SiC-GTO device to realize the monitoring of the state of the SiC-GTO device.
2. The state monitoring method for the SiC-GTO device based on the DTW and the SVM of claim 1, wherein: in step three, two time sequences are defined as R, T respectively, and a matching distance matrix D of the two time sequences is calculatedn×m(ii) a Projecting time series R and T into a two-dimensional rectangular plane, wherein R is a horizontal axis, T is a vertical axis, and a lattice point (i, j) represents an element RiAnd tjThe intersection point of (a); obtaining a time curved path P formed by a series of lattice points in a two-dimensional right-angle plane; building (2)The search constraint range of the vertical dynamic time warping path P.
3. The state monitoring method for the SiC-GTO device based on the DTW and the SVM as claimed in claim 2, wherein: in step four, the distortion sum of the two time series is defined as γ (i, j), γ (1,1) is made to be 0, and from the point (1,1), the minimum distortion sum of the two time series is obtained by repeating recursion to γ (n, m) according to the search path P, that is, γ (i, j) ═ δ (i, j) + min [ γ (i-1, j), γ (i, j-1), γ (i-1, j-1) ], i belongs to (0, n), and j belongs to (0, m).
4. The state monitoring method for the SiC-GTO device based on the DTW and the SVM according to claim 1 or 3, wherein: in the fifth step, the minimum distortion sum is used as a sample characteristic and divided into a training sample and a test sample; and mapping the input vector into a high-dimensional feature vector space based on a support vector machine aiming at the training sample, and constructing an optimal classification hyperplane f (x) in the feature space to divide two different types of samples so as to enable the interval between the two types to be maximum.
5. The state monitoring method for the SiC-GTO device based on the DTW and the SVM as claimed in claim 4, wherein: and step six, classifying the test samples based on the optimal classification hyperplane to realize the monitoring of the state of the SiC-GTO device.
6. The state monitoring method for the SiC-GTO device based on the DTW and the SVM as claimed in claim 4, wherein: in the fifth step, the SVM is based on the mechanism that an optimal classification hyperplane meeting the classification requirement is found, so that the hyperplane can maximize blank areas on two sides of the hyperplane while ensuring the classification precision, and a training sample set (x) is giveni,yi) I ═ 1, 2., l, y ∈ { ± 1} the classification hyperplane is written as (w · x + b) ═ 0, with the following constraints:
introducing Lagrange function to solve to obtain a decision function of the support vector machine as follows:
wherein α is Lagrange multiplier, b is classification threshold, and K (x, x)i) Is a classification kernel function.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810266737.6A CN108519547B (en) | 2018-03-28 | 2018-03-28 | SiC-GTO device state monitoring method based on DTW and SVM |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810266737.6A CN108519547B (en) | 2018-03-28 | 2018-03-28 | SiC-GTO device state monitoring method based on DTW and SVM |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108519547A CN108519547A (en) | 2018-09-11 |
| CN108519547B true CN108519547B (en) | 2020-04-24 |
Family
ID=63430736
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810266737.6A Expired - Fee Related CN108519547B (en) | 2018-03-28 | 2018-03-28 | SiC-GTO device state monitoring method based on DTW and SVM |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108519547B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113189468A (en) * | 2021-05-13 | 2021-07-30 | 中国工程物理研究院电子工程研究所 | Health state on-line monitoring circuit and system of power device |
| CN113533924B (en) * | 2021-08-13 | 2023-06-30 | 中国工程物理研究院电子工程研究所 | Double-sided electrode high-power device testing device |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN202393833U (en) * | 2011-12-30 | 2012-08-22 | 长江三峡能事达电气股份有限公司 | SiC resistance property detector |
| CN104596767A (en) * | 2015-01-13 | 2015-05-06 | 北京工业大学 | Method for diagnosing and predicating rolling bearing based on grey support vector machine |
| CN104935879A (en) * | 2014-03-19 | 2015-09-23 | 通用汽车环球科技运作有限责任公司 | Multi-View Human Detection Using Semi-Exhaustive Search |
| CN205140965U (en) * | 2015-09-18 | 2016-04-06 | 中国工程物理研究院电子工程研究所 | Two -way withstand voltage carborundum solid -state switch |
