Hybrid explainable machine learning models for predicting rapid chloride penetration test and sorptivity of self-compacting concrete with fly ash and silica fume under thermal exposure

Key Points

• Predictive models suggest significant correlation between thermal exposure and chloride penetration in concrete.
• The evaluation showed that self-compacting concrete with fly ash and silica fume had varied sorptivity values, reaching a maximum of 50%.
• Analysis involved the application of hybrid explainable machine learning models to assess concrete properties under stress.
• Understanding these properties is crucial for developing more resilient concrete mixes, especially for construction in extreme conditions.