CN-121114712-B - Multi-environment aging detection and evaluation method and system for electric device

CN121114712BCN 121114712 BCN121114712 BCN 121114712BCN-121114712-B

Abstract

The invention relates to the technical field of power electronics, in particular to a method and a system for detecting and evaluating multiple environment ageing of an electric device. A multi-environment aging detection and evaluation method for an electrical device comprises the following steps of S1, collecting environment factor data and personnel activity data in various application scenes of a gallium nitride power device, generating a test environment stress sequence according to the environment factor data and the personnel activity data, S2, testing the gallium nitride power device according to the test environment stress sequence, monitoring aging parameters of the gallium nitride power device in real time, S3, generating an aging characteristic vector based on the aging parameters, and judging an aging mechanism and a confidence coefficient according to the aging characteristic vector and an aging mechanism judging model. According to the invention, the aging mechanism of the device in different application scenes is judged by simulating the environmental factors and personnel activity data of different application scenes, so that decision support is provided for technicians engaged in device design and maintenance.

Inventors

GAO ZHIYUAN
WANG XIUYU
GAO JUNGUANG
ZHANG ZELI
ZHANG YAN

Assignees

天津海芯微电子技术有限公司
中科光析(山东)检测科技有限公司

Dates

Publication Date: 20260505
Application Date: 20251017

Claims (9)

