Failure analysis method and system for electrical property of gallium nitride integrated circuit

Abstract

The invention discloses a failure analysis method and system for the electrical property of a gallium nitride integrated circuit, and the method comprises the following steps: carrying out the preliminary electrical property test of the gallium nitride integrated circuit, and collecting a first group of electrical property test data; secondly, preprocessing the data to improve the accuracy of subsequent analysis; analyzing the preprocessed first group of electrical test data by using a trained artificial intelligence algorithm, and identifying potential failure modes or abnormal features; performing infrared imaging on the gallium nitride integrated circuit to obtain a heat distribution image of the gallium nitride integrated circuit, and combining layout and circuit design information of the integrated circuit by analyzing hot spots or abnormal temperature areas in the heat distribution image; the method has the beneficial effects that artificial intelligence and an infrared imaging technology are applied to failure analysis of the gallium nitride integrated circuit for the first time, and technical fusion and innovation are realized; through intelligent analysis of an artificial intelligence algorithm and visual display of an infrared imaging technology, the efficiency and accuracy of failure analysis are remarkably improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of integrated circuit failure analysis, and in particular relates to a method and system for analyzing the electrical properties of a gallium nitride integrated circuit. Background Art

[0002] Gallium nitride is a new type of compound semiconductor material with advantages such as high electron mobility, high voltage resistance and high frequency characteristics. Compared with traditional silicon and gallium arsenide, gallium nitride has higher electron mobility and larger energy band width, which gives it significant advantages in achieving high-speed transmission and high power density.

[0003] Gallium nitride's high electron mobility and high-frequency characteristics make it an ideal material for microwave and radio frequency applications, and it can be used to make high-performance microwave integrated circuits and radio frequency integrated circuits; Gallium nitride's high voltage resistance gives it significant advantages in power electronics applications, and it can be used to make high-efficiency, high-power density power management integrated circuits and power amplifiers, etc.; Gallium nitride integrated circuits can also be used in hybrid integrated circuit fields such as optical communications, biomedicine, and sensors to achieve multi-functional or composite functions to meet the needs of different fields; Gallium nitride's high electron mobility and energy band width give it higher switching speeds and power density, thereby improving the performance of integrated circuits; Gallium nitride materials have excellent physical and chemical stability, which makes Gallium nitride integrated circuits have high reliability; with the continuous development and improvement of Gallium nitride technology, Gallium nitride integrated circuits will play an increasingly important role in microwave, radio frequency, power electronics and other fields.

[0004] The patent discloses a failure analysis method and system for integrated circuits with publication number CN109725246A, which includes the following steps: performing an electrical test on the integrated circuit and obtaining first electrical test data; performing charge trap elimination processing on the integrated circuit; performing an electrical test on the integrated circuit and obtaining second electrical test data; if the difference between the first electrical test data and the second electrical test data exceeds a preset difference, it is determined that the integrated circuit has failed due to charge traps, but this method has limitations in identifying specific failure locations and deep-seated failure mechanisms. Summary of the invention

[0005] The object of the present invention is to provide a method and system for electrical failure analysis of gallium nitride integrated circuits to achieve more efficient, accurate and intelligent failure analysis.

[0006] To achieve the above object, the present invention provides the following technical solution: a method for analyzing the electrical properties of a gallium nitride integrated circuit, comprising the following steps:

[0007] Conduct preliminary electrical testing on GaN integrated circuits and collect the first set of electrical test data; then, pre-process the data to improve the accuracy of subsequent analysis;

[0008] Use the trained artificial intelligence algorithm to analyze the first set of pre-processed electrical test data to identify potential failure modes or abnormal characteristics;

[0009] Perform infrared imaging on GaN integrated circuits to obtain their thermal distribution images. By analyzing the hot spots or abnormal temperature areas in the thermal distribution images and combining the layout and circuit design information of the integrated circuits, the possible failure locations can be preliminarily located.

[0010] Combine the results of artificial intelligence failure analysis with the results of infrared imaging detection to conduct comprehensive analysis and diagnosis, and further confirm the cause and specific location of the failure through cross-validation and comparative analysis;

[0011] Generate a detailed failure analysis report based on the comprehensive analysis results.

[0012] As a preferred technical solution of the present invention, a preliminary electrical test is performed on the gallium nitride integrated circuit, and a first set of electrical test data is collected by using professional electrical test equipment or instruments.

[0013] As a preferred technical solution of the present invention, analyzing the first set of preprocessed electrical test data using a trained artificial intelligence algorithm is to input the preprocessed data into a trained model, which is built based on a machine learning or deep learning algorithm and can identify potential failure modes or abnormal characteristics based on the input electrical test data.

[0014] As a preferred technical solution of the present invention, hot spots or abnormal temperature areas in the thermal distribution image are analyzed using infrared imaging technology to capture the thermal radiation of the integrated circuit when it is working; hot spots or abnormal temperature areas in the image are identified through image processing and analysis technology.

