Calculation of Dynamic Viscosity in Concentrated Cementitious Suspensions: Probabilistic Approximation and Bayesian...

Area of Science:

• Rheology
• Materials Science
• Statistical Mechanics

Background:

• The Krieger-Dougherty equation models dynamic viscosity in suspensions.
• Cement pastes, mortars, and concretes exhibit complex rheological behavior.
• Traditional deterministic models have limitations in capturing the inherent randomness of these systems.

Purpose of the Study:

• To introduce a probabilistic perspective to the Krieger-Dougherty equation.
• To enhance the calculation of dynamic viscosity in cementitious suspensions using Bayesian analysis.
• To transform parametric-deterministic models into parametric-probabilistic models for improved results.

Main Methods:

• Application of Bayesian analysis to the Krieger-Dougherty equation.
• Utilizing numerical methods like Markov Chain Monte Carlo (MCMC) and Gibbs Sampling.
• Employing specific software (OpenBUGS) to overcome computational complexity.

Main Results:

• Computation of probability density functions for Krieger-Dougherty equation parameters.
• Successful application to cement pastes, self-compacting mortars, and self-compacting concretes.
• Demonstrated significant improvement in dynamic viscosity calculations compared to theoretical values.

Conclusions:

• Bayesian analysis offers a robust framework for rheological modeling of complex suspensions.
• The probabilistic approach enhances the accuracy and reliability of dynamic viscosity predictions.
• This methodology provides a more enriched understanding of particle-fluid interactions in cementitious materials.