| CN105868770A (en) * | 2016-03-23 | 2016-08-17 | 国网山东省电力公司电力科学研究院 | High-voltage circuit breaker fault diagnosis method based on unsupervised learning model |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8294507B2 (en) * | 2009-05-08 | 2012-10-23 | Cree, Inc. | Wide bandgap bipolar turn-off thyristor having non-negative temperature coefficient and related control circuits |
-
2018
- 2018-03-28 CN CN201810266737.6A patent/CN108519547B/en not_active Expired - Fee Related
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN202393833U (en) * | 2011-12-30 | 2012-08-22 | 长江三峡能事达电气股份有限公司 | SiC resistance property detector |
| CN104935879A (en) * | 2014-03-19 | 2015-09-23 | 通用汽车环球科技运作有限责任公司 | Multi-View Human Detection Using Semi-Exhaustive Search |
| CN104596767A (en) * | 2015-01-13 | 2015-05-06 | 北京工业大学 | Method for diagnosing and predicating rolling bearing based on grey support vector machine |
| CN205140965U (en) * | 2015-09-18 | 2016-04-06 | 中国工程物理研究院电子工程研究所 | Two -way withstand voltage carborundum solid -state switch |
| CN105868770A (en) * | 2016-03-23 | 2016-08-17 | 国网山东省电力公司电力科学研究院 | High-voltage circuit breaker fault diagnosis method based on unsupervised learning model |
Non-Patent Citations (3)
| Title |
|---|
| 《SiC GTO晶闸管技术现状及发展》;王俊等;《大功率变流技术》;20161031;全文 * |
| 《Simulation Study on the Performance of SiC-GTO》;Muhamad Zahim Sujod等;《IEEE》;20070702;全文 * |
| 《Switching Simulation of Si-GTO, SiC-GTO and Power MOSFET》;Muhamad Zahim Sujod等;《IEEE》;20070416;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108519547A (en) | 2018-09-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109934423B (en) | Photovoltaic power station power prediction method and system based on grid-connected inverter operation data | |
| CN105548862B (en) | A kind of analog-circuit fault diagnosis method based on broad sense multi-kernel support vector machine | |
| CN108519547B (en) | SiC-GTO device state monitoring method based on DTW and SVM | |
| CN102930533B (en) | A kind of semi-supervised Hyperspectral imaging dimension reduction method based on improving K-mean cluster | |
| CN105300844A (en) | Hydrophobicity grade judging method of silicone rubber insulator | |
| CN105513039A (en) | Charged insulator string icing bridging degree intelligent image analysis method | |
| Van Brunt et al. | Reliability assessment of a large population of 3.3 kV, 45 A 4H-SIC MOSFETs | |
| CN108169643A (en) | A kind of method and system for cable local discharge pattern-recognition | |
| CN113358992B (en) | Method for acquiring critical parameters of avalanche diode | |
| Bian et al. | The effect of surface roughness on corona-generated electromagnetic interference for long-term operating conductors | |
| Su et al. | Parameter extraction of photovoltaic single-diode model using integrated current–voltage error criterion | |
| Demant et al. | Deep learning approach to inline quality rating and mapping of multi-crystalline Si-wafers | |
| CN105895548A (en) | Grid modulation generation current based method for extracting substrate doping concentration of metal-oxide-semiconductor field-effect transistor (MOSFET) | |
| CN112285519B (en) | Method for measuring series resistance and ideal factor in diode gating array | |
| CN103311374A (en) | Photoluminescence-based solar crystalline silicon wafer quality prediction and control method | |
| KR102677517B1 (en) | Method of testing semiconductor device | |
| CN107544008B (en) | Vehicle-mounted IGBT state monitoring method and device | |
| CN103970961A (en) | Temperature sensor-free LED junction temperature predicting and controlling method | |
| CN116776242A (en) | Vacuum breakdown type discrimination method based on deep learning | |
| US20230197534A1 (en) | Neural Network Based Prediction of Semiconductor Device Response | |
| CN114066005B (en) | CNN network-based silicon carbide diode breakdown voltage prediction method | |
| US20140142882A1 (en) | Methods and systems for prediction of fill factor in heterojunction solar cells through lifetime spectroscopy | |
| CN115389882B (en) | Corona discharge state evaluation method, corona discharge state evaluation device, computer device, and storage medium | |
| Many et al. | Relaxation effects in recombination velocity on germanium surfaces under transverse electrostatic fields | |
| CN117012859B (en) | Preparation method of semiconductor photosensitive high-voltage diode |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20200424 |