Supervised artificial intelligence classification of clinker processing conditions by optical microscopy

Area of Science:

• Materials Science
• Computer Science
• Industrial Chemistry

Background:

• Optical microscopy is crucial for characterizing industrial clinkers, providing insights into production.
• Current limitations include time-consuming sample preparation, image interpretation, and operator subjectivity.
• Automating clinker analysis is highly desirable for large-scale applications.

Purpose of the Study:

• To explore supervised artificial intelligence (AI) methodologies for clinker micrographs.
• To assess AI's capability to assist human operators in result interpretation.
• To evaluate AI's ability to autonomously determine experimental parameters.

Main Methods:

• Application of supervised artificial intelligence (AI) models to clinker micrographs.
• Training AI models to interpret experimental results and identify parameters.
• Comparative analysis of AI-driven interpretation versus human operator analysis.

Main Results:

• AI models demonstrate a promising capability in analyzing clinker micrographs.
• The developed AI model can autonomously determine experimental parameters.
• AI shows potential to reduce subjectivity and increase efficiency in clinker analysis.

Conclusions:

• Supervised AI methodologies show significant potential for automating clinker micrograph analysis.
• AI can serve as a valuable tool to support human operators in industrial clinker characterization.
• Further development of AI in this field can lead to more objective and efficient production processes.