[0015] As a preferred technical solution of the present invention, the failure cause and specific location are further confirmed through cross-validation and comparative analysis, and the implementation method is as follows: the results of artificial intelligence failure analysis are combined with the results of infrared imaging detection, and comprehensive analysis and diagnosis are performed, including comparing the output results of the two methods to see whether they are consistent or support each other.

[0016] As a preferred technical solution of the present invention, physical inspection and X-ray analysis can be combined to further verify and confirm the cause and specific location of the failure.

[0017] As a preferred technical solution of the present invention, a detailed failure analysis report is generated, including failure cause, location, and impact degree information.

[0018] As a preferred technical solution of the present invention, data is preprocessed, including denoising and normalization.

[0019] The present invention also discloses a failure analysis system for electrical properties of gallium nitride integrated circuits, comprising:

[0020] Data acquisition module: used to perform preliminary electrical tests on GaN integrated circuits and collect the first set of electrical test data;

[0021] Artificial Intelligence Analysis Module: Built-in trained artificial intelligence algorithm to analyze pre-processed electrical test data and identify potential failure modes or abnormal characteristics;

[0022] Infrared imaging module: used to perform infrared imaging on GaN integrated circuits to obtain their thermal distribution images; the module also includes an image processing algorithm for preprocessing and analyzing thermal distribution images;

[0023] Comprehensive analysis module: Combines the results of the artificial intelligence analysis module and the infrared imaging module to conduct comprehensive analysis and diagnosis, and confirms the cause and specific location of the failure through cross-validation and comparative analysis;

[0024] Report generation module: Generate detailed failure analysis report based on comprehensive analysis results.

[0025] Compared with the prior art, the present invention has the following beneficial effects:

[0026] For the first time, artificial intelligence and infrared imaging technology were applied to failure analysis of GaN integrated circuits, achieving technological integration and innovation;

[0027] Through intelligent analysis of artificial intelligence algorithms and intuitive display of infrared imaging technology, the efficiency and accuracy of failure analysis have been significantly improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 is a flow chart of the failure analysis method of the present invention;

[0029] Figure 2 This is a schematic diagram of the failure analysis system of the present invention. DETAILED DESCRIPTION

[0030] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0031] Example 1

[0032] See also Figure 1 , which is the first embodiment of the present invention, provides a method for analyzing electrical failure of a gallium nitride integrated circuit, comprising the following steps:

[0033] Conduct preliminary electrical tests on GaN integrated circuits, collect the first set of electrical test data, and pre-process the data, including denoising and normalization, to improve the accuracy of subsequent analysis;

[0034] Use the trained artificial intelligence algorithm to analyze the first set of pre-processed electrical test data to identify potential failure modes or abnormal characteristics;

[0035] Perform infrared imaging on GaN integrated circuits to obtain their thermal distribution images. By analyzing the hot spots or abnormal temperature areas in the thermal distribution images and combining the layout and circuit design information of the integrated circuits, the possible failure locations can be preliminarily located.

[0036] Combine the results of artificial intelligence failure analysis with the results of infrared imaging detection to conduct comprehensive analysis and diagnosis, and further confirm the cause and specific location of the failure through cross-validation and comparative analysis;

[0037] Generate a detailed failure analysis report based on the comprehensive analysis results.

[0038] In this embodiment, preferably, a preliminary electrical test is performed on the gallium nitride integrated circuit, and the first set of electrical test data is collected by using professional electrical testing equipment or instruments. These equipment can measure the electrical parameters such as voltage, current, resistance, capacitance, etc. of the integrated circuit under various conditions, and record these data to form the first set of electrical test data. The test conditions include different input voltages, frequencies, temperatures, etc., to comprehensively evaluate the electrical performance of the integrated circuit.

[0039] In this embodiment, preferably, the preprocessed first set of electrical test data is analyzed using a trained artificial intelligence algorithm by inputting the preprocessed data into a trained model, which is built based on a machine learning or deep learning algorithm and can identify potential failure modes or abnormal characteristics based on the input electrical test data. The model compares the difference between the input data and normal or expected data patterns, thereby identifying any unexpected behavior or characteristics.

[0040] In this embodiment, preferably, the hot spots or abnormal temperature areas in the thermal distribution image are analyzed using infrared imaging technology to capture the thermal radiation of the integrated circuit when it is working. Through image processing and analysis technology, the hot spots or abnormal temperature areas in the image are identified. These hot spots indicate overheating of certain components or areas in the integrated circuit due to excessive resistance, current leakage or other failure modes. By combining these hot spots with the layout and circuit design information of the integrated circuit, the possible failure locations can be preliminarily located.

[0041] In this embodiment, preferably, the failure cause and specific location are further confirmed through cross-validation and comparative analysis, and the implementation method is as follows: This includes comparing the output results of the two methods to see whether they are consistent or support each other; if there are any inconsistencies, further investigation and analysis are required to determine which method is more reliable or more accurate; in addition, other testing and analysis methods (such as physical inspection, X-ray analysis, etc.) can be combined to further verify and confirm the failure cause and specific location; this cross-validation and comparative analysis process helps to improve the accuracy and reliability of failure analysis.