1. A method for environmental aging detection and assessment of an electrical device, comprising the steps of: The method comprises the steps of S1, collecting environmental factor data and personnel activity data in various application scenes of a gallium nitride power device, generating a test environmental stress sequence according to the environmental factor data and the personnel activity data, respectively collecting the environmental factor data and the personnel activity data of the gallium nitride power device in different application scenes, wherein the environmental factor data comprises scene influence factors, temperature, voltage, current, humidity and vibration, the personnel activity data comprises personnel reference electrostatic voltage, personnel comprehensive activity intensity, activity frequency and clothing materials, analyzing time sequence changes of the environmental factor data to obtain a real environmental stress sequence, calculating to obtain electrostatic discharge voltage based on the environmental factor data and the personnel activity data, generating a random electrostatic discharge event sequence through poisson distribution according to the electrostatic discharge voltage and the activity frequency, and integrating the real environmental stress sequence and the random electrostatic discharge event sequence in time and space to generate the test environmental stress sequence; S2, testing the gallium nitride power device according to the test environment stress sequence, and monitoring ageing parameters of the gallium nitride power device in real time; S3, generating an aging characteristic vector based on the aging parameter, and judging an aging mechanism and a confidence level according to the aging characteristic vector and an aging mechanism judging model; and S4, recording the test environment stress sequence, the aging characteristic vector, the aging mechanism and the confidence coefficient, and generating an aging detection evaluation report.
2. The method for detecting and evaluating environmental aging of an electrical device according to claim 1, wherein the step of calculating an electrostatic discharge voltage based on the environmental factor data and the personnel activity data comprises calculating an electrostatic discharge voltage from the scene impact factor, humidity, personnel reference electrostatic voltage, personnel integrated activity intensity, and clothing material by an electrostatic discharge voltage calculation formula, wherein the electrostatic discharge voltage calculation formula is: ; In the formula, In the case of an electrostatic discharge voltage, For a human to reference the electrostatic voltage, For the comprehensive activity intensity of the personnel, Is a coefficient of the material of the clothing, In order to be a degree of humidity, For the purpose of adjusting the constant for the humidity, Is a scene impact factor.
3. The method for detecting and evaluating the environmental aging of the electrical device according to claim 1, wherein the testing the gallium nitride power device according to the testing environmental stress sequence and monitoring the aging parameters of the gallium nitride power device in real time comprises the steps of carrying out multi-factor aging experiments on the gallium nitride power device under the action of the testing environmental stress sequence, and acquiring the aging parameter changes of the device in real time to obtain an aging parameter change sequence, wherein the aging parameter change sequence comprises a threshold voltage change sequence, a dynamic on-resistance change sequence, a static on-resistance change sequence, a thermal resistance change sequence, a leakage current change sequence and a switching time length change sequence.
4. The method for detecting and evaluating the environmental aging of the electric device according to claim 3, wherein the step of generating the aging characteristic vector based on the aging parameter and judging the aging mechanism and the confidence level according to the aging characteristic vector and the aging mechanism judging model comprises the steps of carrying out time series smoothing pretreatment on the acquired aging parameter change sequence, adopting wavelet transformation to eliminate measurement errors, calculating a recoverable aging coefficient according to the dynamic on-resistance change sequence, normalizing the aging parameter change sequence, and then calculating a static on-resistance change rate, a threshold voltage drift rate, a leakage current change rate, a thermal resistance change rate and a switching time delay change amount to form the aging characteristic vector.
5. The method for detecting and evaluating the environmental aging of the electrical device according to claim 4, wherein the calculating the recoverable aging coefficient according to the dynamic on-resistance variation sequence comprises dividing the dynamic on-resistance variation sequence into an aging process sequence segment and a recovery process sequence segment by using a moving window statistical method, obtaining a dynamic on-resistance initial value and a stress-acting dynamic on-resistance stable value from the aging process sequence segment, obtaining a dynamic on-resistance stable value after recovery within a preset recovery time window from the recovery process sequence segment, obtaining a gallium nitride power device trap characteristic recovery rate by using a nonlinear least square fitting method according to the recovery process sequence segment, and calculating a recoverable aging coefficient based on a recoverable aging coefficient calculation formula, wherein the recoverable aging coefficient calculation formula is as follows: ; In the formula, In order to be able to restore the ageing coefficient, The dynamic on-resistance is stabilized for stress, To stabilize the value of the dynamic on-resistance after recovery, As the initial value of the dynamic on-resistance, For the gallium nitride power device trap feature recovery rate, Is a non-zero constant.
6. The method for detecting and evaluating the environmental aging of the electrical device according to claim 1, wherein the step of judging the aging mechanism and the confidence according to the aging characteristic vector and the aging mechanism judging model comprises the steps of inputting the aging characteristic vector into the aging mechanism judging model for analysis to obtain an aging mechanism judging result and the confidence, constructing the aging mechanism judging model based on SVM probability likelihood and Bayesian fusion, and training the aging mechanism judging model through an aging mechanism knowledge base, wherein the aging mechanism knowledge base presets various typical aging modes including gate oxide breakdown, interface trap accumulation, welding spot fatigue, hot carrier effect, electrostatic damage and corresponding aging characteristic vectors.
7. The method for detecting and evaluating the environmental aging of the electrical device according to claim 6, wherein the method for constructing the aging mechanism judging model based on the combination of SVM probability likelihood and Bayesian fusion is characterized by comprising the steps of respectively extracting aging characteristic vectors for a plurality of typical aging modes preset in a knowledge base, training an SVM classifying model by adopting a Gaussian radial basis function, outputting a judging result and a decision function value of each sample by the SVM classifying model, carrying out probability mapping on the decision function value output by the SVM classifying model by adopting a Sigmoid function to obtain posterior probability of each mechanism, calculating the posterior probability of each mechanism according to Bayesian reasoning, selecting the mechanism with the maximum posterior probability as a main aging mechanism of a current sample, calculating a normalized difference value between a maximum posterior probability value and a second maximum probability value, combining probability distribution entropy values of all competing mechanisms, and obtaining the confidence of the judging result by weighting summation.
8. The method for detecting and evaluating the multiple environment aging of the electric device according to claim 1, wherein the recording of the test environment stress sequence, the aging characteristic vector, the aging mechanism and the confidence coefficient generates an aging detection evaluation report, and the method comprises the steps of uniformly storing and marking the collected test environment stress sequence, the aging characteristic vector, the aging mechanism judging result and the confidence coefficient with a time stamp, carrying out fusion analysis on multiple test results of the same application scene, drawing a change curve showing the evolution trend of the aging mechanism according to the aging mechanism judging result and the confidence coefficient, and generating the aging detection evaluation report of different application scenes.
9. A multiple environmental aging test evaluation system for an electrical device for implementing a multiple environmental aging test evaluation method for an electrical device according to any one of claims 1 to 8, comprising: The data acquisition module is used for acquiring environmental factor data and personnel activity data in various application scenes of the gallium nitride power device and generating a test environmental stress sequence according to the environmental factor data and the personnel activity data; the real-time monitoring module is used for testing the gallium nitride power device according to the test environment stress sequence and monitoring the aging parameters of the gallium nitride power device in real time; The mechanism judging module is used for generating an aging characteristic vector based on the aging parameters and judging an aging mechanism and a confidence level according to the aging characteristic vector and an aging mechanism knowledge base of the gallium nitride power device; And the report generation module is used for recording the test environment stress sequence, the aging characteristic vector, the aging mechanism and the confidence coefficient and generating an aging detection evaluation report.

