Data Mining to Atmospheric Corrosion Process Based on Evidence Fusion.

Area of Science:

• Materials Science
• Environmental Science
• Data Science

Background:

• Atmospheric corrosion of carbon steel is a significant issue in outdoor environments.
• Real-time monitoring of corrosion dynamics is crucial for understanding degradation mechanisms.
• Existing statistical methods often struggle with complex, online coupled corrosion data.

Purpose of the Study:

• To develop and validate a novel machine learning model for analyzing atmospheric corrosion data.
• To quantify the influence of various environmental factors on carbon steel corrosion over time.
• To compare the proposed model's performance against established machine learning techniques.

Main Methods:

• Utilized an electrical resistance sensor-based atmospheric corrosion monitor for real-time data acquisition.
• Applied data mining techniques to explore underlying corrosion mechanisms.
• Developed and implemented an information fusion-based machine learning model.
• Compared the new model with artificial neural networks and support vector machines.

Main Results:

• The proposed machine learning model demonstrated superior performance in analyzing online coupled corrosion data.
• The model accurately quantified the contribution of different environmental factors to corrosion.
• Experimental results showed higher corrosion prediction accuracy compared to traditional models.
• The model effectively identified the importance of atmospheric factors influencing corrosion.

Conclusions:

• The novel information fusion-based machine learning model offers a significant advancement in atmospheric corrosion monitoring and analysis.
• This approach provides a more accurate and insightful understanding of carbon steel degradation in outdoor environments.
• The model's ability to quantify environmental factor contributions and predict corrosion holds great promise for materials protection strategies.