Quantitative characterization of carbon/carbon composites matrix texture based on image analysis using polarized...

Area of Science:

• Materials Science
• Composite Materials
• Microscopy

Background:

• The mechanical and physical properties of carbon/carbon (C/C) composites are critically dependent on their matrix texture.
• Accurate characterization of this texture is essential for predicting and optimizing composite performance.
• Existing methods may lack the precision or adaptability for complex microstructures.

Purpose of the Study:

• To develop a quantitative method for characterizing the matrix texture of C/C composites.
• To establish a relationship between optical properties (extinction angle) and microstructural texture.
• To enable automated analysis of C/C composite microstructures from polarized light microscope (PLM) images.

Main Methods:

• Utilized cloud theory for uncertain reasoning and quantitative-qualitative transformation.
• Developed cloud models to describe the relationship between extinction angle and texture types.
• Employed linguistic controllers for matrix texture analysis based on PLM image features.
• Implemented image segmentation techniques to differentiate components and textures within the matrix.

Main Results:

• Successfully calculated extinction angles from PLM images of C/C composites.
• Achieved a low error rate (1-2°) between calculated and measured extinction angles.
• Segmented PLM images by component and further resolved mixed textures within the matrix.
• Demonstrated the feasibility of quantitative characterization of C/C composite matrix from a single PLM image.

Conclusions:

• A novel quantitative characteristic method for C/C composite matrix texture has been successfully developed.
• The integration of cloud theory and PLM image analysis provides an accurate and efficient approach.
• This method allows for detailed microstructural characterization, paving the way for improved material design and performance prediction.