CFD modelling of an high temperature PRSOV for dynamic evaluations

In the aeronautical field, CFD simulations, which were previously limited to the external aerodynamic aspects of the aircraft, are taking on an increasingly fundamental role during the development phases of complex systems, in order to have a data driven design optimization tool. The purpose of the thesis is the development of a solid approach to study the behavior of a high temperature PRSOV (Pressure Regulating and Shut Off Valve) part of the anti-ice system of an operating civil aircraft. An increase of efficiency in the anti-ice system reduces the engine air bleeding, ultimately lowering fuel consumption, toward a cleaner aviation. By implementing evolving conditions and dynamic geometries into a CFD simulation of the PRSOV it is possible to evaluate the behavior of significant variables which cannot be measured directly. Thesis results are validated by experimental test, indirectly through comparison of measurable quantities. Moreover, a lumped parameters model is used, despite the lower accuracies and details. The work can be divided in three main steps, designed to face the problem gradually, each one with increasing complexity and enhanced fidelity to the valve functioning. As first step, a pressure drop analysis is carried out to tune the characteristic parameters of the model using experimental data and academic literature. The second analysis considers the valve in three significative regulating positions, in order to understand how to model the flow under more demanding boundary conditions. These first two studies are preparatory for the last analysis, which consists in a dynamic simulation of the valve, where the regulating piston moves accordingly to the balance of fluid forces, adapting the geometry in response to variable inflow conditions. The proposed approach shows promising results, providing a useful tool for the prediction of the valve dynamics and laying the foundations for future works on stability, still an open aspect for the studied component.