Operational conditions and modelling of backwashing in pressurised porous media filters

ENG- Pressurised sand filters are widely used in drip irrigation systems to ensure water quality by retaining suspended particles, thereby preventing emitter clogging. One of the main operational challenges of these systems is reducing their high energy consumption, which is directly linked to the pressure drop generated during the filtration and backwashing cycles. This energy demand depends on the internal filter design, the properties of the filtering medium, the flow regime, and operating conditions. Backwashing is a critical process for restoring the filter’s functional state and is often operated under conditions that prioritise cleaning efficiency over energy and water efficiency. Within this context, the main objective of this doctoral thesis was to evaluate the operating conditions and model the backwashing process in pressurised porous media filters, proposing design and operational strategies to improve their performance. Three scientific questions were formulated: (1) the validity of classical pressure drop equations under real operating conditions; (2) the predictive capacity of analytical and simulation models to estimate bed expansion; and (3) the impact of the filter medium type and nozzle design on global hydraulic behaviour. Firstly, CFD simulations were conducted in filtration mode, experimentally validated using a laboratory-scale filter. Three types of porous media (glass microspheres and two silica sands) and two bed heights (117 mm and 317 mm) were tested. Five classical equations were compared to model pressure drop (Ergun, Darcy-Forchheimer, Darcy, Kozeny-Carman, and a power function). For superficial velocities below 38.3 m h^(-1), all equations performed adequately. At higher velocities, the Ergun equation yielded the best precision (error <10%). Additionally, an analytical model was developed to account for non-uniform flow by dividing the bed into three zones with different gradients, successfully reproducing the simulated pressure losses. Next, CFD simulations in backwashing mode were carried out using an Eulerian–Eulerian multiphase formulation. These simulations accurately predicted the fluidised bed expansion for various flow rates when using microspheres, although expansion was overestimated for sand at low velocities. Due to the high computational demand, two analytical models based on control volume balances were proposed. Experimental validations showed that one model represents a lower expansion limit, while the other accurately predicted real expansion, especially with microspheres