Corrosion Prediction with Parallel Finite Element Modeling for Coupled Hygro-Chemo Transport into Concrete under...

Area of Science:

• Civil Engineering
• Materials Science
• Computational Mechanics

Background:

• Chloride-induced corrosion significantly impacts concrete structure durability.
• Accurate prediction requires complex models and fine mesh simulations for realistic performance.
• Existing models often lack detailed simulation of surface layers and microclimate variations.

Purpose of the Study:

• To propose a more realistic physical model for coupled hygro-chemical transport in concrete.
• To implement this model using a parallel finite element algorithm.
• To incorporate microclimate effects, including humidity and temperature, into the prediction.

Main Methods:

• Development of a coupled hygro-chemical transport model.
• Implementation using a parallel finite element algorithm for computational efficiency.
• Integration of a microclimate model considering environmental humidity and seasonal temperature.
• Validation of the model on an existing bridge under multi-boundary conditions.

Main Results:

• A prediction model for chloride diffusion in unsaturated concrete was successfully developed.
• The model's computational time decreased with increased processors up to an optimal point.
• Beyond the optimum, computational time increased due to inter-processor communication overhead.
• Model validation on an existing bridge confirmed its applicability under complex conditions.

Conclusions:

• The developed model provides a more realistic prediction of chloride-induced corrosion initiation.
• Parallel computing significantly enhances the efficiency of complex durability simulations.
• The framework is extensible for simulating multi-species ingress, such as de-icing salts.