Predicting Gas Permeability in Cementitious Materials Using Artificial Neural Networks
Predicting Gas Permeability in Cementitious Materials Using Artificial Neural Networks
The composition of cementitious materials has a direct influence on the microstructures on the one hand, and the transfer properties on the other, which ensure the durability of these materials.
According to the literature, very few models exist for estimating these indicators from the composition of a cementitious material. The aim of this work is to indicate gas permeability, using an artificial intelligence approach (neural networks) to predict permeability from the microstructural characteristics of a cementitious material, and to study the link between the composition of this cementitious material and its gas permeability. This model enabled us to predict gas permeability on the basis of the following parameters: water/cement ratio (W/C), volume percentage of cement paste in the material. In the light of the results obtained, which demonstrate the effectiveness of this model, we compared the results of the proposed model for estimating permeability with the results of experimental permeabilities. This comparison is established between model predictions and experimental permeability data, paving the way for significant advances in the durability of cementitious materials.
Abdelkader Kheira; Sidi Mohammed Aissa Mamoune; Houmadi Youcef
18th African Regional Conference on Soil Mechanics and Geotechnical Engineering (ARCSMGE2024)
Behavior of soils, analysis and modeling
Contact Information
- Secretary General - Dr A M McNamara
- Email: [email protected]
- Phone: +44 20 7040 8154
- City, University of London
- Northampton Square
- London EC1V 0HB. United Kingdom