Description

Multi-environment aging detection and evaluation method and system for electric device Technical Field The invention relates to the technical field of power electronics, in particular to a method and a system for detecting and evaluating multiple environment ageing of an electric device. Background Under the background of rapid development of the current electric and electronic technology, gallium nitride power devices are used as high-performance semiconductors and are widely applied to the fields of electric automobiles, communication base stations and industrial power supplies, however, in the complex and changeable actual application scenes, accelerated aging phenomena, such as threshold voltage drift and on-resistance increase, often occur due to the comprehensive influence of environmental factors and personnel activities. As gallium nitride devices develop to higher power densities and higher frequencies, the aging process often involves multiple mechanisms for interleaving, and it is difficult for conventional methods to fully capture dynamic stresses in the real environment, resulting in increased life prediction bias and risk of unexpected failure. The prior art mainly comprises an accelerated aging test based on a single environmental factor, such as a thermal aging test performed through a constant temperature and humidity box, or a device performance change tracking method by utilizing an electrical parameter monitoring device, wherein the methods can partially simulate an aging process, but are limited to a laboratory fixed condition, and cannot effectively integrate the influence of an electrostatic discharge random event generated by the movable property of personnel, so that a multi-environmental aging detection evaluation method and system for an electrical device are needed, environmental factors and personnel activity data of different application scenes of a gallium nitride power device can be simulated, the aging mechanism of the device under different application scenes is judged, and decision support is provided for technicians engaged in device design and maintenance. Disclosure of Invention In order to overcome the defect that the aging mechanism of different application scenes of a gallium nitride power device is difficult to judge in the prior art, the invention provides a multi-environment aging detection and evaluation method and system for an electric device. The technical embodiment of the invention is that the method for detecting and evaluating the environmental aging of the electric device comprises the following steps: S1, collecting environmental factor data and personnel activity data in various application scenes of a gallium nitride power device, and generating a test environmental stress sequence according to the environmental factor data and the personnel activity data; S2, testing the gallium nitride power device according to the test environment stress sequence, and monitoring ageing parameters of the gallium nitride power device in real time; S3, generating an aging characteristic vector based on the aging parameter, and judging an aging mechanism and a confidence level according to the aging characteristic vector and an aging mechanism judging model; and S4, recording the test environment stress sequence, the aging characteristic vector, the aging mechanism and the confidence coefficient, and generating an aging detection evaluation report. The method comprises the steps of respectively collecting environmental factor data and personnel activity data of a gallium nitride power device in different application scenes, wherein the environmental factor data comprises scene influence factors, temperature, voltage, current, humidity and vibration, the personnel activity data comprises personnel reference electrostatic voltage, personnel comprehensive activity intensity, activity frequency and clothing materials, analyzing time sequence changes of the environmental factor data to obtain a real environmental stress sequence, calculating to obtain electrostatic discharge voltage based on the environmental factor data and the personnel activity data, generating a random electrostatic discharge event sequence through poisson distribution according to the electrostatic discharge voltage and the activity frequency, and integrating the real environmental stress sequence and the electrostatic discharge sequence in a space-time manner to generate the test environmental stress sequence. Preferably, the electrostatic discharge voltage is calculated and obtained based on the environmental factor data and the personnel activity data, and comprises the steps of calculating and obtaining the electrostatic discharge voltage according to the scene influence factor, the humidity, the personnel reference electrostatic voltage, the personnel comprehensive activity intensity and the clothing material through an electrostatic discharge voltage calculation formula, wherein the electrostatic