Predicting and optimizing hydraulic transport in drilling processes requires the formulation of transport models, which are able to consider the multi-phase flow in the entire wellbore. Here, from an engineering perspective, the drift-flux model is an appropriate approach due to its simplicity and efficiency. It considers only one mixture momentum and two continuity equations. The first describes the mass conservation of the mixture, the second the conservation of the dispersed phase. In the latter, the dispersed phase velocity is coupled with the mixture velocity through a slip velocity, defined as the averaged velocity difference between the mixture and the dispersed phase. In one-dimensional implementations of this model, the slip velocity needs to be correlated to flow variables (in drilling hydraulics: phase properties, eccentricity, drill string angular velocity, inclination and flow rate) through two parameters known as the distribution coefficient and the averaged drift flux velocity. Such parameters are system dependent and need to be generated through experiments [1]. In the case of high pressure-high temperature (HPHT) drilling systems, providing these data experimentally is basically impossible for technical reasons. Here, computational studies or virtual experiments may greatly enhance the closure of these models and quantification of parameters. However, careful validation of the computational models is essential to guarantee the reliability of the data. In this contribution, we present an Eulerian-Lagrangian multi-phase approach which precisely considers the momentum coupling of single particles with the carrying fluid. The method is validated [2] through detailed experimental data obtained in a vertical multiphase flow loop and processed with the particle image velocimetry (PIV) technique [3]. After validation, a sensitivity analysis to understand the effects of flow variables on the transport of cuttings is performed [4]. This is also the basis for the development of parameter models.
展开▼