[0042] In this embodiment, preferably, a detailed failure analysis report is generated, including failure cause, location, and impact degree information.

[0043] Example 2

[0044] See also Figure 2 , which is a second embodiment of the present invention, provides a failure analysis system for electrical properties of gallium nitride integrated circuits, comprising:

[0045] Data acquisition module: used to perform preliminary electrical tests on GaN integrated circuits and collect the first set of electrical test data;

[0046] Artificial Intelligence Analysis Module: Built-in trained artificial intelligence algorithm to analyze pre-processed electrical test data and identify potential failure modes or abnormal characteristics;

[0047] Infrared imaging module: used to perform infrared imaging on GaN integrated circuits to obtain their thermal distribution images; the module also includes an image processing algorithm for preprocessing and analyzing thermal distribution images;

[0048] Comprehensive analysis module: Combines the results of the artificial intelligence analysis module and the infrared imaging module to conduct comprehensive analysis and diagnosis, and confirms the cause and specific location of the failure through cross-validation and comparative analysis;

[0049] Report generation module: Generates a detailed failure analysis report based on the comprehensive analysis results, including failure cause, location, impact degree and other information.

[0050] Although the embodiments of the present invention have been shown and described, as detailed above, it will be understood by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for electrical failure analysis of gallium nitride integrated circuits, characterized in that: The steps include: Conduct preliminary electrical tests on GaN integrated circuits and collect the first set of electrical test data; then, pre-process the data to improve the accuracy of subsequent analysis; Use the trained artificial intelligence algorithm to analyze the first set of pre-processed electrical test data to identify potential failure modes or abnormal characteristics; Perform infrared imaging on GaN integrated circuits to obtain their thermal distribution images. By analyzing the hot spots or abnormal temperature areas in the thermal distribution images and combining the layout and circuit design information of the integrated circuits, the possible failure locations can be preliminarily located. Combine the results of artificial intelligence failure analysis with the results of infrared imaging detection to conduct comprehensive analysis and diagnosis, and further confirm the cause and specific location of the failure through cross-validation and comparative analysis; Generate a detailed failure analysis report based on the comprehensive analysis results.

2. The method for electrical failure analysis of gallium nitride integrated circuits according to claim 1, characterized in that: Perform preliminary electrical testing on GaN integrated circuits and collect the first set of electrical test data using professional electrical testing equipment or instruments.

3. The method for electrical failure analysis of gallium nitride integrated circuits according to claim 1, characterized in that: Analyzing the first set of preprocessed electrical test data using a trained artificial intelligence algorithm is to input the preprocessed data into a trained model, which is built based on a machine learning or deep learning algorithm and can identify potential failure modes or abnormal features based on the input electrical test data.

4. The method for electrical failure analysis of gallium nitride integrated circuits according to claim 1, characterized in that: Analyze hot spots or abnormal temperature areas in thermal distribution images. Use infrared imaging technology to capture the thermal radiation of integrated circuits when they are working; through image processing and analysis technology, identify hot spots or abnormal temperature areas in the image.

5. The method for electrical failure analysis of gallium nitride integrated circuits according to claim 1, characterized in that: The cause and specific location of the failure were further confirmed through cross-validation and comparative analysis, which was achieved by combining the results of the AI ​​failure analysis with the results of the infrared imaging detection, and performing a comprehensive analysis and diagnosis, including comparing the output results of the two methods to see if they are consistent or supportive of each other.

6. The method for electrical failure analysis of gallium nitride integrated circuits according to claim 5, characterized in that: Physical inspection and X-ray analysis can also be combined to further verify and confirm the cause and specific location of the failure.

7. The method for electrical failure analysis of gallium nitride integrated circuits according to claim 1, characterized in that: Generate detailed failure analysis reports, including failure cause, location, and impact information.

8. The method for electrical failure analysis of gallium nitride integrated circuits according to claim 1, characterized in that: Preprocess the data, including denoising and normalization.

9. A failure analysis system for electrical properties of gallium nitride integrated circuits, characterized in that: include Data acquisition module: used to perform preliminary electrical tests on GaN integrated circuits and collect the first set of electrical test data; Artificial Intelligence Analysis Module: Built-in trained artificial intelligence algorithm to analyze pre-processed electrical test data and identify potential failure modes or abnormal characteristics; Infrared imaging module: used to perform infrared imaging on GaN integrated circuits to obtain their thermal distribution images; the module also includes an image processing algorithm for preprocessing and analyzing thermal distribution images; Comprehensive analysis module: Combines the results of the artificial intelligence analysis module and the infrared imaging module to conduct comprehensive analysis and diagnosis, and confirms the cause and specific location of the failure through cross-validation and comparative analysis; Report generation module: Generate detailed failure analysis report based on comprehensive analysis results.

Images